Digital PDFs

AA-D538A-TC

March 1979

432 pages

Original

19MB

Document:	RMS-11 User’s Guide Order No. AA-D538A-TC
Order Number:	AA-D538A-TC
Revision:	0
Pages:	432
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/rsx11m_s/RSX11M_V3.2_Jun79/5B_FileMgmt_IO/AA-D538A-TC_RMS-11_V1.8_Users_Guide_197903.pdf

OCR Text

March 1979

This manual describes the concepts, structure, and_ usage of Record Management Services (RMS) software for the PDP-11. Though it describes syntax for
all programming languages supported by RMS-11, this Guide is not a definitive
source for that information.

RMS-11 User's Gulde
Order No. AA-D538A-TC

SOFTWARE:

RMS-11

V1 .8

additiona~

lro order
copies of this document, contact the Software Distribution
L.:.enter, Digital Equipment Corporation, Maynard, Massachusetts 01754

digital equipment corporation · maynard, massachusetts

The information in this document is subject to change without notice and
should not be construed as a commitment by Digital Equipment Corporation.
Digital Equipment Corporation assumes no responsibility for any errors that
may appear in this document.
The software described in this document is furnished under a license, and
may only be used or copied in accordance with the terms of such license.
No responsibility is assumed for the use or reliability of software on equipment that is not supplied by DIGITAL or its affiliated companies.

The postage-paid READER'S COMMENTS form on the last page of this
document requests your critical evaluation to assist us in preparing future
documentation.

The following are trademarks of Digital Equipment Corporation:
DEC
DECnet
DECsystem-10
DECSYSTEM-20
DECtape
DECUS
DIBOL
DIGITAL
FOCAL

IAS
MAS SB US
PDP
RSTS
RSX
UNIBUS
VAX
VMS
~I

3/80-14

Cc>ntent!;
Page
xv

PrE~face

Documentation Conventions

xvu

Chapter 1 Introduction.. to RMS-11
1.1
l.2

''-"'

Concepts of Data Organization . . . . . . . .
RMS-11 Implementation of Data Organization

1-1
1-10

1.2.1
1.2.2
1.2.3

Hardware Data Structure.
Software Data Structure .
The RMS-11 Interface . .

1-11
1-12
1-16

1.2.3.1
1.2.3.2
1.2.3.3

Record Formats.
File Organizations.
Record Access Modes

1-16
1-17
1-19

1.2.3.3.1
1.2.3.3.2
1.2.3.3.3
1.2.3.3.4

1-19
1-21
1-23
1-23

Record Operations . .
RM S-11 Utilities . . .

1-25
1-26

Record Processing Environment.

1-27

1.2.4.1
1.2.4.2
1.2.4.3
1.2.4.4
1.2.4.5

1-28
1-30
1-31
1-33
1-33

1.2.3.4
1.2.3.5
1.2.4

1.2.5

1.2.6

Sequential Access Mode·
Random Access Mode .
Access by Record's File Address
Changing Record Access Modes.

Using RMS-11 . . . .
I/0 Buffers in Record Operations.
Record Access Streams . . . . .
IAS/RSX-llM Asynchronous Record Operations
Record Transfer Modes

File Processing Environment

1-33

1.2.5.1
1.2.5.2
1.2.5.3
1.2 .5 .4

1-33
1-36
1-36
1-37

File Processor. .
File Sharing . .
File Operations .
File Attributes .

Bypassing Record Processing .

1-40

Chapter 2 Application Design

~cI
I

2 .1
2.2

When to Design . . . . . . .
Design Considerations . . . .

2-2
2-3

2.2.1
2.2.2
2.2.3

Maximize Speed First
Reducing Space Requirements
Provide Shared Access . . . .

2-3
2-4
2-5

2.2 .3 .1
2.2.3.2
2.2.3.3
2.2 .3.4

2-5
2-7
2-12
2-12

2.2.4
2.3
2.4

System Protection Codes
Sharing among Programs
Sharing among Record Access Streams.
Programming Considerations.

Remember Ease of Design

Design Process. . . . . . . .
Selecting a File Organization .

2-13
2-14
2-14
~~· ~

Chapter 3 Sequential File Applications
3 .1
3.2

3.3

Record Definition . . . .
File Design . . . . . . .

3-2
3-3

3.2.1
3.2.2
3.2 .3

3-4
3-5
3-6

Task Design . . . . . . . .

3-6

3.3.1

Record Operations .

. 3-7

3.3.1.1
3.3.1.2
3.3.1.3
3.3.1.4
3.3.1.5
3.3.1.6
3.3.1.7
3.3.1.8
3.3.1.9

3-7
3-8
3-8
3-9
3-9
3-11
3-11
3-11
3-12

3.3.2

3.3.3

3.3.4

Initial Allocation.
Default Extension Quantity
Contiguity. . . . .

Connect .
Disconnect
Find .
Flush.
Get . .
Put . .
Rewind.
Truncate .
Update . .

Record Transfer Modes.

3-13

3.3.2.1
3.3.2.2

Move Mode.
Locate Mode

. 3-13
. 3-14

I/0 Techniques . . .

. 3-14

3.3.3.1
3.3.3.2
3.3.3.3
3.3.3.4
3.3.3.5

. 3-14
. 3-14
. 3-14
. 3-15
. 3-15

IAS/RSX-llM Asynchronous Record Operations
Deferred Write . . . . . . . . .
Multiple Buffers . . . . . . . .
Multiple Record Access Streams .
Multi-Block Count (MBC).

File Operations

. 3-15

3.3.4.1
3.3.4.2
3.3.4.3
3.3.4.4
3.3.4.5

. 3-16
. 3-16
. 3-16
. 3-16
. 3-16

Close.
Create
Open .
Erase.
Extend .

·~

Chapter 4 Relative File Applications
4 .1
4.2

Record Definition . .
File Design . . . . .

. 4-2
. 4-2

4.2.1
4.2.2

Bucket Size .
Allocation . . .

. 4-2
. 4-3

4.2.2.1
4.2.2.2
4.2.2.3

. 4-4
. 4-5
. 4-5

4.2.3
4.3

Initial Allocation
Default Extension Quantity
Contiguity . . . .

Maximum Record Number .

. 4-6

Task Design . . . . . . . .

. 4-7

4.3.1

Record Operations .

. 4-7

4.3.1.1
4.3.1.2
4.3.1.3
4.3.1.4
4.3.1.5
4.3.1.6
4.3.1.7
4.3.1.8
4.3.1.9

. 4-8
. 4-8
. 4-8
. 4-8
. 4-10
. 4-10
. 4-12
. 4-13
. 4-13

4.3.2

4.3.3

4.3.4

Connect .
Delete . .
Disconnect
Find .
Flush.
Get . .
Put . .
Rewind.
Update.

Record Transfer Modes.

. 4-14

4.3.2.1
4.3.2.2

Move Mode.
Locate Mode .

. 4-14
. 4-14

1/0 Techniques . . . .

. 4-15

4.3.3.1
4.3.3.2
4.3.3.3
4.3.3.4

. 4-15
. 4-15
. 4-16
. 4-17

IAS/RSX-llM Asynchronous Record Operations
Deferred Write . . . . . . . . .
Multiple Buffers . . . . . . . .
Multiple Record Access Streams .

File Operations

. 4-17

4.3.4.1
4.3.4.2
4.3.4.3
4.3.4.4
4.3.4.5

. 4-17
. 4-17
. 4-17
. 4-18
. 4-18

Close.
Create
Open .
Erase.
Extend .

Chapter 5 Indexed File Organization
5.1
5.2

Physical Structure . .
Conceptual Structure

. 5-2
. 5-4

5.2.1

. 5-5

Data .
5.2.1.1
5.2 .1.2

5.2.2
5.2,3
5.2.4

Level 0 of the Primary Index.
Level 0 of the Alternate Indexes .

Indexes . . . . . . . . . . . . . . . . .
Random Access Using the RMS-11 Indexed File Structure .
Why this Structure? . . . . . . . . . . . . . . . . . . .

. 5-5
. 5-6
. 5-6
. 5-8
. 5-8

5.3

Procedures for Performing Random Record Operations .

5.3.1

5.4
5.5

Putting a Record.

5.3 .1.1 Simplest Case.
5.3.1.2 Bucket Splitting
5.3.1.3 Incremental Reorganization
Getting and/or Finding a Record
5.3.2
Updating a Record .
5.3.3
Deleting a Record
5.3.4
Procedures for Performing Sequential Record Operations.
1/0 Cost of Performing Record Operations.

5-10
5-10
5-10
5-12
5-13

£~'

5-13
5-14
5-16
5-17
5-17

Chapter 6 Indexed File Design
6.1
6.2

6.2.1
6.2.2
6.2.3
6.2.4
6.2.5
6.3
6.4
6.5

Bucket Size for Primary Index
Bucket Sizes for Alternate Indexes
Program Syntax .

Allocation .

6.6.1
6.6.2
6.7

Num her of Keys .
Key Size
Key Data Type
Position of Key in Record
Key Characteristics

Areas
Placement Control .
Bucket Size .

6.5.1
6.5.2
6.5.3
6.6

6-1
6-2

Record Definition
Key Selection .

Initial Allocation.
Default Extension Quantity

Population Techniques .

6.7.1
6.7.2

Ascending Order by Primary Key .
Random Insertions after File Population.

6.7.2.1
6.7.2.2

Fill Number
Mass Insert.

6-2
6-3
6-4
6-6
6-7

6-10
. 6-14
. 6-16
. 6-17
6-21
6-22
6-24

6-24
6-29
6-29
6-30
6-31
6-31
6-33

,)--~

Chapter 7 Indexed Task Design
7.1

Record Operations .

7-1

7.1.1
7.1.2
7.1.3
7.1.4
7.1.5
7.1.6
7 .1.7
7.1.8
7.1.9

7-1
7-2
7-2
7-2
7-4
7-4
7-5
7-6
7-6

Connect.
Delete.
Disconnect.
Find
Flush .
Get.
Put.
Rewind
Update

.....

7.2

7.3

7.4

Record Transfer Modes.

. 7-6

7.2.1
7.2.2

Move Mode .
Locate Mode

. 7-7
. 7-7

I/0 Techniques . . .

. 7-8

7.3.1
7.3.2
7.3.3
7.3.4
7.3.5

. 7-8
. 7-8
. 7-8
. 7-8
. 7-8

IAS/RSX-llM Asynchronous Record Operations.
Deferred Write. . . . . . . . .
.
Multiple Buffers . . . . . . . . . . . .
Multiple Record Access Streams . . . .
Sequentially Reading Write-Shared Files

File Operations

. 7-10

7.4.1
7.4.2
7.4.3
7.4.4
7 .4.5

. 7-10
. 7-11
. 7-11
7-11
. 7-11

Close .
Create.
Open .
Erase .
Extend

Chapter 8 Common Optimization Techniques
8.1

Task Building with RMS-11 Routines.
8.1.1

Disk-Resident Overlays. . . .

8-4

8.1.1.1
8.1.1.2

8-7
8-8

Standard ODL Files.
Prototype ODL Optimization
8.1.1.2.1
8.1.1.2.2
8.1.1.2.3
8.1.1.2.4

8.1.2

Techniques . . . .
Possible Task Builder Errors .
Calculating Changes in Task Size
Examples of ODL Optimization

. 8-9
8-14
8-14
8-14

Memory-Resident Overlays . . . . . . . . . . . . .

8-20

8.1.2.1
8.1.2.2
8.1.2.3
8.1.3
8.2

. 8-21
. 8-22
. 8-23
8-23

Program Development . . . . . . . . . . . . . . . . . . .

8-24

Flow of Operations Should Reflect Overlay Structure
Task Builder Considerations

8-24
. 8-25

Virtual-to-Logical Block Mapping.

8-26

8.:3.1

8-26

Retrieval Pointers on Disk
8.3.1.1
8.3.1.2

8.3.2
8.3.3

8.4

Building the RMS-11 Resident Libraries . .
Task Building against an RMS-11 Resident Library.
Installing an RMS-11 Resident Library.

Deciding between Types of Overlays . . . . . . . .

8.2.1
8.2.2
8.3

. 8-1

IAS/RSX- llM .
RSTS/E . . . .

8-26
. 8-26

Retrieval Pointers in Memory.
Optimizing Window Turning .

. 8-27
. 8-27

8.3.3.1
8.3.3.2

Other Optimizations . . . . . .

. 8-27
. 8-28
. 8-29

8.4.1
8.4.2
8.4.3

. 8-30
. 8-30
. 8-30

IAS/RSX-llM
RSTS/E

Data Caching . . . . .
Allocating More Resources to the Task
Disk Usage . . . . . . . . . . . . .

vii

Chapter 9 RMS-11 Utilities
9.0.1
9.0.2

9.0.2.1
9.0.2.2
9.0.2.3
9.0.2.4
9.0.2.5
9.0.2.6
9.0.3
9.1

9.1.4

. 9-10
. 9-10
. 9-12
. 9-12
. 9-12

Permanently Installed Utility
Uninstalled Utility

. 9-12
. 9-14
. 9-15
. 9-17
. 9-17

General Form .
Global Switches.
Outfile Switches
Infile Switches
Command String Examples

9-18

RMSCNV Command Utility

9-19

9.2.1
9.2.2
9.2.3

9.2.4

. 9-21
. 9-21

Permanently Installed Utility
Uninstalled Utility

. 9-21

Command String.
9.2.4.1
9.2.4.2
9.2.4.3
9.2.4.4
9.2.4.5

. 9-27
. 9-28

RMSDEF Interactive Utility .

. 9-29

Purpose .
Effect.
Call and Termination

. 9-29
. 9-29
. 9-32

9.3.1
9.3.2
9.3.3

9.3.3.1
9.3.3.2
9.3.3.3

Permanently Installed Utility
Uninstalled Utility
Terminating The Utility.

. 9-21
. 9-22
. 9-23
. 9-26
. 9-26

General Form .
Global Switches.
Outfile Switches
Infile Switches
Command String Examples

Cautions
VO Techniques

9.2.5
9.2.6

iii-

9-19
. 9-19
. 9-21

Purpose .
Effect.
Call and Termination
9.2.3.1
9.2.3.2

. 9-12

Command String.

Cautions

9.1.5

. 9-10

Purpose .
Effect .
Call and Termination

9.1.4.1
9.1.4.2
9.1.4.3
9.1.4.4
9.1.4.5

9.3

. 9-9

Documentation Conventions

9.1.3.1
9.1.3.2

. 9-5
. 9-6
. 9-6
. 9-7
. 9-7
. 9-7

Command vs. Interactive
Installed vs. Uninstalled.
Indirect Files .
Command String Continuation.
Patch Level .
Command Utility Error Messages

RMSBCK Command Utility
9.1.1
9.1.2
9.1.3

9.2

. 9-2
. 9-4

Using the RMS-11 Utilities.
Utility Conventions

. 9-32
. 9-33
. 9-33

9.3.4

......,..

Process
9.3.4.1
9.3.4.2
9.3.4.3
9.3.4.4
9.3.4.5
9.3.4.6
9.3.4.7
9.3.4.8

9.4

9.4.1
9.4.2

. 9-49
. 9-50

9.4.3

Purpose .
Effect .

9.4.4

Permanently fustalled Utility
Uninstalled Utility

Command String.
9.4.4.1
9.4.4.2
9.4.4.3

·~

For Disk Sequential Files
For Magnetic Tape Files.
For Relative Files .
For fudexed Files .
For fudexed Files With FU Switch Specified

Call and Termination
9.4.3.1
9.4.3.2

General Form .
Switches
Command String Examples

. 9-50
. 9-51
. 9-51
. 9-51
. 9-51
. 9-52
. 9-52
. 9-52
. 9-53
. 9-53
. 9-54
. 9-54

Cautions

. 9-57

RMSIFL Command Utility.

. 9-57

9.4.5
9.5.1
9.5.2
9.5.3

9.5.4

Purpose .
Effect .
Call and Termination

. 9-57
. 9-57
. 9-61

9.5.3.1
9.5.3.2

. 9-61
. 9-61

Permanently Installed Utility
Uninstalled Utility

Command String.
9.5.4.1
9.5 .4.2
9.5.4.3
9.5.4.4

General Form .
Global Switches.
Outfile Switches
Infile Switches

. 9-61
. 9-61
. 9-62
. 9-63
. 9-66

Cautions

. 9-66

RMSRST Command Utility

. 9-67

9.5.5
9.6

. 9-33
. 9-34
. 9-35
. 9-38
. 9-41
. 9-45
. 9-47
. 9-48
. 9-49

'-"'

'-"

Command File
File Specification .
Data Structure
Key Definition
File Structure.
Data Allocation .
Protection
File Creation .

RMSDSP Command Utility

9.4.2.1
9.4.2.2
9.4.2.3
9.4.2.4
9.4.2.5

9.5

. 9-33

9.6.1
9.6.2
9.6.3

Purpose .
Effect .
Call and Termination
9.6.3.1
9.6.3.2

Permanently Installed Utility
Uninstalled Utility

. 9-67
. 9-68
. 9-69
. 9-69
. 9-69

'-"
ix

9.6.4

Command String.
9.6.4.1
9.6.4.2
9.6.4.3
9.6.4.4
9.6.4.5

9.6.5

General Form .
Global Switches.
Outfile Switches
Infile Switches
Command String Examples

9-69
. 9-69
. 9-71
. 9-72
. 9-73
. 9-74
. 9-75

Cautions

Appendix A RMS-11 and the Operating Systems
A.1

A.1.1
A.1.2
A.1.3
A.1.4

A.2

A.3

A.4

. A-1

IAS.
RMS-11 Restrictions on IAS
IAS Restrictions on RMS-11
Compatibility with Other File Managers
Asynchronous Operations. . . . . . . · .

. A-1
. A-1
. A-1
. A-2

A.1.4.1
A.1.4.2

. A-2
. A-2

RMS-11 Synchronous Environment
RMS-11 Asynchronous Environment.

RSTS/E . . . . . . . . . . . . . . . . .

. A-3

A.2.1
A.2.2
A.2.3

. A-3
. A-3
. A-4

RMS-11 Restrictions on RSTS/E .
RSTS/E Restrictions on RMS-11 .
Compatibility with Other File Managers

RSX-llM . . . . . . . . . . . . . . . . .

. A-4

A.3.1
A.3.2
A.3.3
A.3.4

RMS-11 Restrictions on RSX-llM .
RSX-llM Restrictions on RMS-11 .
Compatibility with Other File Managers
Asynchronous Operations. . . . . . . .

. A-4
. A-4
. A-4
. A-4

A.3.4.1
A.3.4.2

. A-4
. A-5

RMS-11 Synchronous Environment
RMS-11 Asynchronous Environment.

VAX/AME . . . . . . . . . . . . . . . .

. A-6

A.4.1
A.4.2
A.4.3

RMS-11 Restrictions on VAX/AME.
VAX/AME Restrictions on RMS-11.
Asynchronous Operations. . . . . .

. A-6
. A-6
. A-6

A.4.3.1
A.4.3 .2

. A-6
. A-7

RMS-11 Synchronous Environment
RMS-11 Asynchronous Environment .

Appendix B RMS-11 and the Programming Languages
B.l
B.2
B.3

Implementation in Languages
Function Chart . . .
Error Code Mapping . . . . .

. B-1
. B-1
. B-4

Appendix C RMS-11 Disk-Resident Overlays
C .1
C.2

Overlay Structures. . .
RMS-11 ODL Files . .

. C-1
. C-6

C.2.1
C.2.2

. C-6
. C-7

RMS16X.ODL.
RMS20X.ODL.

At;

Ap1>endix D RMSDFN Command Utility
D .1
D .2
D .3

Purpose . . . . . . .
Ef:fect . . . . . . . .
Call and Termination

. D-1
. D-1
. D-2

D.3.1
D.3.2

. D-2
. D-2

Permanently Installed Utility.
Uninstalled Utility"

D.4

Command String . .
D.4.1 General Form
D.4.2 Switches.
D.4.3 Examples

D.5

Cautions . . . .

. D-2
. D-2
. D-3
. . D-6
. D-7

Appendix E Utility Error Messages
E.l
E.2

RMSDEF Interactive Utility
Command Utilities"

. E-1
. E-9

Ap1>endix F Magnetic Tape Handling
F .1
F.2

General Magnetic Tape File Processing .
RMS-11 Magnetic Tape File Processing.

. F-1
. F-3

Rewinding Tape Volumes . .
Positioning for the Next File . .

· F-3
. F-4

F.2.1
F.2.2

Glossary

Index

Figures
PRE-1 Map for the RMS-11 User's Guide
1-1 Files . . . . . . . .
1-2 Files with Attributes .
1-3 Records in a File. . .
1-4 Record Formats . . .
1-5 Record Access Modes
1-6 Sequential File Organization .
1-7 Relative File Organization
1-8 Indexed File Organization .
1-9 Random Access Example . .
1-10 RMS-11 in Its Environment
1-11 Physical Storage Structure .
1-12 Logical Data Structure . . .
1-13 Virtual-to-Logical-Block Mapping.
1-14 Sequential File Organization .
1-15 Relative File Organization . . . .

xvi
1-2
1-2
1-3
1-4
1-6
1-7
1-8
1-8
1-9
1-11
1-13
1-14
1-15
1-17
1-18

1-16 Primary Index in RMS-11 Indexed File .
1-17 Program Sequentially Reading a Sequential File.
1-18 Program Sequentially Reading a Relative File .
1-19 Program Sequentially Writing to a Relative File.
1-20 Program Sequentially Reading an Indexed File
1-21 Program Randomly Reading a Relative File .
1-22 Program Randomly Reading an Indexed File
1-23 Sequential Account File
1-24 System Memory Layout
1-25 RMS-11 Task Structure
1-26 Record Operations and Stream Context .
1-27 RMS-11 's Environment
1-28 Records Spanning Blocks.
2-1 Time Factors in an VO Operation
2-2 System Protection Concepts
2-3 Bucket Locking Example.
2-4 Record-Length Field on Disk and Tape
3-1 RMS-11 Task Structure
4-1 RMS-11 Task Structure
5-1 Indexed File with and without Areas
5-2 Formatted Bucket .
5-3 Index as a Pyramid
5-4 Format for Secondary Index Data Record .
5-5 Example of a Primary Index
5-6 Search Time Curves .
6-1 Single-Area Indexed File .
6-2 Example of Single-Area Indexed File
6-3 Two-Area Indexed File .
6-4 Example of Multi-Area Indexed File
7-1 RMS-11 Task Structure
8-1 Source-to-Task Sequence.
8-2 RMS-11 in Tasks
•'
8-3 Sample Overlay Structure and ODL File
8-4 Incremental Optimization Example .
8-5 Concatenation around Overlays Example
8-6 Duplicate Module Example No. 1.
8-7 Duplicate Modulee Example No. 2
8-8 Overlay Structure Diagram .
9-1 RMS DEF Processing Flowchart.
C-1 RMSllS.ODL and RMSllX.ODL Overlay Structures

xii

1-19
1-20
1-20
1-20
1-21
1-22
1-24
1-25
1-27
1-29
1-32
. 1-34
. 1-35
2-4
2-6
2-10
2-15
3-13
4-14
5-3
5-4
. 5-5
. 5-6
. 5-7
. 5-9
. 6-11
. 6-12
. 6-12
. 6-13
. 7-7
. 8-2
. 8-3
. 8-5
. 8-9
. 8-10
. 8-11
. 8-12
. 8-13
. 9-31
. C-3

Tables
1-1
2-1
2-2
2-3
3-1
3-2
4-1
5-1
6-1
9-1
9-2
9-3
9-4
9-5
B--1
B--2
D--1
D-2

Records Formats and File Organizations . . . . .
Shared Access Criteria . . . . . . . . . . . . . .
File Organization Characteristics and Capabilities .
File Organization Advantages and Disadvantages
End-of-File Indicators . . . . . . .
Sequential File Data Sizes (in bytes) . . .
Relative File Data Sizes (in bytes) . . . .
1/0 Cost of Performing Record Operations.
Key Data Types . . . . . .
RMSBCK Utility Switches.
RMSCNV Utility Switches .
RMSDSP Utility Switches .
RMSIFL Utility Switches .
RMSRST Utility Switches .
RMS-11 Features Supported by Programming Languages
Language Error Code Mapping . .
RMSDFN Utility Switches . . . .
Key Characteristic Combinations.

. 1-37
. 2-8
. 2-18
. 2-19
3-2
3-3
4-2
5-18
. 6-3
. 9-14
. 9-22
9-53
9-63
9-70
B-2
B-5
. D-3
. D-4

Commercial Engineering Publications Typeset this manual usingDIGITAL's
TMS-11 Text Management System.
827ALL

xiii

""'!

PrE!face
Hecord Management Services for the PDP-11 (RMS-11) provides powerful
data management capabilities. The RMS-11 User's Guide tells you about
those capabilities and how to use them. This manual is designed for instruction and for reference:
• If you read the manual according to the map in PRE-1 (see "Structure of
the Manual"), you can learn practically everything you need to know about
RMS-11, except its language implementation.
• If you use the manual as a reference, its structure and extensive index make
it easy to consult.

RMS-11 is a set of software routines that transfers data between a running
program, which uses data in a logical form called records, and the file processor portion of an operating system, which maintains the physical structure of
the data on a storage device.

NOTE
You can use RMS-11 Indexed files only if you have purchased
the RMS-1 lK software product. Your system manager can tell
you if your system includes this capability.

Prerequisites
To read this manual, you should be either a MACR0-11 or higher level
language programmer who wants to improve the performance of an application you have written or are planning to write. In other words, you should have
a working knowledge of a DIGITAL programming language that supports
RMS-11 and some motivation to inquire into the workings of RMS-11.

NOTE
This manual describes the functions of RMS-11. Only
MACR0-11 programs can use the full set of capabilities. The
higher level languages support a subset of those capabilities.
You must use your language documentation to determine:
• what RMS-11 facilities you can actually use in your higher
level language
• the syntax to use them

Figure PRE-1: Map for the RMS-11 User's Guide

L§]
B
NEED
MORE
INFO
LANGUAGE
MANUALS

A,.._-~

ENOUGH
INFO
INTRODUCTION
TO
RMS-11
CH.1

WRITE
APPLICATION

ENOUGH
INFO

APPLICATION
DESIGN
CH.2

RUN
APPLICATION
INDEXED
FILE
ORGANIZATION
CH.5
INDEXED
FILE
DESIGN
CH.6
INDEXED
TASK
DESIGN
CH.7

UNSATISFIED
A

EVALUATE
APPLICATION

SATISFIED

RMS-11
UTILITIES
CH.9

DESIGN
APPLICATION

B
.f-MK-00098-00

xvi

Striucture O'f the Manual
Figure PRE-1 shows the paths through this Guide you should follow in the
process of writing your applications. The following paragraphs describe the
parts of the manual shown in that map:

Chapter 1 addresses the RMS-11 beginner: it discusses basic RMS-11
concepts that are common to all programming languages in which RMS-11
applications can be written.
Chapter 2 introduces RMS-11 design, why it is necessary, and its general
premises.
Chapters 3-7 describe the file organizations, their structures, and file and task
design considerations.
Chapter 8 covers common techniques that can be used to optimize RMS-11
applications, regardless of the file organization selected.
Chapter 9 describes the RMS-11 utilities. These tasks provide RMS-11
facilities to all users, programmer or operator, higher level language or
MACR0-11.
Appendixes provide supplementary information, such as:
Implementation of RMS-11 on the various operating systems
RMS-11 features supported by DIGITAL programming languages
How the higher level languages map RMS-11 error codes
Documentation of the RMSDFN utility
Utility error messages

'-"''

The Glossary reprises all terms defined in this manual.

Ass;ociatE~d

Documents

The RMS-11 documentation set contains the following manuals:

RMS-11 User's Guide
RMS-11 MACR0-11 Reference Manual
RMS-11 Installation Guide
You must also use operating system and language documentation. See the
Documentation Directory for your operating system.

Implementation in Languages
RMS-11 is the record management software for the following PDP-11 programming languages. See Appendix B for more information.
BASIC-PLUS-2
DIBOL
MACR0-11
PDP-11 COBOL
RPG II

xvii

Implementation in Operating Systems
RMS-11 is available as an interface between user application programs and
data storage devices on the following DIGITAL operating systems. See
Appendix A for more information.
IAS
RSTS/E
RSX-llM
VAX/AME

NOTE
The TRAX system uses RMS-11 as part of its central data
manager. See TRAX documentation for a discussion of the
RMS-11 interface on that operating system.

Documentation Conventions
RMS-11 operates similarly on the supporting operating systems (see Appendix A); RMS-11 is the common record access software for most higher level
languages. Therefore, it should be possible to produce a single manual describing that operation. However, the differences among operating systems
and languages present a barrier to that unification. The following conventions
are designed to enable you to hurdle that fence.

Differences in Operating System Functions
Details that are common to the operating systems are printed in black, and
the details that are specific to one operating system or another are printed in
color as follows:
• RSTS/E-specific information is printed in red.
• IAS-, RSX-llM-, and VAX-specific information is printed in blue.

Differences in Language Support for RMS-11
Differences in language functionality are noted only in Appendix B. You
should also consult the language documentation. However, note that
MACR0-11 programs can use all facilities described. Therefore, MACR0-11
error codes are used in the discussions. Appendix B contains a chart showing
how the higher level languages map these codes.

xviii

J•11

TEtrmlnology
The operating systems may have different terms for the same process.
Example "Installed" to an RSX-UM user means essentially the same as "CCL" to
a RSTS/E user in that the user just has to enter a short command to
activate the task.

In this manual, these terms are reconciled under a cover term with a specific
definition of what that term means to the different operating systems. The
most pervasive cover terms are defined here. Other terms are defined in
CONVENTION notes as they are needed.
Cover Term

Definition

filespec

file specification; see your operating system documentation for syntax and
Appendix A for RMS-11 restrictions on that syntax

account

the User File Directory (UFD) or its number; also known as User Identification Code (UIC) or the Project, Programmer Number (PPN)

[act nbrJ

account number; used in filespecs

file-name

the name of the file

extension

the file-name extension to the right of the dot also called file-type

protection code

the letter codes R, W, E, and/or D and the number codes 0 through 255 that
represent the level of protection for a file by the operating system

wild card

the presence of an asterisk ( *) in a field of the filespec indicates that any
value is acceptable for that field

Usier Interface

• Even though the IAS/RSX-llM monitor prompt is not shown, it is assumed
before any utility or other system task is called.
• Even though the RSTS/E system response is not shown, it is assumed after
any utility or other system task is terminated.

Terminal Displays
Any examples of terminal displays show the software messages with underlines; your inputs are not underlined.
Example From "Task Building Against the RMS-11 Resident Library," Section
8.1.2.3:

TKB >USER. TSK, USER.MAP= US ERA, US ERB, USERC
TKB>/
ENTER OPTIONS:
TKB>LIBR=RMSRES:RO
TKB>//

xix

Chapter 1
Introduction to RMS-11

Your business, whether commercial, scientific, governmental, or educational,
relies on data. That data indicates the state of your business; that data helps
you control the future of the business. Therefore, you want fast, effective
access to it.
Computer hardware, with its speed (times in nanoseconds) and its mass data
storage (millions of characters), provides the means for that kind of access.
The problem is converting the data from the way you talk about it to the way
the computer system can handle it-and back again.

RMS-11 is the translator between you and your system. A set of software
routines used by your programs, RMS-11 processes data for you in units
called records. This introduction takes you slowly into RMS-11, beginning
with common concepts of data organization. Then the discussion turns to how

RMS-11 implements those concepts and fits into its environment, that is,
your programs, language, and operating system.

1.11 Concepts of Data Organization
First, let's examine the concepts involved in organizing data, using images
from a noncomputer environment you're familiar with.
Fiie

When data is stored on paper, people gather it in containers called files (see
Figure 1-1). A file not only keeps related data in one place, but it also segregates that data from other information. The term file applies to the data as
well.
Example Personnel information, including names, address, job titles, pay rates, and so on.
Example Product information, including part numbers, prices, specifications, discount rates,
and more.

1-1

Figure 1-1: Files

H-MK-00091-00

Access

Figure 1-2: Files with Attributes

H-MK-00090-00

An advantage of the segregation provided by files is controlled access. Some
files, like payroll and budget files, are available only to a restricted group of
people. Other files, like transaction or inventory files, are used by a larger
group of different people. And there are some files that everyone uses, such as
a telephone book.
You can extend the access concept further and say that within the group of
people who can use a file, only certain ones are allowed to put data into it or
change the data already in it; others can only read what's in the file.
Example Telephone numbers, especially those of customers, should be changed only by
authorized people so that the numbers remain current and accurate.

1-2

Introduction to RMS-11

Fiie
Attributes

A file cabinet has features that identify it, that tell what's in it, even how the
data is organized. These features, or attributes, can be as a simple as "the
gray three-drawer" or "the one by the window."
As file requirements become complicated and cabinets multiply, more complete and precise identification is necessary. Then, the files acquire names or
numbers, signs announcing who can use the file, cards cross-referencing the
data in the drawers, and so on (see Figure 1-2).

Record

Each file contains discrete groups of items whose form is repeated throughout
the data. Each group, or record, represents an entity, and a file consists of a
series of such entities.
Example In a personnel file, all the information on an employee constitutes a record; therefore, the number of records in the file equals the number of employees.
Example In an inventory file, all the information on a stock item comprises a record.

~·

Within each record are the specific items of data with which you are
concerned.

Figure 1-3: Records in a File

H-M K-00086-00

Introduction to RMS-11

1-3

Record
Format

On paper, a record can be a form (see Figures 1-3 and 1-4), and different
records require different forms.
Some forms are always the same length. Their information does not expand
with time or use.
Example A product information form: if the facts about a product change, you fill out a new
form; if you add products to the file, you fill out a new form.

Other forms vary over time and use. They have continuation sheets or some
other extension.
Example An employee payroll record: an employee with the company ten years has more data
on file than a new employee.

Other records use a combination of these two formats.
Figure 1-4: Record Formats
PRODUCT RECORD

...............

,..,,,,,,,.,,,~

SALARY HISTORY

E.C.O. HISTORY

__.

..,,.,,,._.,-..,.,,..-~~

:::~iQ~

:~~~

....................
---~

_,.,..,

END

(MAY CONT.)

Q-M K-00073-00

Access
Modes

Once you have records in a file, you get, or access, them in two ways (see
Figure 1-5) :

NOTE
Record access not only means retrieving a record from a file; it
also includes putting the record into the file.
Sequentially
You pick a point in the file and then access records one at a time. Often
you start at the beginning of the file because you want to look at each
record in the file-probably for a report you are compiling. At other times,
you start partway through the file.
Context

To read each record, you take it out of the file. You mark the position each
time with a card or upside-down folder so that you know:
• where to put the record back into the file
• where the next record is

1-4

Introduction to RMS-11

To :insert records sequentially, you reach into the drawer to the place
where you want the records to go and use two fingers to hold open a space.
Then you take a sheet from the stack of records and slip it into position
between your fingers. You move your finger behind the record you added,
holding open a space for another form. Repeating these steps, you insert
the entire stack into the file in the sequence they are available to you.
In either type of access, you move through the records consecutively. Each
record inserted or retrieved relates to the record accessed right before it.
Randomly
You determine the record you want on some basis other than its order
of occurrence in the file. Perhaps you have a list oflocations. Then you reach
into the file at the exact record's location. Each record selection is
independent of the previously accessed record and of the next record
processed.
To randomly access records in a file, you use an identifier. You can assign
a number to each record and use that to put the record into the file and get
it out again. But it is simpler when the identifier is part of the record
itself. In fact, it helps if you can locate a record via more than one identifier,
or key, within the data.
Or you can use the physical location of the record. For example, you have
a file stored in a three-drawer cabinet. Inside each drawer you make tiny
slots, each capable of holding one record form. Each slot has a number.
You can, therefore, identify a record's location, or address, within the file,
using a drawer number and a slot number within that drawer.
There are times, of course, when you access records in a file sequentially and
randomly. Usually, you randomly access the first record in a series and then
sequentially access the records in that series.
Example The records in a personnel file contain departmental codes. If the records of all
personnel in a department are grouped together, you can produce a report on one
department by randomly accessing the first record in the file with the department's
code and then reading the consecutive records with that code.

Record
Op1eratlons

What will you do with a record when you come to its place in the file?
• Verify that the record exists in the place it should.
• Read the record, that is, examine its contents._
• Insert a record in the position you've located.
• Revise the contents of the record, that is, change some data in it.
• Remove the record from the file.

Fiie Organizations

Generally, the person who uses a file establishes a method of organizing the
records within it. This method reflects the file's use and dictates the information and time needed to locate a record.

Introduction to RMS-11

1-5

Figure 1-5: Record Access Modes

H-MK-00093-00

Sequential

For instance, you want to file a series of records:
• You have little or no need to access the records randomly.
• You generally use all the records in the file in an unvarying order when you
open it.
With these requirements, you can organize the records by stacking them in a
file:
• The records assume the physical sequence in which they are put into the
file.
• There are no spaces, where records could be inserted later, left in the series.
Each record, except the first, has a record before it; each record, except the
last, has a record following it.

1-6

Introduction to RMS-11

""'

This is a sequentially organized file (see Figure 1-6). Its overhead and upkeep
are minimal. To put a record into the file, you just put it after the last record
already there.
Figure 1-6: Sequential File Organization

H-MK-00087-00

Rtelatlve

However, if you want more access flexibility, you can change the organization.
This time, set up a series of file folders, numbered in sequence from "1" to the
last folder in the file. Each folder is the same size; it holds only one record, but
it can be empty.
This Relative organization (see Figure 1-7) enables you to access specific
records more easily. You do not have to look sequentially through the records
before the one you want (though you can). You use the numbers on the folders
to locate or insert records. You can even code the forms to make use of some
internal data items as the basis of these relative record numbers.

Indexed

You have a large file and most of the time, you randomly access its records.
The list of relative record numbers is taking more time than it's worth, because you use several kinds of information to look for records. You are ready
for an Indexed file, though you may still want to sequentially access the whole
file to compile reports.

Introduction to RMS-11

1-7

Figure 1-7: Relative File Organization

H-MK-00089-00

When you open the file drawer, you see papers neatly stacked, with numbered
tabs sticking up (see Figure 1-8). At the front of the drawer there is a set of
small card files; inside them are groups of cards separated by dividers. The
cards in each file are an index to the records in the back of the drawer. You
look at the record you have to insert and find the data item marked "KEY."
Using the information there, you consult an index to determine where you
should insert the record.

Figure 1-8: Indexed File Organization
SEQUENTIAL
ACCESS

BY
KEY 0

Q-MK-00078-00

1-8

Introduction to RMS-11

.....i.

Or you can find a record by looking for a specific value in one of the key files.
Example You want the record of a transaction with Q,R,&S, so you open the Transaction File
drawer. Inside, the records are filed at the back and there are five indexes at the
front (see Figure 1-9). You know that "Q,R,&S" is the Primary Key of the record
you want and you open the index labeled "O".
The first record in the index is the Root. It has a list of Primary Key values on it;
each value is paired with a number. You look down the alphabetical list until you
find a value equal to or greater than "Q,R,&S". You find "Rhesus INC" and the
number next to it is "3".
You put the Root back in the index and go to the first divider and the third record
behind it. You see that "Rhesus INC" is the last entry on this card, but you scan up
the list to find a value closer to the one you're looking for. You find "Queeg Company" and the number "7".
So you reach into the back of the drawer, trailing down the tabs to the seventh one.
Behind its divider there is a stack of records. You search sequentially through them
until you find the Q,R,&S transaction.

Figure 1-9: Random Access Example

OUEEG CO

]L--------

ROOT

RHESUS INC

H-MK-00088-00

Example Using the Transaction File described in the last example, you want to find a record,
but all you know is its transaction number. Fortunately, the Third Alternate Key for
the file is transaction number. You open the index labeled "2" and look at its Root
card. Employing the technique illustrated in the previous example, you move from
the Root through the rest of the index. Behind the last divider, you find a record on
which the transaction number you are looking for is listed. Next to the number is the
code "7/5".
You reach into the back of the drawer, pulling the seventh tab forward. Then you
count the records behind it. The fifth one has the right transaction number on it.
You notice that the transaction was made with the firm of Q,R,&S.

Introduction to RMS-11

1-9

Utilities

Once you establish files and their records, you'd like to do some things with
each file as a whole:

Back up
The data in a file is valuable, or you would not keep it. You should have a
duplicate of your records in some other place in case something happens to
the original. Therefore you need the ability to back up a file quickly and
efficiently.
Restore
If something does happen to the original records, you must replace them
with the back ups quickly and smoothly.
Display
You need the ability to produce a list of your files, with their names and
other attributes.
Convert
Files do grow beyond your estimate, and to increase their usefulness, you
should change their organization, say, from a sequentially organized files
to ones with indexes. Or the opposite could happen to a file and you need
to make it simpler. Then, there are times when a file set up for one
purpose could be used for another application. In that case, you need to
copy the organization and contents of the original file into the new one,
changing some attributes.
Define
Then, when you've designed the file, you should also have a procedure for
creating the file.
Indexed File Load
Creating Indexed files can be complicated and time-consuming. You could
use a procedure that takes a file of any organization and produces an
optimal Indexed file quickly and efficiently.
That's a synopsis of data organization concepts. You should recognize them
from your experiences with paper. But now you're using a computer to organize your data: anything you can do at your desk, you ought to be able to do
with your computer.

1.2 RMS-11 Implementation of Data Organization
The software routines called RMS-11 organize data on your DIGITAL computer, implementing the concepts we have just discussed. RMS-11 interfaces
your data processing programs with the rest of the computer system.
Your computer system consists essentially of layers of software and hardware
(see Figure 1-10), each of which has a responsibility in the data management
process regulated by RMS-11:
• The hardware devices store the data.

1-10

Introduction to RMS-11

• The operating system controls the hardware to provide the record containers
called files.

• RMS-11 controls the internal structure of the computerized files, providing
file organizations and record formats, access modes, and operations.
• Your application program drives the process by initiating data processing
operations.

Figure 1-10: RMS-11 in Its Environment

KEY·

PROGRAM

RMS-11

~DATA FLOW

...... CONTROL FLOW
~TASK SEGMENT FLOW

H-MK-00085-00

The rest of this introduction discusses these system layers.

1.2.1 Hardware Data Structure
The prime storage device on today's computer systems is the disk drive 1 • The
disk itself consists of one or more circular pieces of metal (called platters) and
the drive is the mechanical and electronic equipment to read and write information on the platters.
1

Although magnetic tapes also provide significant data storage capability, they require sequential access. Therefore, in modern business environments, where fast random access is
important, disk drives are used for on-line storage. Magnetic tape, however, does have its
uses; details of magtape handling via RMS-11 are covered in Appendix F.

Introduction to RMS-11

1-11

Data is stored on disk platters magnetically, much the same as sound is
recorded on tape. The structure of that data is modular, expressed in hierarchical units (see Figure 1-11):

~-~.

Bit
A bit is the smallest storage location recognized by the hardware. A bit is
an area of the disk surface where magnetic orientation can be changed to
one of two recognized values, conventionally designated "O" and "1".
Byte
A byte contains eight bits and is frequently used to represent an alphanumeric character with the American Standard Code for Information Interchange (ASCII) codes. Other methods of representing data, particularly
numeric data, require two or more bytes at a time.
Sector
A sector consists of 512 bytes on most disks supplied by DIGITAL.
Track
A track consists of the sectors at a single radius on one disk platter. One
read/write head of the disk drive can access a track without changing
position.

-~,

Cylinder
A cylinder consists of the tracks at the same radius on all disk platters.
The disk drive's head structure can access a cylinder without changing
position.

1.2.2 Software Data Structure
The operating system software makes the computer's hardware available to
the user. As it turns out, this user is RMS-11, but it is RMS-ll's purpose to
complete the logical chain, that is, to make the capabilities of the computer
system available to you.
The following operating system components are involved in the data management process:
Device Drivers
Each device driver is software written for a specific type of hardware unit.
Drivers instruct their devices during data access operations.
File Processor
The file processor maintains the structure and integrity of data stored on
file-structured devices. It provides volume directory and space management functions, as well as translating RMS-11 data requests for the device
drivers.
Monitor
The monitor coordinates the other components of the operating system,
including the device drivers and the file processor.

1-12

Introduction to RMS-11

Figure 1-11: Physical Storage Structure

[]
I

I
I

BYTE=
8 BITS

't'11111w
.....

I SECTOR== 512 BYTES] '\I' I I I I I I
''

A TRACK IS COMPRISED OF
THE AREA AT A SINGLE RADIUS
ON ONE RECORDING SURFACE.

I I I I

II II

I I

I I I I I II I I
~-

' ''.!

I',
I

A CYLINDER CONSISTS OF
THESE TRACKS IN THE SAME
RADIUS ON ALL THE RECORDING
SURFACES.

NOTE
RECORDING OCCURS ON BOTH
SURFACES OF EACH PLATTER.
THE EXTREME TOP AND BOTTOM
SURFACES OF SOME DISK MODELS
ARE NOT USED FOR RECORDING.

REMAINDER OF VOLUME
/CONTAINING OTHER CYLINDERS.

H-MK-00096-00

Introduction to RMS-11

1-13

CONVENTION
The cover term monitor in this manual has the
same meaning as the following system-specific
terms:
System
Term
IAS
executive
RSTS/E
monitor
RSX-llM
executive

Disks As Logical Devices - To be independent of the disk drivers, the file
processor places a logical structure over the data on each disk (see Figure
1-12). In essence, the processor treats a disk as a logically contiguous series of
data units called blocks. A block contains 512 eight-bit bytes. Logical blocks
are numbered from 0 to n-1, where n is the number of blocks on the disk.

NOTE
Three On-Disk Structures (ODS) are currently supported by
the DIGITAL operating systems on which RMS-11 operates:
• RSTS/E has its own disk structure.
• IAS and RSX-llM use a standard named Files-11 ODS-1.
• VAXNMS supports both Files-11 ODS-1 and Files-11
ODS-2.

Figure 1-12: Logical Data Structure
LOGICAL
BLOCK
NUMBER

0
SECTOR 1,
TRACK 1-----1~.'<·:'.
LOGICAL
BLOCK
NUMBER
n-1

/
/

// /
//
//

H-MK-00095-00

PHYSICAL
STORAGE

1-14

Introduction to RMS-11

Files As Virtual Devices - The file processor supplies containers for blocks of
data. Because these logical structures serve the same purpose as paper files,
they are also called files.

The processor treats each file as a device containing virtually contiguous
blocks: it can ignore all blocks on the disk except those in the file being
processed. The processor assigns Virtual Block Numbers (VBNs) from 1 ton,
where n is the last block in the file.

NOTE
Logical block and virtual block describe the same physical unit
of storage; only the numbering scheme is different. Virtual
blocks have Logical Block Numbers (LBNs), but logical blocks
do not have Virtual Block Numbers unless they are allocated to
a file.

"-''

Virtually contiguous does not necessarily mean logically contiguous. There is
more than one file on a disk. As these files take room on the disk, there are
fewer contiguous logical blocks. Eventually the file processor creates or extends a file so that portions of it reside in different parts of the disk (see Figure
1-13). The blocks retain their serial Virtual Block Numbers, but they are no
longer logically contiguous.
Figure 1-13: Virtual-to-Logical-Block Mapping

LOGICAL
DEVICE

A A

A A B

.11

VIRTUAL
DEVICES

FILE A

I
I
I

~.,,,

/!
/·

.,..,

.' ..
..

I- - b

,..~·:.~.: ?'"

_,,> =-::
/.

.,,,,,

.·

,,r

·.
FILE C

A C C B

~->··"·_..;-~
- - . . : > >."'.'"· . .. ':;)

,,,·
. ..,,,·
.,,,:·
,..· ,,, .,,., ,,, ,,. .. :

.:
I:
. ..
./:.;,,·

k~·

77\

I .:i" ,,, ,,, ,..,.. .: ,,, ,.. ""' . ~·" "" .... .,,,,,
11.·,,,,
./
·"·
•/

FILE B

B B C C C C C C A A

KEY
- - - MAPPING FOR FILE A
- · - • - MAPPING FOR FILE B
· .... • ... MAPPING FOR FILE C
H-MK-00068-00

Introduction to RMS-11

1-15

Virtual to Logical to Physical Blocks - The device-independent file processor controls the virtual and logical structures applied to data and translates
from one to the other. For instance, if a system user requests access to a block
within a file, the file processor calculates the logical block on the disk that
equates to that Virtual Block Number. In doing so, the file processor takes
into account the fact that the virtual qlocks may not be logically contiguous.

After calculating the Logical Block Number, the file processor requests that
block from the disk driver for the device containing the file. The driver then
translates that request into the cylinder/track/sector, or physical block, location that the device hardware must read or write.

1.2.3 The RMS-11 Interface
The file handling components of the operating system do not handle records
as such. Your business is structured around the use of records within files .
. RMS-11 manages records, translating your requirements for the operating
system and vice versa.
RMS-11 does this by controlling the internal structure of the files supported
by the operating system with:
• Record formats
• File organizations
• Record access modes
• Record operations

NOTE
The following discussion is just an introduction. Details on
most topics are provided in the file organization-specific chapters later in this manual.
To meet the requirements for record formatting
discussed previously in this chapter, RMS-11 provides the following record
formats:
1.2.3.1

Record Formats -

Fixed
Every record must have the same length.
Variable
Each record can have a different length, but it may not exceed a maxim um record size you set for the file.
Variable-with-Fixed-Control (VFC)
Each record has a section that is always the same length and a section
that can vary in length. However, no record may exceed the maximum size
you set for the file.

1-16

Introduction to RMS-11

RMS-11 also supports. the following record formats to be compatible with
other PDP-11 file systems (see also Appendix A):
Stream
Each record is a contiguous series of characters, with no maximum length
set.
Undefined
Essentially no records are defined in the file. Each access operation reads
or writes a block.
You control the amount, content, and arrangement of data within records and
its entry and interpretation.
~ .2.3.2

Fiie Organizations -

RMS-11 provides three methods of organizing

records within a file:
" Sequential
" Relative
" Indexed
Sequential File Organization - Figure 1-14 shows how RMS-11 implements
the sequentially organized file described in Section 1.1. The organization is
defined as follows:

An RMS-11 Sequential file is a series of virtually contiguous records stored in
the order they were written.
Therefore, Sequential files can be used on any medium recognized by
RMS-11 (tape, disk, or unit record devices).

Pigure 1-14: Sequential File Organization
FOURTH RECORD WRITTEN IS
LOCATED BETWEEN
THIRD AND FIFTH RECORDS WRITTEN

END OF FILE

FIRST
SECOND
THIRD
FOURTH
FIFTH
SIXTH
RECORD RECORD RECORD RECORD RECORD RECORD
WRITTEN WRITTEN WRITTEN WRITTEN WRITTEN WRITTEN

•

RECORD

0-M K-00067-00

Relative File Organization - Figure 1-15 illustrates how RMS-11 implements the Relative file concepts discussed in Section 1.1. The organization is
defined as follows:

An RMS-11 Relative file is a series of record storage cells with a fixed size.
" The cell size is based on the length you specify as the maximum for any
record in the file.

Introduction to RMS-11

1-17

• RMS-11 numbers the cells consecutively from 1 ton, where n indicates the
last cell in the file. A cell number relates its location to the beginning of the
file.
• RMS-11 stores records in the cells and associates them with their cellnumbers.
Example Record number 1 is in the first cell.
Example Record number 17 is in the seventeenth cell.

Only one record can be put into a cell, but all cells do not have to contain
records.
You can use cell numbers to identify and access the records in the cells. The
cell numbers are then known as relative record numbers.
Relative files have two capabilities not available with Sequential files:
• Random access by record number
• Record deletion
With Relative files, you still have fast sequential access.

NOTE
You can store Relative files only on disks.

Figure 1-15: Relative File Organization
CELL
NUMBERS

n-1

•••

FIRST
SECOND
RECORD RECORD
WRITTEN WRITTEN

THIRD
RECORD
WRITTEN

0-M K-00066-00

Indexed File Organization - Figure 1-16 shows how RMS-11 implements
the Indexed file described in Section 1.1. The organization is defined as
follows:

An RMS-11 Indexed file contains data records sorted in ascending order by
Primary Key value and one or more indexes that point into the data records.
RMS-11 stores each record in an Indexed file according to the value of the
data in a part of the record itself. Specifically, when you create an Indexed
file, you must identify a section of each record as a Primary Key. Thereafter,
when you store a record in the file, RMS-11 puts it between a record with a
lower or equal Primary Key value and a record with a higher key value (see
Figure 1-16). You can also identify sections of the records as Alternate Keys,
but their values do not affect the placement of the records in the file.

1-18

Introduction to RMS-11

l~igure 1-16:

Primary Index in RMS-11 Indexed File

PRIMARY INDEX (EMPLOYEE NAME)'
ABLE

[ABLE

ELM AV

24379

JONES

MAIN ST

SMITH

19724

SMITH

PRIMARY KEYS

HOLT RD

35888

H-M K-00072-00

Each key provides a logical access path to locate a record within a file. You
can specify up to 255 keys for an Indexed file. For each one, RMS-11 constructs an index in the file. This structure embodies an exact and economical
search pattern for RMS-11, enabling it to locate any record rapidly.
Because RMS-11 stores records in ascending order, you have fast sequential
access to them.

NOTE
You can store Indexed files only on disks.
1.2.3.3

Record Access Modes -

RMS-11 provides three Record Access

Modes:
• Sequential Access Mode
• Random Access Mode
• Access by Record's File Address (RFA)
RMS-11 guarantees that every unit of data it retrieves is a record you (or
others with access to the file) put into it.
1.2.3.3.1 Sequential Access Mode - Record access starts at some point in the
file and continues with consecutive records. The location of the next sequential record is determined by the file organization.

Sequential Access to Sequential Files - Records in a Sequential file are
virtually adjacent. To retrieve a specific record using sequential access, you
must open the file and look through the records before the one you want (see
Figure 1-17). From that point, you can still get any record after the one you
just looked at, but if you want to retrieve a record before it, you must go back
to the beginning of the file.

Introduction to RMS-11

1-19

Figure 1-17: Program Sequentially Reading a Sequential File
RMS-11

PROGRAM

READ
NEXT
RECORD

PROGRAM
H-MK-00094-00

When you sequentially insert records into an Indexed file, they must be
ordered in nondescending sequence by Primary Key. RMS-11 inserts the
records into the file based on those values, using the same procedure it uses
for random insertion (see Random Access to Indexed Files, Section 1.2.3.3.2).
1.2.3.3.2 Random Access Mode - You, rather than the organization of the
file, establish the order in which records are processed. You must specify a
record identifier with each random access. Each record access is independent
of the previous record used. Successive operations in the random mode can
identify and access records anywhere in the file.

You cannot use Random Access Mode with Sequential files. Both the Relative
and Indexed file organizations permit random access to records.
Random Access to Relative Files by specifying (see Figure 1-21):
t1

a relative record number (RRN)

whether you want:

You can access a record in a Relative file

- only the record with the specified RRN: equal match on the RRN

Introduction to RMS-11

1-21

- the first record after the specified RRN: greater than match on the RRN
-

the record with the specified RRN or if that cell is empty, the first record
after the specified RRN: greater than or equal match on RRN

RMS-11 locates a file cell according to this criteria and checks for a valid
record:
• Read a record
If there is a valid record in the cell, RMS-11 returns it to you; otherwise,

RMS-11 gives you an error code.
• Writing a record
If there is a valid record in the cell, RMS-11 writes over it only in special

circumstances; otherwise, RMS-11 gives you an error code.
If there is no valid record in the cell, RMS-11 writes the record into the

cell; from then on, you randomly read that record with the relative record
number.

Figure 1-21: Program Randomly Reading a Relative File
START OF

----------

PROGRAM

__i::;;;.;.~-;

---~--

1. READ RECORD 6 1--+-----1
RMS-11

2. READ RECORD 21--+-----\

~EMPTY
Q-MK-00080-00

The circled numbers indicate the order of the access operations; in this case,
record number 2 is "F".
Random Access to Indexed Files by identifying it as follows:

You can read a record in an Indexed file

• a key number (Primary, First Alternate, and so on)
• a value
• value match (equal to, greater than, or either)
• number of characters for value match
The key number determines the index RMS-11 follows. The number also
indicates the section, or key field, of the records RMS-11 must compare to the
specified value for the length indicated by the match criteria.

1-22

Introduction to RMS-11

'-"'

Example You have an Indexed personnel file; the Second Alternate Key is the social security
number. You can read the first record that contains a social security number starting with "560" by identifying it as follows:

• key number 2
• value in key field must be equal to "560" for first three characters
Figure 1-22 shows the search RMS-11 makes during this read operation.
However~ to insert a record randomly, you do not have to indicate a key.
RMS-11 uses the Primary Key value in the record to place the record in the
file. Then, RMS-11 revises the index(es) if necessary. The example for Indexed File Organization in Section 1.1, shows how RMS-11 determines record
location for all types of record access.

RMS-11 establishes a unique
identifier within a disk file for every record it writes. This Record's File Address (R.FA) remains valid for that record alone for the life of the file. If the
record is deleted, its RFA is not reused; an attempt to read that record returns
the information that the record was deleted.
1.2.3.3.3 Access by Record's Fiie Address -

RF A access is the fastest way to find or read a record randomly:
• For Sequential files, it is the only way to access records randomly.
• For Relative files, it bypasses relative record number processing.
~·

• For Indexed files, it eliminates reading the index.
To read a record by RFA, specify the address for that record (see Figure 1-23).
You cannot use RFA access to write a record.

1.2.3.3.4 Changing Record Access Modes - You can change Record Access
Mode at any time while you are accessing a file. The file organization and
storage medium must support the access mode selected. Generally, you use
Random Access Mode or Access by RF A to access the first record of a series
and then use Sequential Access Mode to access the records in that series.
Example A personnel file has department code as one of its Alternate Keys. You can produce
a report on a department by randomly accessing the first record in the file with a
specific departmental code and then using Sequential Access Mode to read consecutive records.
Example The Sequential file in Figure 1-23 is written in account-number order, one record per
transaction; each account has more than one record. To list the transactions for
account C, open the file and sequentially read each record until you find the first one
for account C. Then you start the report.

However, if you had saved the RFA for account C's first record when you wrote it,
you could access that record with its RF A and then switch to Sequential Access
Mode to produce the list.

Introduction to RMS-11

1-23

Figure 1-22: Program Randomly Reading an Indexed File

PRIMARY INDEX-.

Fl AST AL TERNA TE
INDEX
...

SECOND ALTERNATE
INDEX--~

PROGRAM

READ RECORD WITH
560 IN SECOND
ALTERNATE KEY

RMS-11

NOTE
THE FILE IS STORED
ON A CYLINDER.
Q-MK-00100-00

1-24

Introduction to RMS-11

Figure l-23: Sequential Account File
A. DAT A ENTRY
PROGRAM
WRITE
RECORD

B. REPORT GENERATION
PROGRAM
RETRIEVE
ST RECORD
ACCOUNT

·~·

c.
RETRIEVE RECORDS
FOR ACCOUNT C
SEQUENTIALLY ..

H-MK-00092-00

RFA =RECORD'S FILE ADDRESS

You write programs to process the data units
you designate as records. Processing involves record operations that your
program initiates and RMS-11 performs. Through RMS-11, your programs
cam:
1.2.3.4 !Record Operations -

• read a record, retrieving data from disk. These operations include:
Find
RMS-11 locates and retrieves the specified record according to the requirements of the Record Access Mode. However, RMS-11 does not
make the record available to your program.
Get
RMS-11 locates and retrieves the specified record according to the requirements of the Record Access Mode. Then RMS-11 makes the record
available to your program.
• write a record, storing data on disk. These operations include:
· Delete
You mark a record in a file, indicating that the record no longer exists.
The space used by the record can be reclaimed by future operations
according to the requirements of the file organization and the record
format.

Introduction to RMS-11

1-25

NOTE
You can delete records at the end of a Sequential file
only, by truncating the file (described in "Truncate,"
Section 3.3.1.8).
Update
You replace a record in a file with a revised version.
Put
You store a new record in a file, according to the Record Access Mode.

NOTE
RMS-11 requires that a successful find or get operation
precede a delete or update operation. However, some
higher level languages hide this prerequisite.
Example Using PDP-11 COBOL, you can use a REWRITE statement without establishing the record being updated with a
READ or START statement.

• perform other operations, including:
Connect
You make the records of the file available to your program in preparation for a stream of operations. This operation is hidden in most higher
level languages.
Disconnect
You terminate a stream of operations, making the buffers assigned to
the stream available for other operations.
Flush
You ensure that all records you have written, updated, or deleted are
written to disk before you terminate or change processing.
Rewind
You return to the beginning of the file for sequential access.
"Buffers in Record Operations," Section 1.2.4.2, contains details of the processes RMS-11 uses during these operations.
1.2.3.5 RMS-11 Utllltles - DIGITAL provides you with programs, called utili-

ties, that use RMS-11 to accomplish standard file-related jobs. Aligned with
the requirements discussed in Section 1.1, these utilities are:

RMSBCK
Used to back up your data, the RMSBCK utility copies files in a special
format that cannot be mistaken for the original data and therefore cannot
be altered during normal operations.

1-26

Introduction to RMS-11

RMSRST
Used to read RMSBCK's special format, the RMSRST utility replaces
your data files with back-up versions whenever you want.
RM SD SP
The RMSDSP utility lists files and their attributes at your request.
RMSCNV
Used for any combination of RMS-11 file organizations and record formats, the RMSCNV utility copies one file into another, while preserving
the source file.
RMSDEF
During an interactive process, the RMSDEF utility helps you define attributes for a file and then creates it.
RMSIFL
Bypassing normal RMS-11 methods, the RMSIFL utility quickly loads an
RMS-11 Indexed file with records from a file you designate, optimizing
the structure of all indexes.

1.2.4 Record Processing Environment
RMS-11 processes records at the command of your program, operating in
virtual conjunction with its executable form. In fact, to the operating system,
RMS-11 is part of your program (see Figure 1-24).
Figure 1-24: System Memory Layout
DISK
DRIVERS

FILE

PROCESSOR

1/0
PAGE

OPERATING SYSTEM

I
I
I

FILE PROCESSING ENVIRONMENT

RECORD PROCESSING ENVIRONMENT

I
I
I

PROGRAM

RMS-11

1/0 BUFFERS
INTERNAL TABLES

A. UNSHARED RMS-11
H-MK-00070-00

(PART 1 of 2)

(continued on next_ page)

Introduction to RMS-11

1-27

Figure 1-24: System Memory Layout (Cont.)

E..,,

DISK
DRIVERS
FILE PROCESSOR

RMS-11
RESIDENT
LIBRARY

L___,,..---J ·........._
OPERATING SYSTEM

1/0
PAGE

TASK

I ·'-..
'-..

·-..........

""'-.

'-.,_

I
·i

'·........._I

·,.

'-.

\ . '--.

I
I

/ FILE PROCESSING ENVIRONMENT
'-.,

'-..

'--

/ RECORD PROCESSING ENVIRONMENT

., I

'I.

. . . . __
'-..

I '-.,

'·

"''

PROGRAM

B. SHARED RMS-11
(PART 2 of 2)

SHARED RMS-11
· - - 1/0 BUFFERS AND
INTERNAL TABLES
H-M K-00070-00

1.2.4.1 Using RMS-11 - RMS-11 is a set of file access routines. These
routines implement a standard file structure and interface across DIGITAL
operating systems and programming languages.

You use these routines by combining them with a program you have written in
a language that implements RMS-11 (see Appendix B). You must write this
program so that it uses the appropriate RMS-11 functions, obeying the language syntax. Then you convert your program to object code, through either a
compiler or an assembler.
Once your program is in object form, combinejt with the RMS-11 routines via
a utility called the Task Builder. This software converts object modules to an
executable form called a task. In the process, the Task Builder not only
combines different object modules, but can also arrange the task so that some
executable modules overlay each other when the task is run.
You can combine RMS-11 routines with your object code in either of the
following ways:
• in the task itself-with nonoverlaid routines or a disk-resident overlay
structure
• in memory-resident overlays-a form apart from your task
The primary difference between the techniques is that memory-resident overlays can be shared among programs; the other forms cannot, that is, each
program has its own copy of the routines. In addition, memory-resident over-

1-28

Introduction to RMS-11

lays eliminate the 1/0 operations needed to bring disk-resident overlays from
disk; therefore, your tasks run significantly faster.
Bither way, your task takes the logical form shown in Figure 1-25. Your
program code exists in one part of the task. The RMS-11 routines run in
another part. When your program performs an RMS-11 operation, it sets up
the parameters and data and then calls the RMS-11 routine. Control jumps to
that part of the task, the routine runs to completion, and control returns to
your program.

Figure 1-25: RMS-11 Task Structure
fS1zE DEPENDS ON:- -

I• NUMBER OF FILES OPENED SIMULTANEOUSLY I
1•

BUCKET SIZES

~ NUMBE~f'._RECrD _!CCESS STREAMS _ _
USER BUFFERS ~

(

110

BUFFERS
VIRTUAL
MEMORY

RMS-11

PROGRAM

INTERNAL
CONTROL
STRUCTURES

fS1zEoEPENDs

O"N:-

- ~

l • RMS-11 FUNCTIONS USED l
~ ~VERLA0~U~TURE~~E~

H-M K-00069-00

Also part of the task are storage buffers that usually come in three forms:
User Buffers
Your program usually has room to store one record for each open file. This
buffer is available to your program and the data in it can be manipulated,
read, changed, used for calculations, and so on. RMS-11 can also access
this buffer.
1/0 Buffers
For each file your program has open, RMS-11 requires at least one 1/0
buffer. All data meant for or arriving from disk is stored here:

• RMS-11 requests the file processor to move block(s) from disk into this
buffer to satisfy its or your program's requirements. Each request from a
file always specifies the same number of blocks, termed collectively an
1/0 unit. The size of the 1/0 unit depends on the file organization and
your file design.
• RMS-11 moves records between the 1/0 buffer and the user buffer. Your
program can also access this buffer in restricted circumstances.

Introduction to RMS-11

1-29

Control Structures
RMS-11 control structures communicate with your program and with
each other.
1.2.4.2 1/0 Buffers In Record Operations -

These buffers are used in record

operations in the following ways:
Delete
Your program indicates the record to be deleted. RMS-11 has the record's
I/0 unit in memory because a delete operation must be preceded by a
successful get or find operation. RMS-11 then changes the record in the
I/O buffer to indicate that the record is deleted and requests the file
processor to rewrite the 1/0 unit on disk. Finally, signalling success,
RMS-11 returns control to the program.

NOTE
The space in the deleted record is re-used according to the
requirements of the file organization and the record format.
Find
RMS-11 follows the process used for a get operation, except it does not
move the record from the I/0 buffer to the user buffer.
Flush
Your program tells RMS-11 to write an I/O buffer to disk if the buffer has
not already been written.
Get
Your program specifies the record to be read from disk and the user buffer
in memory where RMS-11 should put it. RMS-11 attempts to locate the
record in the file, using techniques required by the file organization, record
access mode, and the program's request. Each technique involves the
movement of one or more 1/0 units of the file into the 1/0 buffer, where
RMS-11 looks for the record. If RMS-11 does not find the record specified,
it returns the appropriate error code.
If RMS-11 finds the record, it normally moves the record from the 1/0

buffer to the user buffer and signalling success, returns control to the
program.
Put
Your program specifies the user buffer containing the record. RMS-11
locates the point in the file where the record belongs and has the file
processor bring that 1/0 unit from disk into the 1/0 buffer. Then RMS-11
moves the record from the user buffer to its place in the 1/0 buffer.
RMS-11 requests the file processor to rewrite the 1/0 unit (including the
new record) to the disk and signalling success, returns control to the program.
Update
Your program specifies the user buffer containing the revised record.
RMS-11 has the record's 1/0 unit in memory because an update operation

1-30

Introduction to RMS-11

must be preceded by a successful get or find operation. RMS-11 then
moves the data from the user buffer to the I/0 buffer, writing over the old
record. Finally, RMS-11 has the file processor write the 1/0 unit to disk
and signalling success, returns control to the program.

1.2.4.3 Record Access Streams - Before your program can access records in a
file, it must open that file and connect a Record Access Stream to it. This
stream is a channel between your program and the file. You use the stream for
each record operation.

NOTE
Most higher level languages do not provide Record Access
Streams and the connect operation at the user level. They use
these facilities to implement their own file access techniques.
Context of a Record Access Stream - Each stream processes record operations for one record at a time. RMS-11 keeps track of the stream's position in
a file. This position is called a context and consists of the following entities
(see Figure 1-26):

Current Record
The Current Record is:
• established by a successful find or get operation.
• the target of the following operations:
Delete
RMS-11 marks the Current Record as deleted.
Get Immdediately Preceded by a Find
Normally, RMS-11 moves the Current Record into the user buffer.
Current Record does not change.
Truncate (Sequential files only)
RMS-11 logically deletes the Current Record and all records following it in the file by establishing a new end-of-file position at the first
byte of the Current Record.
Update
RMS-11 replaces the Current Record with the one in the user buffer.
Since only get or find operations set the Current Record, one of these operations must precede an update or delete operation. Other operations leave
the stream without a Current Record. RMS-11 rejects any update or delete
operation attempted without a Current Record.
Next Record
The Next Record is the target of a sequential get, find, or put operation
(put operations on Sequential and Relative files only).
• If you specify a get operation, RMS-11 locates the Next Record and puts it
in the user buffer.

Introduction to RMS-11

1-31

• If you specify a find operation, RMS-11 locates the Next Record.
• If you specify a put, RMS-11 moves the record in the user buffer into the
file at the position of the Next Record.

Next Record is affected by record operations in specific ways explained
later in this manual.
Comparable to the marker you leave in a paper file, stream context is important to sequential access only. Record operations affect context in ways designed to facilitate normal processing.
Example When you update records in a paper file, you must locate the record first. You either:

• take the form out of the file so that you can look at it (a get operation)
• verify that the form you've located is the proper one by checking the relative
record number or key value (a find operation)
In this process, you are establishing the Current Record.
Then, when you update the record, you either change the one you've gotten or
replace the one you've found. And you insert the new version where the context
(Current Record) indicates.
In addition to setting Current Record with your get or find operation, you establish
the position of the Next Record. Then, after you complete the update operation, the
context indicates which record you locate next.

NOTE
The specific effects of each record operation on the stream
context depend on file organization. Each of the organizationspecific chapters describes these effects.

Figure 1-26: Record Operations and Stream Context
OPERATIONS
UPDATE
DELETE
GET
FIND
WRITE
Q-M K-00082-00

Multiple Record Access Streams - A stream can handle only one record at a
time, but you can connect more than one Record Access Stream to a Relative
or Indexed file if you want to:

• process more than one record in a file at a time with asynchronous record
operations (see Section 1.2.4.4)
• maintain more than one context during the processing of a file

1-32

Introduction to RMS-11

··~

Each stream represents an independent, concurrently active sequence of record operations. Again, most higher level languages hide this capability.
Example A program opens an Indexed file and connects two Record Access Streams. In one
stream, the program uses the Primary index to access records in random mode. In
the other stream, it sequentially gets records in the order specified by an Alternate
index.

1.2.4.4 IAS/RSX-11 M Asynchronous Record Operations - Within each Record
Access Stream, your program can perform any record operation either synchronously or asynchronously. In synchronous operations, RMS-11 returns
control to your program after the operation ends, either successfully or with an
error.
When you execute an asynchronous operation, RMS-11 may return control to
your program before the operation is finished. The program continues processing while the physical transfer of data between disk and memory is carried
out. However, you must not initiate another record operation on that stream
until the first operation ends. See your language documentation for asynchronous techniques.

NOTE
If you intend to use asynchronous RMS-11 record operations

and/or Asynchronous System Traps (AS Ts) in other parts of
your program, see the section on your operating system in
Appendix A.
1 . 2.4.5 Record Transfer Modes - Your program can manipulate the data in a
record while it resides in the user buffer or while it is still in the I/0 buffer.
These choices are called Record Transfer Modes. The organization-specific
chapters discuss these modes more thoroughly.

1.2.5 File Processing Environment
Now that we have discussed the data management process, layer by layer,
from hardware to your program, let's examine more details of RMS-ll's
relationship with the operating system.
RMS-11 manipulates files so that it can process records. The file processing
environment involves RMS-11 with complex flows of data, control, and overlay segments (see Figure 1-27). Although its requests initiate activity in the
operating system and its devices, RMS-11 is not aware of the file management process.
1.2.5.1 File Processor - Each operating system has a file processor:
• Files-11 Ancillary Control Processor (FllACP) on IAS/RSX-llM
• File Processor (FIP) on RSTS/E
The file processor performs I/0 and other operations on files. RMS-11 must
make requests in a certain format so that the file processor changes the files
properly. Thus, the file processor's operations, while logically invisible to
RMS-11~ can affect the performance of your program.
However, the file processor is not concerned with the data contents of a file. It
only knows Virtual and Logical Block Numbers, directories and other support

Introduction to RMS-11

1-33

information, and the disk drivers involved. Therefore, RMS-11 can manipulate the contents of a file as long as it makes proper requests to the file
processor. In this manner, RMS-11 maintains the following file formatting or
structures:

Block Spanning - Basically, RMS-11 lays out the records in a Sequential
file one right after the other, in the order they are written. However, you must
decide whether those records can cross block boundaries. When records span
blocks, RMS-11 can pack them with optimal density into the file because a
record can be stored in one or more blocks (see Figure 1-28). When block
boundaries restrict records, each one must be less than 512 bytes long, and
RMS-11 might leave unused bytes at the end of each block.
Figure 1-27: RMS-ll's Environment

DISK DRIVER
----FILE PROCESSOR

A. .I

MONITOR/EXECUTIVE

KEY
<::=::> DAT A FLOW
........ CONTROL FLOW
~TASK SEGMENT FLOW

PROGRAM

RECORD
BUFFERS

RESIDENT

RMS-11

H-MK-00085-00

Buckets - The I/0 unit for Relative and Indexed files is called a bucket. A
bucket consists of one or more blocks that RMS-11 treats as a unit. Indexed
files, in fact, consist of buckets formatted with control information. Records
can cross block boundaries, but they cannot cross bucket boundaries.
When RMS-11 initiates an I/O operation for a file of one of these organizations, it requests the file processor to move a bucket. Since buckets are an
RMS-11 concept, the request specifies the Virtual Block Number for the first
block in the bucket and the size of the bucket in bytes. Note that buckets are
fixed within a file; once created, a bucket contains the same virtual blocks at
all times.

1-34

Introduction to RMS-11

l~igure 1-28:

t..._BLOCK

]

Records Spanning Blocks

·1 •

BLOCKJBLOCKIBLOCKIBLOCKIBLOCKI

_____________
! _ _ _ __

.A. RECORDS LESS
THAN 512
BYTES

I~.

RECORDS GREATER
THAN 512
BYTES

C. VARIABLE-LENGTH
RECORDS
H-MK-00071-00

The operating systems limit bucket sizes:

Operating
System

Maximum
Bucket Size

IAS
RSTS/E
RSX-llM

32 blocks
15 blocks
32 blocks

NOTE
The I/0 unit for Sequential files is not the bucket, but the
block. You can adjust the block count for each Record Access
Stream, so that more than one block can be moved during each
1/0 operation.
Areas - Maintained and used by RMS-11, areas are portions of an Indexed
file that are treated independently for initial allocation, extensions, placement, and bucket sizes. Like subfiles, but invisible to the operating system,
areas allow you to divide Indexed files logically into separate units for each
index and for the data records. You do this to improve performance.
Placement Control - Through the file processor, RMS-11 allows you to
place a file, as a whole or by areas, on a disk at specific location(s). You do
this to :improve performance, taking advantage, for example, of tracks and
cylinders.

Introduction to RMS-11

1-35

1.2.5.2 File Sharing - Timely access to critical file~ often requires more than
one program to use those files at the same time. With the help of the file
processor, RMS-11 enables programs to share files.

The way programs can share a file depends on the file organization:
• With the exception of magnetic tape files, every RMS-11 file can be shared
by any number of programs for read-type operations.
• Only one program at a time can access a Sequential file for write-type
operations, while multiple writing programs can share Relative and Indexed
files.
File sharing is controlled by the programs and by the order that the programs
open a file. Basically, the first program to open a file sets the sharing type;
programs attempting to open the file after that generally must specify the
same type of sharing. More details on file sharing are provided in Chapter 2.
Although the file processor does most of the work,
RMS-11 provides the following file-level functions:
1.2.5.3 File Operations -

Creating a File
In addition to the file specification, RMS-11 passes the following information to the file processor when it creates a file:
• An initial allocation of blocks for each area in the file. You specify both
the areas and their allocations in your instructions to RMS-11.
• The specific locations on a device where the processor should allocate
those blocks. You also supply this information.
• The following file attributes:
File organization
Record format
Forms control
Record size
Number of virtual blocks in the file
End-of-file (Sequential files only)
Bucket size (Relative and Indexed files only)
Default extension quantity
RMS-11 stores the other file attributes, such as key and area descriptions for Indexed files, in the file (see "File Attributes," Section 1.2.5.4).
Opening a File
RMS-11 initiates access to the specified file, reading its attributes.
Extending a File
RMS-11 requests the file processor to add blocks to a file's allocation, in
two circumstances:
• A program explicitly directs the extension. RMS-11 passes the request
almost directly to the file processor, using the extension quantity supplied by the program.

1-36

Introduction to RMS-11

~.I

"'-".·

• A put or update operation cannot be completed because there is not
enough room in the file. RMS-11 requests that blocks be added to the
file, using either the default extension quantity or if that is zero, a
minimum number of blocks depending on the file organization.
Closing a File
RMS-11 writes all I/O buffers to the file, if they haven't already been
transferred, and terminates access to the file.
Erasing a File
RMS-11 requests the file processor to delete the file from the device directory and release its blocks for re-use. You can erase a file you or other
programs are accessing. However, the file processor does not actually erase
the file until all accessing programs close it.
1.2.'5.4 Fiie Attributes - When you create an RMS-11 file, either through a
program or an RMS-11 utility, you must specify the following information:

Medium
Your selection depends on the file's organization. You can create permanent Sequential files on disk devices or magnetic tape volumes. You can
also write transient files on devices such as line printers and terminals.
However, RMS-11 restricts Relative and Indexed files to disk devices.
File Specification
The name you assign to a new file enables RMS-11 to find the file later.
You follow the file specification conventions for your operating system; see
also Appendix A.
Protection
RMS-11 also allows you to assign a protection code to a file when you
create it; again, the format of this specification depends on the operating
system.
File organization
You have a choice of three organizations described in "File Organizations," Section 1.2.3.2: Sequential, Relative, and Indexed.
Record format
.
Your choice of the record formats described in "Record Formtts," Section
1.2.3.1, is restricted by the file organization (see Table 1-1):

Table 1-1: Record Formats and File Organizations

[

Record Format
File
Organi:iation

Fixed

Variable

VFC*

Stream

Sequential

Yes

disk only

I
I

Undefined

Yes

Relative

Yes

Indexed

Yes

* Variable-with-Fixed-Control

No
No

1
Introduction to RMJ-11
I

1-37

Record size
The meaning of the record size information depends on the record format:

For fixed-length records, record size is the same length for every record in
the file. RMS-11 rejects any write-type record operation that specifies a
record of the wrong size.
For variable-length and stream records, record size is a maximum length.
RMS-11 rejects any write-type record operation using a record size greater
than the maximum-unless. the maximum is zero; then RMS-11 does not
check the length of records added to the file. RMS-11 also keeps the
length of the longest record actually stored in the file.
For variable-with-fixed-control records, there are two size specifications:

• length of the fixed control area
• maximum length of the variable area
~~

RMS-11 treats these specifications the way it treats the sizes for fixedlength records and variable-length records, respectively. RMS-11 also
keeps the length of the longest record actually stored in the file.
Block spanning (for Sequential files)
You decide whether or not records can cross block boundaries.
Bucket size (for Relative and Indexed files)
You establish the number of blocks in each bucket. Bucket size impacts
performance.
Maximum record number (for Relative files)
If you set a nonzero Maximum Record Number (MRN), RMS-11 rejects
any record operation using a higher relative record number. If you establish an MRN of zero, RMS-11 does not check relative record numbers.
Keys (for Indexed files)
You must decide the following:
• number of keys
• position and size of each key
• data type for each key (including string, two- and four-byte integer and
binary, and packed decimal)
• whether records can duplicate key values
• whether Alternate Key values can change during update operations
• null key value for Alternate Keys
Areas (for Indexed files)
You must decide the following:
• the number of areas in the file

1-38

Introduction to RMS-11

• what logical portions of the file go in which areas
I

• the fill number for each area (space and performance op~imization)

Forms control
I
You can specify two types of forms control for records of a1y format in a
file of any organization:
Carriage Control
When records from the file are written directly to a unit ecord device,
the device driver puts a line feed character in front of th~ record and a
carriage return character after the record before pas~ing it to the
device.
I
I
Example You use RMSCNV to write the records of an Indexed file tt a line printer. If
you have specified carriage control for that file, the re corf s are printed on
separate lines. If you have not, the records are printed ontinuously, the
only breaks coming at the physical ends of lines.

FORTRAN
I
When records from the file are written directly to a unit tecord device,
the device driver interprets the first byte of each I record as a
FORTRAN forms control character.
·
You are not required to specify either type of forms control
Default extension quantity
You specify how many blocks you want RMS-11 to add to the file when
each allocation has been completely used for data storage. RMS-11 extends the file automatically when it needs space to complete n operation.
CONVENTION
The cover term file directory in this manual
has the same meaning as the following systemspecific terms:

System
IAS
RSTS/E
RSX-llM

Term
directory entry and file header(s)
file directory
directory entry and file header(s)

During the creation process, RMS-11 stores this information, called the file
attributes, in the file directory and for Relative and Indexed files, in the first
blocks of the file (called the Prologue).
NOTE
Attributes also include the file's current size, in blocks. You
may specify an initial allocation quantity when you create the
file, but this initial size probably changes as you use the file.

Introduction to RMS-11

1-39

After creation, for the life of the file, RMS-11 gets information about a file
from the file itself. This ability gives you several advantages:

• The file will not change its characteristics.
• You can design your RMS-11 files off-line. No program accessing the files
need specify attributes (except those required by the higher level
languages), because RMS-11 uses only a file specification from a program
when it opens a file. The files act as virtual devices for the programs.
• You can open an RMS-11 file with only its file specification. After that,
RMS-11 enables you to read the file attributes. You can write your own
program or use the RMSDSP utility to display those attributes.

1.2.6 Bypassing Record Processing
Finally, your program can bypass RMS-11 record processing and process any
RMS-11 file block-by-block in a mode called Block 1/0. However, RMS-11
requires files that will be written using Block 1/0 to be created with the
following attributes:
• disk or magnetic tape medium
• sequential organization
• undefined record format
However, you can read an RMS-11 file with Block 1/0 regardless of the organization or record format.
Using Block 1/0, your program reads or writes multiple blocks of the file by
identifying a starting Virtual Block Number and the number of blocks affected. Your program, of course, must interpret the contents of the blocks once
RMS-11 retrieves them.

1-40

Introduction to RMS-11

Chapter 2
A~•plication Design

You're writing an application. You want a program or a set of programs to
take in data, process it, store it, update it if necessary, and at intervals output
it in the proper formats.
You want all this to happen simply, quickly, and accurately. You must therefore take the time to design your application with RMS-11 considerations in
mind. These considerations include initial allocation, record format, overlays,
key selection, disk usage, and others.
If you don't consider RMS-11, you won't get the best performance possible
from your application, and you'll probably get less performance than Luck
would allow because of the defaults you're accepting without knowing it (see
HWhen To Design" in this chapter).
Example If you do not design your file, you could end up with a file like one user did:
The first time he created the file, he used a higher level language program and took
all defaults. Then he loaded records into the file: the process was quite lengthy.
However, he re-examined the file and recreated it, applying a couple of design
considerations. With the new file, the record insertion process went ten times faster.
Example If you do not understand the implications of RMS-11 file structure, you could end
up like some users accustomed to programming with BASIC-PLUS Record 1/0:
They picked up the facts that RMS-11 uses 15 bytes of control data in each bucket
and seven bytes of control data for each fixed-length record (more in Chapter 6).
Then, because they were used to working with whole blocks, they set up single-block
buckets (512 bytes) and subtracted RMS-11 overhead (22 bytes) to come up with a
record size of 490 bytes.
But when they used those files, the users were alarmed to see them grow at high
rates. They had not read that RMS-11 preserves its fast sequential and Alternate
Key access during random insertions by moving records and leaving behind sevenbyte pointers (more in Chapter 5). Therefore, when one of those 490-byte records
was moved, it left behind seven bytes, which meant that no other record fit into that
bucket. Soon the file was filled with practically empty buckets that could not
be used because the designers did not allow for the full implications of RMS-11
structure.

2-1

Example If you slap together an application with a higher level language, you probably don't
worry about RMS-11. In this process, you accept the language's concept of design, if
any. The chances are good that the defaults the language uses in its interface with
RMS-11 are not suited for your application.

2.1 When To Design
There are two times to design an application:
• Before you write the application, especially if you have:
-

large file (s)

many users simultaneously accessing the file(s)

a high level of activity (many records read, written, updated, or deleted
in a given time period)

• After you write the application, if you're not happy with its performance.
Often, poor performance results from default values. You can often find
improvements by studying the nature and source of the defaults and how
they affect the structure of your application and your file.
Basically, defaults have three sources:
Source Language Compilers
In many instances, source language compilers such as PDP-11 COBOL
or BASIC-PLUS-2 supply default values for RMS-11 file attributes
and/or facilities.
Example RMS-11 does not calculate an optimal bucket size for Indexed files. Rather,
the program creating the file must specify a bucket size. When that program
is the product of a compiler, the bucket size can be explicitly specified in the
source code or it can be implicitly set by the compiler, using a default value.

Specifically, PDP-11 COBOL provides the BLOCK CONTAINS clause in
the file-description-entry. You can use this clause to set the number of bytes
or the number of records in a bucket. However, if you do not include this
clause, the PDP-11 COBOL compiler sets the bucket size to the minimum
disk blocks required to contain one record.

RMS-11
The intratask interface between the RMS-11 routines and your program
has the same structure in all tasks, regardless of their source, PDP-11
COBOL, RPG, MACR0-11, and so on. This interface consists of control
blocks (see the RMS-11 MACR0-11 Reference Manual for details). The
information provided by your program in these blocks effectively
controls RMS-11, causing it to create, open, access, and close files.
However, when explicit information is not provided, RMS-11 uses its
default values.
Operating System
RMS-11 acts as a middle man between your task and the operating
system. As such, RMS-11 can supply control information for system
functions such as protection codes. However, if RMS-11 supplies no
control data, the system uses its defaults.

2-2

Application Design

2.2 Design Considerations
When you design your application, you are primarily concerned with four
things:
Speed
You want to maximize the speed with which the programs process data.
Space
You want to minimize the room for the data and the task on disk and the
memory the task takes to run.
Shared Access
You want your data to be exactly as accessible to the people using the
computer system as necessary, no more, no less.
Ease of Design
You do not want to spend more time than necessary writing the application.
Remember, the importance of design is proportional to the complexity of the
file organization. That is, design is least important for applications using
Sequential files and most important for applications using Indexed files.

2.2.1

Maximize Speed First

You can make many performance (speed) decisions before you have to consider anything else. Therefore, the first criterion to apply throughout the
design process is:
MINIMIZE I/0 TIME
The mechanics of the mass storage devices on your system consume most of
the time for any RMS-11 operation. The memory-resident routines that prepare the data for I/0 or process it afterwards are very much faster (one to
three orders of magnitude).

An application's entire environment (see Figure 2-1) affects I/0 time:
File structure
A variety of file attributes impact I/O time, including:
bucket size
number of keys
number of duplicate key values
initial file allocation
default extension quantity
File size
The number of records in the file affects the VO operations required to
scan a file sequentially or follow an index.
Program
Your program impacts I/0 time by requiring I/0 operations for file operations (open, close, and so on), record operations (get, put, and so on), and
overlays.

Application Design

2-3

RMS-11
The RMS-11 routines can be overlaid.

~.1

File processor
Besides requiring overlay segments from disk, the file processor can also
request I/Os to map virtual blocks of the file to logical blocks on the
storage device.
Device hardware
The storage device is the primary contributor to the length of an I/0
operation. The type of device chosen (moving-head, fixed-head, and so on)
to contain the task and the data files is crucial to I/0 performance.

Figure 2-1: Time Factors in an 1.''J Operation

DEVICE

H-M K-00084-00

2.2.2 Reduce Space Requirements
RMS-11 requires space for three reasons:
• to store data in a file
• to store the RMS-11 routines
- on disk when they're not in use
- in memory when they're being executed
• to buffer data in memory while the task runs

2-4

Application Design

Data Storage Space - The space RMS-11 requires to store data is proportional to the organization of the file-and the processing capabilities of that
organization:
Sequential File Organization
RMS-11 adds to the size of your data an empty byte, if necessary, to align
each record with a word 1 boundary. Also, when the file contains variablelength records, RMS-11 adds a record-length field to each record.

Relative File Organization
RMS-11 constructs a series of record storage cells based on the length of
the records. The cells are one byte longer than the fixed size of fixedlength records or three bytes longer than the maximum size specified for
variable-length records.
Indexed File Organization
RMS-11 adds to your data:
• an index for each defined key
• fifteen bytes of formatting information for each bucket
• a seven-byte header for each record
• a record-length field for each variable-length record
• other overhead of varying lengths for records RMS-11 moves during file
activity and for deleted records
You should keep the size of records to the minimum required for your application.
Task Size - The space RMS-11 routines occupy in a task depends on the
method you use to link the routines with your program. See "Task Building
with RMS-11 Routines," Section 8.1, for more details.
Buffer Sizes - You can vary the size of the 1/0 buffers RMS-11 uses to store
data in memory (see "Using RMS-11," Section 1.2.4.1). Generally, the larger
the buffers, the faster the task processes data. See ''1/0 Techniques," Section
3.3.1.3, 4.3.1.3, or 7 .1.3, for the file organization(s) you are interested in.

2. 2. 3 Provide Shared Access
Shared access revolves around the question: who is allowed to read or write to
a file? The answer involves two levels of permission to access the file:
• system protection codes
• sharing specifications in accessing programs
2.2.3.1 System Protection Codes - Operating systems allow you to assign a
protection code to each file when it is created. This code describes concentric
circles of users who are allowed different levels of access to that file (see
Figure 2-2). See your operating system documentation for specific protection
conventions.
1

A word equals two bytes.

Application Design

2-5

Figure 2-2: System Protection Concepts
READ ACCESS

WRITE ACCESS

RSTS/E

READ ACCESS

WRITE ACCESS

GROUP

EXTEND ACCESS

DELETE ACCESS

IAS/RSX-11 M

2-6

Application Design

F-MK-00065-00

Before you can share an RMS-11 file, or for that matter, run the task that
accesses the file, you must log into your computer system under an account
number compatible with the protection code(s) assigned to the file and task.
2.2.3.2 Sharing among Programs - Once the operating system allows a program access to a file, the program's own specifications take effect. Whenever a
program opens a file, it must declare:

• the record operations it intends to perform on the file (find, get, put,
update, delete, and/or truncate)
• the operations it will allow other programs to perform on the file. These
operations are categorized as either 2:
Read-type
Other programs may access the file for get and find operations only.
Write-type
Other programs may access the file for put, update, delete, and truncate
operations, as well as get and find operations.

Shared Access Criteria - The first program to open a file determines how
other (rwt-first) programs can access that file (see also Table 2-1):
• T.he access declaration of the not-first programs must agree with the first
program's allow declaration.
Example The first program allows read access. Any program declaring update intentions is
denied access to the file.
Example The first program allows write access. Any other program, regardless of access
declaration, is allowed to open the file-if it meets the other requirements.

• The allow declaration of the not-first programs must agree with the first
program's access declaration.
Example The first program has accessed the file for write operations. All not-first programs
must allow write access.

CAUTION
On RSTS/E, if the first program has specified access write
and allow no-write, other programs with access read, allow
no-write declarations can still open the file. However, the
reading programs are not protected against the file changes
being made by the writing program; the following subsection,
"Bucket Locking," discusses this topic.
• The allow declaration in the not-first programs must be the same as the
allow declaration in the first program.
2

Some higher level languages have a broader range of access options; they break down to
language-specific checks, then either read-type or write-type sharing when RMS-11 opens the
file.

Application Design

2-7

Example H the first program allows write operations, all not-first programs must allow
write operations.

NOTE
On IAS/RSX-llM, there is one exception to this criterion: if
the first program declares read access, but allows writers, a notfirst program with any access declaration and allowing no
write access can open the file. However, from that point,
programs attempting to open the file must have a no-write allow
declaration.
The operating system only allows not-first programs that meet all these
criteria to open the file.
CAUTION
The first program opens a file with access read, allow no-write
declarations. According to the assigned system protection code:
• If the program has write-access privileges to that file,

RSTS/E grants write access to the program. No other program can open the file with write access.
Other access read, allow no-write programs open the file.
Then, the first program closes the file. Write access to the file
is then available to other programs.
• If the program has only read-access privileges to. that file,

RSTS/E does not grant write access to the program; instead,
the program opens the file for read access only. Write access
to the file is still available.
When the write access is available, a program declaring access
write, allow no-write can open the file. Any reading programs
also accessing the file are not protected against the changes
caused by the writing program.
Table 2-1: Shared Access Criteria
First Program Declarations
Not-first
Program
Declarations

Access Write
Allow Write

Access Write
Allow No Write

Access Read
Allow Write

Access Read
Allow No Write

Access Write
Allow Write

Opens file

Access denied

Opens file

Access denied

Access Write
Allow No Write

Access denied

Access denied
opens file

Access denied

Access denied
opens file

Opens file

Access denied

Opens file

Access denied

Access denied
opens file

Access denied

Opens file

Access Read
Allow Write
Access Read
Al.low No Write

2-8

Application Design

~~· - ·-,.~,

NOTE
You cannot ensure that a program has exclusive access to a file.
Bucket Locking - The concern here is data integrity. Anyone who updates a
record should be assured that the data written back to the file is good until
that record is accessed again.

Conflict occurs when more than one program tries to update a file at the same
time. If no control is placed on access, two or more programs could read the
same record, one after the other, then update it, one after the other. Only the
last update remains in the file.
Therefore, RMS-11 activates bucket locking for a Relative or Indexed file
when the first program to open it allows write sharing. From that point,
RMS-11 requests the operating system to lock each bucket read from disk
until RMS-11 explicitly releases the bucket. Typically, after a get, find, or
mass insert put 3 operation, only the bucket containing the data record remains locked. While that bucket is locked, no other program can access it.

RMS-11 requests the operating system to unlock such a bucket when one of
the following occurs:
• 'The get, find, or put operation fails.
• The get or find operation ends successfully-if the program has declared
read only access to the file.
• The program initiates another record operation that accesses a different
data record bucket.
After the bucket is unlocked, other programs may access it.
Example Programs A and Bare write-sharing a file named RMSREL.DAT (see Figure 2-3).
Both try to update relative record number 12. However, program B initiates the
prerequisite get operation first, locking the bucket containing the record. The operating system keeps program A from accessing that bucket while program B uses it.
After program B updates record 12, RMS-11 unlocks the bucket and the operating
system allows program A to get record 12 (including program B's updated data).

A performance-oriented 1/0 technique used with Indexed files. See Chapter 7.

Application Design

2-9

Figure 2-3: Bucket Locking Example

RELATIVE
RECORD
#12-~

PROGRAM
B
READ RELATIVE
RECORD #12

A. PROGRAM B GETS RECORD 12

PROGRAM
A
READ RELATIVE
RECORD #12

PROGRAM B
PROCESS
RECORD

B. PROGRAM A IS DENIED RECORD 12

(continued on next page)

2-10

Application Design

:Figure 2-3: Bucket Locking Example (Cont.)

PROGRAM.B
PROGRAM A

UPDATE
RECORD

TRY AGAIN

C. PROGRAM B UPDATES RECORD 12, UNLOCKING BUCKET

PROGRAM A
PROGRAM B
DO NEXT
RECORD

D. PROGRAM A GETS RECORD 12
F-MK-00097-00

Application Design

2-11

Cost The operating system administers the bucket-locking process. It establishes, for each file, a list of virtual blocks thatare locked. The system
must scan this list every time RMS-11 performs a read-type operation
and then either permit the read or return an error.

In addition to this lock-list overhead, extra instructions are executed to
lock and unlock the buckets. The unlock sequence is particularly costly
because RMS-11 makes a special monitor call.
File Organization Restrictions and bucket locking.

The file organizations restrict file sharing

Sequential Files
Programs can share Sequential files for read access only. If a program
accesses such a file to perform write-type operations, no other programs
can open that file. Conversely, if a reading program opens a Sequential
file, the operating system prevents writing programs from accessing it.
Therefore, programs attempting to open Sequential files must allow no
write access.
The primary reason for this restricted access is that Sequential files are
not sufficiently formatted to permit simple and economical control of
sharing.
Relative and Indexed Files
Sharing of Relative and Indexed files works as described in this section.

2.2.3.3 Sharing among Record Access Streams - In addition to the bucket
locking used when programs allow sharing, RMS-11 activates its own version
of bucket locking when a program accesses a file for write-type operations.
This locking allows multiple Record Access Streams to share the file.

RMS-11 bucket locking works in the same manner as the locking administered by the operating system, except that the locks can be encountered only
by different Record Access Streams within the same program.
Cost The overhead for RMS-11 bucket locking is small.

2.2.3.4 Programming Considerations - For the greatest flexibility at run
time, you should always assume that any record your program attempts to
access can be denied because the bucket containing the record is locked.
RMS-11 returns the error code ER$RLK when the bucket is locked by another
Record Access Stream in the same or in another program.

Therefore, you should use the following techniques when you write RMS-11
programs that involve shared access:
• Never keep a bucket locked longer than necessary. You should follow any
successful get or find operation with another record operation of any type as
soon as possible. The second operation unlocks the bucket locked by the
read-type operation.

2-12

Application Design

Wilht ua :am: am : a ; : ; '"'"'

Alternatively, you can release the bucket explicitly with a free operation. A
free operation releases only the bucket locked by the Record Access Stream
associated with the operation.
MACR0-11
Issue a $FREE macro.
BASIC-PLUS-2
Use an UNLOCK or FREE statement.
PDP-11 COBOL
PDP-11 COBOL does not support the free operation.

RPG II
RPG II does not support the free operation.
DIBOL
DIBOL does not support the free operation.
• If your program detects an ER$RLK error (or its higher level language

equivalent described in Appendix B), its error processing depends on the
number of Record Access Streams active on the file:
Single Stream
Set up a loop that waits, then re-initiates the record operation until
RMS-11 indicates a successful completion.
Multiple Streams
Do not set up a loop that continuously re-initiates the record operation.
You should either:
•

continue processing on the other streams, attempting the record
operation on the locked-out stream periodically

•

release the buckets locked by all the other streams, then re-initiate
the record operation that failed. Any get-update or find-update
sequences interrupted on the other streams must be restarted, since
the release of a bucket destroys context (see "Record Access
Streams," Section 1.2.4.3).

2.2.4 Remember Ease of Design
When you design and write your application, you should consider yourself and
the person who will maintain the application. Keep in mind the following:
• Keep things simple. You can apply this criterion to the whole development
process: from program flowcharts to the record layouts to the file organizations and design.
Example From Sequential through Indexed, the RMS-11 file organizations offer more capability, but they are also more complex. Choose the organization that supplies
enough capabilities, but no more. For instance, if you want to random access a file
by a single key only, you might use a Relative file and some hashing instead of an
Indexed file.

Application Design

2-13

• Apply optimizations one by one until you reach a satisfactory level of performance. Generally, further improvements are not necessary.
Example The optimization of Indexed performance can be involved, but you do not have to
use every technique discussed in this manual. You should only satisfy current
performance requirements. For instance, recently a PDP-11 COBOL program
needed optimization. The Indexed file being read was made contiguous (discussed
in Chapter 6) and the RMS-11 overlay structure was changed (discussed in
Chapter 8): execution time dropped from 16 minutes to 8.5. Since this performance was adequate, no more optimizations were considered.
Example Some optimizations apply to one type of record operation, but not to others.
Determine if an optimization will benefit your processing before you implement
it.

2.3 Design Process
The first step in the design process is the selection of the file organization.
Table 2-2 shows the capabilities of the RMS-11 file organizations. Table 2-3
describes their advantages and disadvantages.
Once you've selected, go to the appropriate chapter(s):
Sequential
Relative
Indexed

Chapter 3
Chapter 4
Chapter 5

Each chapter discusses file structure (physical and conceptual) as well as
design considerations. Indexed files are the most complex to design because of
their power and flexibility. You must consider bucket sizes, areas, placement
control, index levels, and so on.
After you read the file organization chapter(s), go to Chapter 8, "Common
Optimization Techniques".
Finally, you employ the design considerations described in this manual. Write
the application programs. Create and populate the files, using the RMS-11
utilities when they are useful. Use the programs and files in a simulated
environment while you evaluate performance. You may have to return to this
manual, changing your design and/or combining attributes and RMS-11
facilities in different ways, until the application runs to your satisfaction.
Design is important to the success of your RMS-11 application.

2.4 Selecting a File Organization
Table 2-2 lists important features of each file organization to help you decide
which one(s) you need. Table 2-3 points out advantages and disadvantages.
But first some information about certain of those features to help you decide:
Record Formats - RMS-11 supports all of the following record formats for
Sequential files, but restricts Relative and Indexed file organizations (see
Table 2-2):

2-14

Application Design

Fixed
Records in the file are the same size, which is a file attribute. The fixed
record format requires no RMS-11 overhead.
RMS-11 limits fixed block-spanning records to 32,765 bytes, while the
minimum valid record is one byte of data on disk, 18 bytes on magnetic
tape.
Variable
Records in the file can be any length, up to a maximum stored as a file
attribute. For each record, RMS-11 maintains a record-length field specifying the number of data bytes in the record. The size of this field depends
on the storage medium for the file (see also Figure 2-4):
• On disk, the field is a two-byte binary count that does not include the
two bytes for the field.
• On ANSI magnetic tape, the field is a four-character decimal count that
does include the four characters for the field.
I~igure 2-4:

Record-Length Field on Disk and Tape

[ LE~~:H I-

l'ON DISK

DATA RECORD

- I, 1,~

LE;~TH

[

.......

_.tl\ _____
I

", .......

•

DATA RECORD

---------- --------

,,ON MAGTAPE

•

___ _._,,,,../

0-M K-00064-00

The variable record format:
• should be used when the data truly varies in length, because the format
adds the record-length field to each record's size
• can be used in a new application when future uses may require records
to change length

NOTE
Changing a record's length during an update operation is
restricted by file organization. See the ''Record Operations''
sections of the organization-specific chapters.
RMS-11 limits variable-length block-spanning records on disk to 32,763
bytes because of the record-length field. RMS-11 allows records to reach
this maximum only in Sequential files; other file organizations place
further restrictions on record size. The minimum valid record is a two
bytes of zeroes representing a null record.

Application Design

2-15

Variable-With-Fixed-Control (VFC)
A VFC record consists of two areas:
• a fixed control area from one to 255 bytes in length; the length is maintained as a file attribute.
• a variable area that can vary in length from zero bytes to the maximum
record size stored as a file attribute.
For each record, RMS-11 maintains a record-length field specifying the
number of data bytes in the record including fixed and variable areas.
The size of this field depends on the storage medium for the file (see
Figure 2-4):
• On disk, the field is a two-byte binary count that does not include the
length of the field.
• On ANSI magnetic tape, the field is a four-character decimal count that
does include the length of the field.
RMS-11 limits VFC block-spanning records to 32, 763 bytes because of the
record-length field. The minimum valid record is three bytes: the recordlength field plus the minimum fixed area of one byte. The maximum
variable area is the difference between 32, 763 and the length of the fixed
area.
Stream
A stream record consists of a series of contiguous ASCII characters.
RMS-11 detects the end of a stream record only by the presence of one of
the following terminators:
Carriage Return-Line Feed(015 8 /012 8 )
CTRL/Z (032 8 )
Escape (033 8 )
Form Feed (014 8 )
Line Feed (012 8 )
Vertical Tab (013 8 )
RMS-11 limits stream format to Sequential disk files. Additionally, the
format causes the most CPU overhead because RMS-11 must examine
each record character-by-character for the terminator.
During record operations, RMS-11 processes stream records as follows:
Find and Get Operations
RMS-11 scans the stream of ASCII characters, removing leading NUL
(000 8 ) characters and searching for the first occurrence of one of the
terminators:
• If it finds a Form Feed, Vertical Tab, Line Feed, or Escape, RMS-11

includes the terminator character with the record.

2-16

Application Design

• If it finds a CTRL/Z:

and it has encountered only NUL characters, RMS-11 returns
ER$EOF error code.

and it has encountered non-NUL characters, RMS-11 includes
the terminator character with the record. RMS-11 also notes endof-file has occurred; ER$EOF will be returned on the next find or
get operation.

• If it finds a Carriage Return, RMS-11 checks the character following
the Carriage Return:
-

If the next character is a Line Feed, RMS-11 discards both terminator characters (Carriage Return and Line Feed) and considers
the record complete.

If the next character is not a Line Feed, RMS-11 includes the
characters in the record and resumes its search for a terminator.

During a get operation, RMS-11 moves each character included in the
record into the user buffer as it scans the stream of ASCII characters.
RMS-11 does not move any data into the user buffer during a find
operation.
Put and Update Operations
RMS-11 checks the last character of the record in the user buffer:
~• If it finds a Line Feed, Vertical Tab, Form Feed, or Escape, RMS-11

moves the record as it is to the I/O buffer.
0

If it does not find one of these terminators, RMS-11 moves the
record to the I/0 buffer and adds a Carriage Return-Line Feed
character sequence to the end of the record.

Undefined
The undefined format means that RMS-11 reads only blocks, not records.
Your program must interpret the contents of each block.

1/0 Techniques - RMS-11 supports the following techniques so you can
adjust the performance of record operations:
Deferred Write
Normally, every write-type record operation to a Relative or Indexed file
results in a physical I/0 operation. However, you can have RMS-11 defer
this write function until the I/0 buffer is full or must be used for another
bucket. Deferred write is the normal mode of I/O for Sequential files.

Application Design

2-17

Table 2-2: File Organization Characteristics and Capabilities
Characteristics
and
Capabilities

Sequential

Relative

Indexed

Medium
Disk
Magnetic Tape
Unit record

Yes
Yes
Yes

Yes
No
No

Record Formats
Fixed-length
Variable-length
VFC
Stream
Undefined

Yes
Yes
Yes
Yes
Yes

Yes
Yes
Yes
No
No

Yes
Yes
No
No
No

Record Overhead

None

One byte per .record

Seven bytes per record

Access Modes
Sequential
Random
Access by RF A

Yes
No
Yes

Yes
Yes
Yes

Record Operations
Connect
Delete
Disconnect
Find
Flush
Free
Get
Rewind
Truncate
Update
Put

Yes
No
Yes
Yes
Yes
No
Yes
Yes
Yes (disk only)
Yes
Yes

Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes

One or more blocks

Bucket

Normal mode of operation
Yes
No

Selectable
Use bucket size
Yes

No
No

Yes
No

Yes
Yes

File Sharing*

Multiple readers only

Multiple readers
and writers

Other Features

Block-spanning records

Maximum Record
Number

1/0 Unit
1/0 Techniques
Deferred Write
Multi-Block Count
Multiple Record
Access Streams
Multiple Buffers
Mass Insert

* Important: see system-specific exceptions in "Provide Shared Access," Section 2.2.3.

2-18

Application Design

Multi-Block Count (MBC)
You can open a Sequential file so that RMS-11 reads or writes more than
one block of the file into the 1/0 buffer at a time. This capability speeds
file processing, though the buffer gets bigger. For Relative and Indexed
files, you achieve a similar effect by increasing bucket sizes.
Multiple Buffers (MBF)
You can allocate I/O buffers for a Relative or Indexed file beyond
RMS-ll's minimum requirements: one for Relative; two for Indexed. If
only one task is accessing the file, RMS-11 uses the buffers to save in
memory, or cache, buckets from the file, so that they do not have to be
read from disk again if needed.
For Indexed files, RMS-11 caches the Root buckets from indexes that
are used, saving one I/O operation on every random record operation.
However, for Relative files, RMS-11 makes no distinction between buckets, saving them until it has to use the buffer.
Mass Insert
Turned on before the insertion of a series of records already sorted in
ascending order by Primary Key, this mode enables RMS-11 to store the
records tightly and quickly in the file. Records can be mass inserted only
at the logical end of an Indexed file. Mass Insert significantly improves
performance for single-key Indexed files. However, with each additional
key defined for the file, the improvement is smaller.

Table 2-3:

File Organization Advantages and Disadvantages
Sequential

Advantages

Simplest organization

Relative

Indexed

Random access in all
languages

Most flexible random
access:
• by any one of multiple keys or RFA
• key access by generic
or approximate value
• you access records by
record contents

Optimal use of disk and
memory:
• minimum overhead on
disk
• block spanning

Allows random get and
put operations

Optimal if application
accesses all records on
each run, except if file
must be write-shared

Optimal if application
accesses all records on
each run and file must be
write-shared

Most versatile in record
formats:
• exchange data with
nonRMS-11 systems
• compatible with IAS
and RSX-llM FCS
files*
• compatible with ANSI
magnetic tape format

Random and sequential
access with low overhead

Allows deletions

Duplicate key values
possible
Fast sequential access
Automatic sort of records by Primary and
Alternate keys; available during sequential
access
Record location is
transparent to user.

(Continued on next page.)

Application Design

2-19

Table 2-3: File Organization Advantages and Disadvantages (Cont.)
Sequential
• compatible
with
RSTS/E stream files*

Relative
Can be write-shared

Most versatile in storage
media; file is portable
Disadvantages

To get a record, most
higher level languages
must access all records before it (no access by RFA).
PDP-11 COBOL program
cannot access a record already passed without closing and re-opening file
(rewind is not available).
YOU can add records only
at end of file.
You can delete records
only at end of file; use
truncate record operation.
Interactive process is awkward: operator must wait
as a program searches for a
record.
Sharing restricted to multiple readers

Restricted to disk
File contains a cell for
each cell number between 1 and last record
in file; data may not be
stored densely.

Indexed
Can be write-shared
Potential range of key
values not physically
. present as in Relative
file organization
Highest overhead on
disk and in memory
Restricted to disk
Least simple programming

Program must know relative record number or
RFA of record before it
can randomly access the
data; no generic access
as in Indexed file organization.
Interactive access can be
awkward if you do not
access records by relative
number.
You can insert records
only into unused record
cells, but you can update
existing records.
RMS-11 does not allow
duplicate relative record
numbers.

* RMS--11 can read these file structures and return a record to your program. However, differences in data
storage techniques among programming languages can keep the pfogram from properly interpreting the
contents of that record. See also Appendix A.

2-20

Application Design

Chapter 3
Se~quentlal File Applications

CONVENTION
The cover term file directory in this manual
has the same meaning as the following systemspecific terms:

System

Term

IAS
RSTS/E
RSX-llM

directory entry and file header(s)
User File Directory entries
directory entry and file header(s)

Physical Structure - Sequential files carry almost no RMS-11 overhead.
The operating system's file management software stores attributes in the file
directory. RMS-11 stores data records beginning with Virtual Block 1:
0

If records cross block boundaries, RMS-11 packs records into the file end-

to-end, allowing for control information and padding.
•• If you do not allow records to span blocks, RMS-11 packs records into each

block, allowing for control information and padding.

NOTE
You will waste space in your file if both of the following are
true:
• You do hot allow records to span blocks.
• Your records do not exactly fit into a block.

3-1

To be compatible with other file management systems (see Appendix A),
RMS-11 flags space not used at the end of each block as shown in Table 3-1.
When you allow records to span blocks, the only unused space starts after the
last record in the file.

"·

Table 3-1: End-of-File Indicators
Medium

Record Format

Disk

All but Stream

-1 in word following last valid byte

Disk

Stream ASCII

nulls (OOOs) to end of block

Magnetic tape

All

circumflex ( "' ) to end of block

Unit record

All

CTRL/Z (032s)

EOF Indicator

However, for disk Sequential files, RMS-11 uses the end-of-file attribute,
stored in the file directory, to tell where tbhe valid data in a fbile ends. Itddoes
not rely on the indicators shown in Ta le 3-1. This attri ute inclu es a
Virtual Block Number and a byte offset within this block. The virtual block
containing the logical end-of-file may not be the last block allocated to the
file.

Example The end-of-file indicators shown in Table 3-1 are like the words "THE END"
printed on the last page of a book. On the other hand, the end-of-file attribute that
RMS-11 uses is like listing the last page or "THE END" in the table of contents.

RMS-11 reads the end-of-file with the other file attributes when it opens a
file. RMS-11 also updates the end-of-file in the file directory when it closes
the file if the end-of-file changed while the file was open. The end-of-file
changes if records were added to the end of the file or if the file was truncated.
Conceptual Structure - RMS-11 stores records in the sequence that programs write them, one after the other from the first record in the file to the
last. RMS-11 can access the records in the same order or randomly via
Record's File Address (disk files only).

3.1

Record Definition
Records in disk Sequential files are word aligned, which means that RMS-11
adds a pad byte to the end of any record with an odd number of bytes.
RMS-11 uses this convention to maintain structural compatibility with
FCS-11 sequential files.
You can define a Sequential file so that RMS-11 writes records across the
boundaries between blocks. -Such a Sequential file is optimally dense; all
bytes within its allocated space are used, except at the end of the file where no
data has been written.
Table 3-2 shows the maximum data sizes for records in a Sequential file.
These are the sizes of your data; they are adjusted for RMS-11 restrictions
and overhead.

3-2

Sequential File Applications

Table 3-2: Sequential File Data Sizes (in bytes)
Maximum Data Size
Format

With Block-Spanning

Without Block-Spanning

Data Size Calculation

Fixed

32766

512

Your data + MOD(yd/2) 1

Variable

32765

510

Your data + 2 + MOD(yd/2) 1

VFC

32765 5

509

Fixed + variable + 2 + MOD(yd/2) 1

Stream

None

511 2

Data + terminator(s)

MOD(yd/2) is the remainder after the length of your data (yd) is divided by 2:
• MOD(yd/2) =: 0 if the data size is an even number of bytes.
• MOD(yd/2) =, 1 if the data size is an odd number of bytes.
For VFC, yd= fixed + variable.

Assuming one-byte terminator character; however, if terminator is CR-LF, then maximum length without
block-spanning records is 510 bytes. Note that these figures do not include the terminator characters.

3.2 File Design
With Sequential files, design includes:
Record format selection
Record Formats in Section 2.4 completely discusses your choices of record
formats.
Medium selection
Sequential files can be accessed on disk and magnetic tape. When you
select the medium for your file, consider the following:
Speed of access
How long can each record operation take? Tape is significantly slower
than disk.
Frequency of use
How often do you use the file? If you use it once a month, a quarter, and
so on, you could store the file on tape and save your disk for more
immediate purposes.
Transportability
RSTS/E disk structure is not compatible with IAS, RSX-llM, or VAX,
and vice-versa. If you need to use the file across these systems, you
should consider using a magtape file.
Allocation
Allocation involves two different quantities:
Initial Allocation Quantity
The number of blocks assigned to a file when you create it
Default Extension Quantity
The number of blocks added to a file when RMS-11 extends it
automatically

Sequential File Applications

3-3

The concept of contiguity involves both these quantities. Contiguity significantly impacts performance, but its use differs by operating system
(discussion in Chapter 8).

3.2.1

Initial Allocation

Even with Sequential files, where a file extension requires only an allocation
of blocks by the operating system, total allocation of the file when you create
it is much more efficient.
You calculate the allocation amount for block-spanning records as follows:
ALQ = (NRF*RSZ)/512
where:
ALQ

is the allocation quantity in blocks

NRF

is the largest number of records that will reside in the file at one time

RSZ

is the size of the record in bytes:.

• For a variable record format (VAR or VFC), use the average record
size, including two bytes for the record-length field.
• For fixed-length records, use the actual record size.
Be sure to round RSZ up to a multiple of two to account for word alignment.
This allocation can be done by the RMSDEF utility or by your application
program as follows:
MACR0-11
Use the initialization macro or $STORE to set the ALQ field in the FAB
(or XAB if you are using placement control) of the file to the calculated
number of blocks before issuing $CREATE.
BASIC-PLUS-2
Use the FILESIZE clause in the OPEN statement that creates the file.
PDP-11 COBOL
Use the /AL:n switch on the file specification in the ASSIGN clause or the
VALUE OF ID.
RPG II
Use the RPGASN utility to override the default value set by the compiler
with a switch.
DIBOL
Use RMSDEF; DIBOL does not support allocation quantities during the
creation of Sequential files.

3-4

Sequential File Applications

""'

3.2.2 Default-Extension Quantity
You should establish a reasonable Default Extension Quantity (DEQ)
whether the file is totally allocated at creation time or not. A reasonable
DEQ minimizes the number of file extensions. The time required for each file
extension is significant; involved are:
• A call to the system file processor
• Possible I/0 operations to bring file processor routines into memory
• I/O operations to read and change file directory information
• I/O operations to read and change the disk free-block bit map
A good basis for calculation is the number of records added to the file in a
given period of time, such as a day; use the formula for allocation quantity to
determine the number of blocks.

"-"''

If you do not specify a DEQ, it defaults to zero whether you create the file

with RMSDEF or a higher level language. RMS-11 responds to a DEQ of zero
by requesting five blocks from the file processor each time it automatically
extends the file.
Example You are inserting 1000 fifty-byte fixed-length records into a Sequential file. Records
do not span blocks; therefore, each block contains ten records. The file is currently
full, that is, no more records may be added without an extension.
• If DEQ is zero, RMS-11 extends the file by five blocks each time it runs out of
space. Therefore, in this example, RMS-11 extends the file twenty times.
• If DEQ is 1, RMS-11 extends the file for every tenth put operation after the first,
for a total of 100 extensions.
• If DEQ is 25, RMS-11 extends the file four times.
• If DEQ is 100, RMS-11 extends the file only once.

The DEQ for the file can be set by the RMSDEF utility or by your application
program as follows:
MACR0-11
Use the initialization macro or $STORE to set the DEQ field in the FAB
(or XAB if you are using placement control) of the file to the calculated
number of blocks before issuing $CREATE. You can also set a run-time
DEQ.
BASIC-PLUS-2
Use RMSDEF; BASIC-PLUS-2 does not support DEQ specifications.
PDP-11 COBOL
Use the /EX:n switch on the file specification in the ASSIGN clause or the
VALUE OF ID.
RPG II
Use the RPGASN utility to override the default value set by the compiler
with a switch.

Sequential File Applications

3-5

DIBOL
Use RMSDEF; DIBOL does not support DEQ specifications during the
creation of Sequential files.

.zt!!t~

3.2.3 Contiguity
Finally, you should consider contiguity for a Sequential file to minimize the
time spent in each VO operation. If the blocks in a file are not contiguous,
they can be on different parts of the disk. The device must therefore move its
heads to access the file contents. However, physical contiguity ensures that
the file is stored on one track, or at worst, adjacent tracks. Since the disk can
read a track without moving the heads, file contiguity reduces head movement. This statement assumes that no other software is accessing the disk at
the same time.
Contiguity also enhances virtual-to-logical-block mapping (discussed in Section 8.3).

To ensure that the blocks in the file are physically contiguous, allocate the
whole file when you create it (see "Initial Allocation," Section 3.2.1).
MACR0-11
Use the initialization macro or $SET to set the FOP field in the FAB (or
XAB if you are using placement control) of the file to include FB$CTG
before issuing $CREATE.

BASIC-PLUS-2
Use the CONTIGUOUS clause in the OPEN statement that creates the
file.
PDP-11 COBOL
Use the /CO switch on the file specification in the ASSIGN clause or the
VALUE OF ID.

;!llllh

RPG II
Use RMSDEF; RPG II does not create contiguous files.
DIBOL
Use RMSDEF; DIBOL does not create contiguous files.

3.3 Task Design
The record and file processing capabilities described in Chapter 1 are available for Sequential files. This section discusses the operations and their implementation and restrictions with Sequential files.

3-6

Sequential File Applications

3.3.1

Record Operations

RMS-11 performs a record operation at the request of a program. The available operations include:
Connect
Disconnect
Find
Flush
Get
Put
Rewind
Truncate
Update

:'-'

In all record operations, except truncate, RMS-11 establishes Current Record
(if any) and Next Record (if applicable). If any record operation fails,
RMS-11 normally sets Current Record to NONE and does not change Next
Record. "Record Access Stream," Section 1.2.4.3, introduces the concepts of
Current· Record and Next Record.

3.3.1.1 Connect - A connect operation affects the Current Context for the
Record Access Stream as follows:

Current Record
There is no Current Record. Any operation requiring Current Record fails
at this point.
Next Record
• If you did not specify that you were going to append records to the file,
the Next Record is the first record in the file.
• If you did specify that you were going to append records to the file, the
Next Record is the end-of-file.

Your program specifies that it will append records to the file as follows:
MACR0-11
Use the initialization macro or $SET to set the value RB$EOF in the
ROP field of the RAB before issuing $CONNECT.
BASIC-PLUS-2
Use ACCESS APPEND in the OPEN statement that creates the file.
PDP-11 COBOL
Use the keyword EXTEND in the OPEN statement.
RPG II
RPG II does not support this feature.
DIBOL
DIBOL does not support this feature.

Sequential File Applications

3-7

3.3.1.2 Disconnect -A disconnect operation destroys the Current Context for
the Record Access Stream. You cannot resume this context by reconnecting
the stream.

3.3.1.3 Find - To perform a find operation on a Sequential file, RMS-11:

1. determines the location of the record in the file according to the specified

access mode:
• Location is indicated by Next Record pointer in Sequential Access
Mode.
• Location is determined by specified RFA in Access by RFA.
2. reads the block containing the record, or the first part if it spans blocks,

from disk into the task's I/0 buffer, if it is not already in memory. The
block may be in memory if it was required by a previous operation.
3. returns the RFA to the program, but does not transfer the record to the
program's user buffer
If no valid record exists in the location specified, the response depends on the

access mode:
• In Sequential Access Mode, the error code is ER$EOF, meaning that no
record was located because there are no more records in the file.
• In Access by RF A, the error code is ER$RF A, meaning that no record was
located at the RFA specified.

A find operation affects the Current Context for the Record Access Stream as
follows:
• find in Sequential Access Mode:
Current Record
Set to value of the record found, that is, the Next Record before operation started.
Example You've connected a stream to a Sequential file without specifying append.
There is no Current Record, but the Next Record is the first record in the file.
If you execute a sequential find operation, the Current Record is set to the
first record in the file.

Next Record
Set to the record virtually following the Current Record.
Example From the previous example, Next Record is the second record in the file.

• find in Access by RF A:
Current Record
Set to the record found, that is, the reocrd identified by the RF A.

3-8

Sequential File Applications

"-".·

Next Record
Unchanged.
Example In the preceding examples, you've done a sequential find after connecting the
stream to the file. You now execute a find by RFA. The Current Record is set
to the record specified, but the Next Record is not changed. Therefore, when
you do another sequential find, Current Record is set to the second record in
the file, not the record following the one found by RFA.

You use find instead of a get operation because:
• it is quicker because the record is not moved to the user buffer. Although the
time required to move a record from one part of memory to another is very
short, do not expend it unnecessarily.
• it does not change Next Record in access by RFA. This convention allows
you to branch off sequential processing for updating, deleting, or truncating,
and yet keep your place.
You can use a find operation in the following ways:
• To skip records in Sequential Access Mode by initiating successive find
operations.
• To establish a random starting point for sequential processing with RFA.
You could then initiate successive get operations, where the first one gets
the record found by RFA.
• To establish a Current Record for an update or truncate operation.
3.3.1.4 Flush - See "Records Operations," Section 1.2.4, for a summary of
the flush operation.

A flush operation does not affect the Current Context for the Record Access
Stream.
3.3.1.5 Get -To perform a get operation on a Sequential file, RMS-11:

1. determines the location of the record in the file according to the specified

access mode:
• In Sequential Access Mode, location is indicated by:
• Next Record pointer, if the get operation was not immediately preceded by a successful find operation.
• Current Record pointer set by an immediately preceding successful
find operation.
• Location is determined by speCified RF A in Access by RF A.
2. reads the block containing the record, or the first part if it spans blocks,

from disk into the task's I/0 buffer, if it is not already in memory.
Example Your records are 50 bytes long. When you read sequentially through the file,
RMS-11 must request a disk 1/0 operation for every tenth get record operation
your program executes.

Sequential File Applications

3-9

3. returns the RF A to the program and moves the record from the VO buffer
to the specified user buffer in the program unless the program is operating
in Locate Record Transfer Mode. If the buffer does not contain the entire
record, RMS-11 reads more blocks into the I/O buffer and assembles the
record in the program's user buffer regardless of record transfer mode.

If no valid record exists in the location specified, the response depends on the
access mode:

• In Sequential Access Mode, the error code is ER$EOF, meaning no record
was located because there are no more records in the file.
• In Access by RF A, the error code is ER$RF A, meaning no record was located
at the RFA specified.
A get operation affects the Current Context for the Record Access Stream as
follows:
• get in Sequential Access Mode not immediately preceded by a successful
find operation:
Current Record
Set to value of the record read, that is, the Next Record before operation
started.
Example You've connected a stream to a Sequential file without specifying append.
There is no Current Record, but the Next Record is the first record in the file.
If you execute a sequential get operation, the Current Record is set to the first
record in the file.

Next Record
Set to record virtually following Current Record.
Example You've connected a stream to a Sequential file without specifying append.
There is no Current Record, but the Next Record is the first record in the file.
If you execute a sequential get operation, the Next Record is set to the second
record in the file.

• get in Sequential Access Mode immediately preceded by a successful find
operation:
Current Record
Unchanged (from Current Record set by find operation).
Next Record
Set to record virtually following Current Record (possibly changing Next
Record set by find operation).
• get in Access by RFA:
Current Record
Set to record specified by RFA, that is, the record read.
Next Record
Set to record virtually following Current Record. This convention differs
from find by RF A which does not change Next Record.

3-10

Sequential File Applications

-J

3.3.1.6 Put -

To perform a put operation on a Sequential file, RMS-11:

1. determines if Sequential Access Mode is specified; if it is not, RMS-11
returns the error code ER$IOP.
2. determines if Next Record is end-of-file; if it is not, RMS-11 returns the
error code ER$NEF.
Your program gets to the end of a Sequential file by:
• specifying append when the program connects the Record Access
Stream to the file (see "Connect," Section 3.3.1.1).
• initiating sequential find and/or get operations until RMS-11 returns an
ER$EOF error code.
3. reads the last block in the file into the 1/0 buffer, if it is not already in
memory
4. moves the record from the user buffer in the program to the task's 1/0
buffer
5. writes the VO buffer to disk only if the buffer is full. If there is not room
for the block(s) in the file, RMS-11 extends the file (see "Default Extension Quantity," Section 3.2.2) and then writes the buffer.
A put operation affects the Current Context for the Record Access Stream as
follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
End-of-file. A sequential find or get operation fails with error code
ER$EOF.
3.3.1.7 Rewind - A Rewind operation sets the context of the Record Access
Stream to the beginning of the Sequential file. In doing so, it affects the
Current Context for the stream as follows:

Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Set to first record in file.

3.3.1.8 Truncate - A truncate operation declares an end-of-file at the position of the Current Record. In doing so, the operation deletes the Current
Record and all records in the Sequential file following that record.

Truncate requires a valid Current Record. It therefore should follow a successful get or find operation; otherwise, RMS-11 returns the error code ER$CUR.

Sequential File Applications

3-11

A truncate operation affects the Current Context for the Record Access
Stream as follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
End-of-file.
After a truncate operation, you can immediately add to the file using put
operations.

NOTE
The truncate operation does not reduce the size of a Sequential
file.
3.3.1.9 Update -In an update operation, RMS-11 moves the specified record
from the task's user buffer to the 1/0 buffer, replacing the Current Record set
by the prerequisite get or find operation. However, RMS-11 does not immediately write the buffer to the file. RMS-11 requests the file processor to write
the changed buffer over its original location on the disk only when the buffer
must be replaced in memory by another operation.
Example You get a record by RFA and then update it. Then, you get another record by RFA.
RMS-11 writes the buffer containing the first record you updated only when it must
replace the data in the buffer to satisfy the second get operation.

Update operations have the following restrictions:
• The operation is valid only on disk Sequential files. If you attempt it on
magnetic tape files or unit record devices, RMS-11 returns the error code
ER$IOP.
• The operation requires a valid Current Record. It therefore should follow a
successful get or find operation; otherwise, RMS-11 returns the error
ER$CUR.
• The size of the record cannot change during an update operation. If it
changes, RMS-11 returns the error code ER$RSZ.
• You cannot update Stream format records. If you attempt it, RMS-11 returns the error code ER$IOP.
None of these errors affects the original record in the file on disk.
An update operation affects the Current Context for the Record Access
Stream as follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Unchanged.

3-12

Sequential File Applications

3.3.2 Record Transfer Modes
You can manipulate records either in the 1/0 buffer or in your program's user
buffer (see Figure 3-1). Each of these options is called a Record Transfer
Mode. You can change Record Transfer Mode at run time, even between
record operations.

Figure 3-1: RMS-11 Task Structure

fS1zE DEPENDS ON:- -

-1

I• NUMBER OF FILES OPENED SIMULTANEOUSLY :

I • BUCKET SIZES

~ NUMBER_.9~RECOrRD~CCESS STREAMS _ _

Us ERB UFFERS

'
1/0
.BUFFERS

VIRTUAL
MEMORY

PROGRAM

RMS-11

INTERNAL
CONTROL
STRUCTURES

fSizEDEPENDS ON:- ]O_

l • RMS-11 FUNCTIONS USED l
~~VERLA ~TRUCTURE~~E~

H-M K-00069-00

3.3.2.1 Move Mode - Move Mode requires that each record be copied between the user and 1/0 buffers:

Sequential File Applications

3-13

3.3.2.2 Locate Mode - Locate Mode enables your program to manipulate
records in the 1/0 buffer, eliminating the data transfers between it and the
user buffer. However, when you specify Locate Mode, RMS-11 uses the 1/0
buffer only when such usage does not compromise data integrity. Otherwise,
RMS-11 uses Move Mode. Therefore, your program must still contain a user
buffer.
Example RMS-11 uses Move Mode instead of Locate Mode when records span buffers in a
Sequential file.
Example RMS-11 uses Move Mode instead of Locate Mode if you opened the file indicating
that you were going to perform update operations on it.

RMS-ll's use of Move Mode instead of Locate Mode is transparent to your
program as long as you use RMS-11 facilities to access the record data.
For Sequential files, your program can both get and put records in Locate
Mode. See your language documentation to determine if the language supports Locate Mode and if it does, what the programming techniques are.

3.3.3 1/0 Techniques
You can use the following techniques to improve the performance of record
operations.
3.3.3.1 IAS/RSX-11 M Asynchronous Record Operations - Within each Record
Access Stream, your program can perform any record operation either synchronously or asynchronously. In synchronous operations, RMS-11 returns
control to your program after the operation ends, either successfully or with an
error.

When you execute an asynchronous operation, RMS-11 may return control to
your program before the operation is complete. The program continues processing while the physical transfer of data between disk and memory is carried
out. However, you must not initiate another record operation on that stream
until the first operation ends; otherwise, RMS-11 returns the error code
ER$ACT. See your language documentation for asynchronous techniques.

NOTE
If you intend to use asynchronous RMS-11 record operations
and/or Asynchronous System Traps (ASTs) in other parts of
your program, see the section on your operating system in
Appendix A.

3.3.3.2 Deferred Write -The normal mode of operation for Sequential files is
similar to operations using Deferred Write with the other file organizations.
Using this technique does not change or improve performance.
3.3.3.3 Multiple Buffers -

Sequential files.

3-14

Sequential File Applications

The multiple buffer capability is not available to

3.3.3.4 Multiple Record Access Streams - RMS-11 allows each program to
use only one stream on a Sequential file because Sequential files are not
sufficiently formatted to permit simple and economical sharing.
3.3.3.5 Multl-Block Count (MBC) - Your task can be set up so that more than
one block from a disk Sequential file is read or written at one time. This
multiple-block 1/0 can improve processing as it tends to reduce the number of
physical VO operations. However, it also increases the size of the task, on a
one-for-one basis; that is, for each increment of MBC, the 1/0 buffer in the
task grows by 512 bytes.

An MBC greater than one is therefore useful for sequential processing, including file population.
Example You are using fifty-byte records. During sequential processing, if MBC is one,
RMS-11 requests a disk 1/0 operation for every tenth record operation your program
executes, whether the operations are gets or puts. If you set MBC to five, for
instance, RMS-11 causes a physical 1/0 operation for every fifty record operations.

Since MBC is a run-time parameter, the quantity is set by the application
program:
MACR0-11
Use the initialization macro or $STORE to set the MBC field in the RAB
of the Record Access Stream being used to the desired number of blocks
before the stream is set up ($CONNECT).
BASIC-PLUS-2
When you specify the organization as SEQUENTIAL in the OPEN statement for the file, the compiler equates the BUCKETSIZE to the MultiBlock Count for the Sequential file. If you try to allocate more buffer space
than is available to your task at runtime, the task terminates with an
error (?Maximum memory exceeded).
PDP-11 COBOL
When you specify the organization as SEQUENTIAL in the OPEN statement for the file, the compiler equates n in the RESERVE n AREAS
clause to the Multi-Block Count for the Sequential file.
RPG II
RPG II does not support this feature.
DIBOL
When you specify a value after the PROC statement, the compiler uses
that value as the Multi-Block Count for all Sequential files opened by the
program.,

3.:t4 FUe Operations
You can perform the following file operations on Sequential files. File operations do not involve records and can only perform synchronously.

Sequential File Applications

3-15

3.3.4.1 Close - A close operation disconnects the Record Access Stream
before RMS-11 releases access to the file. You can also specify magnetic tape
volume operations during a close operation. See Appendix F.
3.3.4.2 Create - In addition to the file specification, RMS-11 passes the
following information to the file processor when it creates a file:

• An initial allocation of blocks for the file.
• The location on a specific device where the processor should allocate those
blocks.
• The following file attributes:
File organization
Record format
Forms control
Record size
Number of virtual blocks in the file
End-of-file
Default extension quantity
3.3.4.3 Open -

You can specify the file you want to open in two different

ways:
By filespec
The first time you open a file, you must use the file specification.
By File ID
When you create or open a file by filespec, RMS-11 returns an identifying
notation to your program. You can store this File ID, either in memory or
in a file, and use it to open the file from that point on.
On IAS/RSX-1 lM systems, open by File ID is significantly faster than open
by filespec, because the process bypasses directory reads and other overhead.
However, on RSTS/E, open by File ID is no faster than open by filespec.
You can also specify magnetic tape volume operations during an open operation. See Appendix D. See also "File Operations," Section 1.2.5.3, for an
introduction to the open file operation.
3.3.4.4 Erase - You cannot erase a magnetic tape or unit record file;
RMS-11 returns the error code ER$IOP. "File Operations," Section 1.2.5.3,
introduces the concept of erasing files.

You cannot extend a magnetic tape file. "File Operations,"
Section 1.2.5.3, introduces the concept of extending files.

3.3.4.5 Extend -

3-16

Sequential File Applications

Chapter 4
Relative File Applications

CONVENTION
The cover term file directory in this manual
has the same meaning as the following systemspecific terms:
System

Term

IAS

directory entry and file header(s)

RSTS/E

file directory

RSX-llM

directory entry and file header(s)

Physical Structure - Relative files contain at least one block of RMS-11
information known as the Prologue. The operating system's file management
software stores attributes in the file directory. RMS-11 stores the Prologue in
Virtual Block 1 - unless bucket size is two, four, or eight blocks. In that case,
RMS-11 makes the Prologue equal to one bucket in size; this step can improve performance by aligning buckets with file clusters. Data records begin
in the block following the Prologue.

RMS-11 allocates Relative files in bucket increments. The first bucket begins
with the first data block. To support deleted record control, RMS-11 initializes each bucket (sets all bits to 0) when it allocates the blocks.
The fixed-length cells are set up in each bucket starting with byte 0 and
packed end-to-end, byte-aligned, until no more cells can fit in the bucket (no
padding necessary). Cells cannot span bucket boundaries, though they can
cross block boundaries in multi-block buckets. The first byte of each cell is
used by RMS-11 to provide deleted record control.

4-1

Conceptual Structure - RMS-11 stores records in a series of fixed-size cells.
Only one record can be put into a cell, but all cells do not have to contain
records. The cell size is based on the length you specify as the maximum for
any record in the file. RMS-11 numbers the cells consecutively from 1 ton,
where n indicates the last cell in the file. A cell number relates its location to
the beginning of the file and is associated with the record in the cell, if any, as
a relative record number.

RMS-11 can access records in a Relative file either sequentially or randomly,
via both relative record number and RFA.

4.1

Record Definition
RMS-11 calculates the size of a Relative record cell as follows:
1

rfo
+ ds

byte for RMS-11 overhead
bytes for record format overhead (0 for fixed; 2 for variable)
bytes in the data itself
bytes in each record cell in the file

The data size used for variable records is the Maximum Record Size set for
the file.
Table 4-1 shows the maximum data sizes for records in a Relative file. These
are the sizes of your data; they are already adjusted for RMS-11 restrictions
and overhead.
Table 4-1:

Relative File Data Sizes (in bytes)

Format

Maximum

Record Cell Size Calculation

Fixed

16,383 or 7,679

Data size + 1

Variable

16,381 or 7,677

Maximum record size + 3

VFC

16,381 or 7,677

Fixed + variable + 3

4.2 File Design
Though not as critical as for Indexed files, design is still important for an
application using a Relative file. Design considerations include:
1. Bucket size
2.

File allocation

3. Maxim um Record Number

4.2.1

Bucket Size

Buckets are the 1/0 units for Relative files. Their size is therefore critical to
the space required by a task and the speed with which it performs. Sequential

4-2

Relative File Applications

access, especially, benefits when there are multiple records per bucket. There
is, of course, a trade-off: the larger a bucket, the larger the task, but the faster
it reads data sequentially:
• Each block added to the bucket size increases the task size by 512 bytes.
• The speed of an RMS-11 operation is closely proportional to the number of
I/0 operations involved. RMS-11 requests an 1/0 operation each time it
requires a new bucket to locate a record. Therefore, the more record cells in
a bucket, the fewer 1/0 operations RMS-11 needs to read a file sequentially.
However, write sharing a Relative file counteracts this optimization if your
program has read-only access to the file. RMS-11 reads a bucket from disk
during each get operation - even if the Next Record is in the bucket in
memory - because the bucket isn't locked after each get operation and a
writing program may have changed the bucket since it was last read.
Bucket size can be set by RMSDEF or by your application program:
MACR0-11
Use the initialization macro or $STORE to set the BKS field in the FAB
(or XAB if you are using placement control) of the file to the chosen
number of blocks before issuing $CREATE. Note that the default value is
one block per bucket.

BASIC-PLUS-2
Use the BUCKETSIZE clause in the OPEN statement that creates the
file. See "Program Syntax," section 6.5.3, for cautions in the use of the
BUCKETSIZE clause.
PDP-11 COBOL
In the file-description-entry (FD), use the BLOCK CONTAINS clause.

RPG II
Use the RPG ASN utility to override the default value set by the compiler.
DIBOL
Use RMSDEF; DIBOL does not create Relative files.

4.2.2 Allocation
File allocation involves two different quantities:
Initial Allocation Quantity
The number of blocks assigned to a file when you create it
Default Extension Quantity
The number of blocks added to a file each time RMS-11 automatically
extends it
The concept of contiguity involves both these quantities. Contiguity has a
significant impact on performance, but its use differs by operating system.

Relative File Applications

4-3

4.2.2.1 lnltlal Allocatlon -Total allocation of a file when you create it is the
most efficient technique regardless of file organization, but with Relative files,
pre-allocation becomes most critical. Each allocation, whether at creation or
during an extension, requires RMS-11 to initialize the new buckets by setting
all bits to zero. You can avoid time-consuming file extensions during normal
processing by totally allocating the file when you create it or by explicitly
extending the file when it is not being used for processing.

You calculate the allocation amount as follows:
ALQ = PLG + (NRF/NRBKT)*BKS
where:
PLG

is equal to one block or to BKS if BKS is 2, 4, or 8.

NRF

is equal to MRN or to the number of records that will be written
into the file.

BKS

is the bucket size in blocks

NRBKT

is the number of records in a bucket:
NRBKT = (512*BKS)/(RSZ + RFO)
where:
RSZ is the size of the record:
• data size for fixed-length records
• maximum record length for variable-length records
• size of fixed control area + maximum variable area size
for VFC

RFO is the record format overhead:
• RFO = 1 byte for fixed-length records
• RFO = 3 bytes for variable-length and VFC records
RMS-11 rounds ALQ up to the next bucket size if you don't.
This allocation can be done by the RMSDEF utility or by your application
program as follows:
• during file creation
MACR0-11
Use the initialization macro or $STORE to set the ALQ field in the FAB
(or XAB if you are using placement control) of the file to the calculated
number of blocks before issuing $CREATE.
BASIC-PLUS-2
Use the FILESIZE clause in the OPEN statement that creates the file.
PDP-11 COBOL
Use the I AL:n switch on the file specification in the ASSIGN clause or
the VALUE OF ID.

4-4

Relative File Applications

RPG II
Use the RPGASN utility to override the default value set by the compiler.

~·

DIBOL
Use RMSDEF; DIBOL does not create Relative files.
0

by putting a record with the Maximum Record Number (MRN) in the file
first. Before RMS-11 can write this record, it must allocate all record cells
from 1 to MRN and initialize the new blocks. When the put operation is
finished, the Relative file is completely allocated.

However, if the file cannot be totally
allocated at creation, then you should establish a reasonable Default Extension Quantity (DEQ) to minimize the number of (and the time spent on) file
extensions. Even if the file is totally allocated when you create it, you should
establish a reasonable DEQ in case the file gets bigger than planned.
4.2.2.2 Default Extension Quantity -

A good basis for calculation is the number of records that are added to the end
of the file in a given time period, such as a day; use the formula for allocation
quantity in "Initial Allocation," Section 4.2.2.1.

The default extension quantity should be equal to a multiple of the bucket
size.
If you do not specify a DEQ, it defaults to zero whether you create the file

with RMSDEF or a higher level language. RMS-11 responds to a DEQ of zero
by requesting four times bucket size in blocks from the file processor each
time it automatically extends the file.
The DEQ for the file can be set by the RMSDEF utility or by your application
program as follows:
MACR0-11
Use the initialization macro or $STORE to set the DEQ field in the F AB
(or XAB if you are using placement control) of the file to the calculated
number of blocks before issuing $CREATE. You can also set a run-time
DEQ.
BASIC--PLUS-2
Use RMSDEF; BASIC-PLUS-2 does not support DEQs.
PDP-11 COBOL
Use the /EX:n switch on the file specification in the ASSIGN clause or the
VALUE OF ID.
RPG II
Use the RPGASN utility to override the default value set by the compiler.
DIBOL
Use RMSDEF; DIBOL does not create Relative files.
4.2.2.3 Contiguity- Finally, you should consider contiguity for a Relative file
to minimize the time spent in each I/0 operation. If the blocks in a file are not

Relative File Applications

4-5

contiguous, they are, by definition, in different parts of the disk. The device
must therefore move its heads to access the file contents. However, physical
contiguity ensures that the file is stored on a single track, or at most, adjacent
tracks. Since the disk can read an entire track without moving the heads, file
contiguity reduces head movement. This statement assumes that no other
software is accessing the disk at the same time.
Contiguity also enhances virtual-to-logical-block mapping (discussed in Section 8.3).
Therefore, if possible, you should allocate a Relative file contiguously, and the
only way to ensure that all blocks in the file are physically contiguous is to
allocate the whole file when you create it.
MACR0-11
Use the initialization macro or $SET to set the FOP field in the F AB of
the file to include FB$CTG before issuing $CREATE.
BASIC-PLUS-2
Specify CONTIGUOUS in the OPEN statement that creates the file.
PDP-11 COBOL
Use the /CO switch on the file specification in the ASSIGN clause or the
VALUE of ID.
RPG II
Use RMSDEF; RPG II does not create contiguous files.
DIBOL
Use RMSDEF; DIBOL does not create Relative files.

4.2.3 Maximum Record Number
The Maximum Record Number (MRN) associated with a Relative file limits
the size of the file. RMS-11 will not put a record into a file with a relative
record number greater than the assigned MRN. However, if an MRN is not
set, that is, MRN is zero, RMS-11 only checks if the record number is greater
than zero before attempting to store a record in a Relative file.
MRN determines the maximum useful size of a file because RMS-11 allocates
a record cell for each record between relative record number 1 and the highest
relative record number used. You can explicitly make the file larger than this
maximum, but RMS-11 will not use the space. The actual size can be smaller
than the size that would be set if a record with the Maximum Record Number
were written into the file.
You can calculate the file size (FSZ) in buckets from the largest relative
record number (LRN) in the file (greatest value equals MRN):
LRN
FSZ = (BKS*512)/(RSZ+RFO)
where:
BKS

4-6

is the bucket size in blocks

Relative File Applications

·~

RSZ

is the size of the record:
• size of your data for fixed-length records
• maximum record length for variable-length records
• size of fixed control area + maximum variable area length for VFC
is the format overhead:
• RFO = 1 byte for fixed-length records

RFO

• RFO = 3 bytes for variable-length and VFC records
MRN can be set by RMSDEF or by your application program:
MACR0-11
Use the initialization macro or $STORE to set the MRN field in the FAB
of the file to the desired number of records before issuing $CREATE.
Note: if you want no limit checks, do not include F$MRN in the FAB's
initialization block or set the field to zero at run time prior to initiating
the $CREATE macro.
BASIC-PLUS-2
Use RMSDEF; BASIC-PLUS-2 does not support MRN specifications.
PDP-11 COBOL
Use RMSDEF; PDP-11 COBOL does not support MRN specifications.
RPG II
Use RMSDEF; RPG II does not support MRN specifications.
DIBOL
Use RMSDEF; DIBOL does not create Relative files.

4.3 Task Design
The record and file processing capabilities described in Chapter 1 are available fo:r Relative files. This section discusses the operations and their implementation and restrictions with Relative files.

4.3.1

Record Operations

RMS-11 performs a record operation at the request of a program. The available operations include:
Connect
Delete
Disconnect
Find
Flush
Get
Put
Rewind
Update

Relative File Applications

4-7

In all record operations, RMS-11 establishes Current Record (if any) and
Next Record (if applicable). If any record operation fails, RMS-11 normally
sets Current Record to NONE and does not change Next Record. See "Record
Access Stream," Section 1.3.1.5, for an introduction to the concepts of Current Record and Next Record.

4.3.1.1 Connect - A connect operation affects the Current Context for the
Record Access Stream as follows:

Current Record
There is no Current Record. Any operation requiring Current Record fails
at this point.
Next Record
The Next Record is the first record cell in the file.
4.3.1.2 Delete - In a delete operation, RMS-11 flags the Current Record cell
to indicate that it contains a deleted record. RMS-11 does this by setting the
RMS-11 control byte in the cell to a certain value. The prerequisite get or find
operation brought the cell's bucket into the 1/0 buffer.

Then, RMS-11 writes the bucket over its original location on the disk, unless
you have specified Deferred Write (discussed in "1/0 Techniques," Section
4.3.3).
A delete operation requires a valid Current Record. Therefore, a delete should
follow a successful get or find operation; otherwise, RMS-11 returns the error
code ER$CUR. This error does not affect the original record in the file on disk.

A delete operation affects the Current Context for the Record Access Stream
as follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record.
Unchanged.

4.3.1.3 Disconnect -A disconnect operation destroys the Current Context for
the Record Access Stream. You cannot resume this context by reconnecting
the stream.
4.3.1.4 Find -

To perform a find operation on a Relative file, RMS-11:

1. determines the location of the record in the file according to the specified
access mode:
• Location is indicated by the Next Record pointer in Sequential Access
Mode.
• Location is determined by the specified relative record number and
match criterion in Random Access Mode.
• Location is determined by the specified RF A in Access by RF A.

4-8

Relative File Applications

2. reads the bucket containing the indicated cell from disk into the task's I/0
buffer, if it is not already in memory. The bucket may be in memory if it
was required by a previous operation.
3. checks the contents of the cell:
• If the cell contains a valid record, RMS-11 returns the RF A to the

program, but does not transfer the record to the program's user buffer.
• If the cell is empty or contains a deleted record, the response depends on

the access mode:
- In Sequential Access Mode, RMS-11 repeats steps 1 through 3.
- In Random Access Mode, RMS-11 reacts according to the specified
match criterion:
On equal match, RMS-11 returns the error code ER$RNF
On greater-than or greater-than-or-equal match, RMS-11 adds one
to the relative record number and repeats steps 1 through 3.
- In Access by RFA, RMS-11 returns the appropriate error code:
ER$RFA
No valid record has ever existed at the specified location.
ER$DEL
The control byte in the cell indicates that the record in it was
deleted.
A find operation affects the Current Context for the Record Access Stream as
follows:
• find in Sequential Access Mode:
Current Record
Set to the relative record number of the record found, that is, the Next
Record before operation started.
Example You've connected a stream to a Relative file. There is no Current Record, but
the Next Record is the first record in the file. If you execute a sequential find
operation, the Current Record is set to the first record in the file.

Next Record
Set to relative record number one higher than relative record number for
Current Record.
Example From the previous example, Next Record is the second record cell in the file.

• find in Random Access Mode or Access by RF A:
Current Record
Set to record found, that is, the record identified by the relative record
number or RFA.

Relative File Applications

4-9

Next Record
Unchanged.
Example In the previous examples, you've done a sequential find after connecting the
stream to the file. You now execute a find by RFA. The Current Record is set
to the record specified, but the Next Record is not changed. Therefore, if you
do another sequential find, Current Record will be set to the second record
cell in the file, not the cell following the one found by RFA.

You use find instead of a get operation because:
• it is quicker because the record is not moved to the user buffer. Although the
time required to move a record from one part of memory to another is very
short, there is no use expending it if you do not need to.
• it does not change Next Record in Random Access Mode or Access by RFA.
This allows you to branch off sequential processing for purpose of updating
or deleting and yet keep your place.
You can use a find operation in the following ways:
• To skip records in Sequential Access Mode by initiating successive find
operations.
• To establish a random starting point for sequential processing with RF A.
You could then initiate successive get operations, where the first one gets
the record found by RFA.
• To establish a Current Record for a delete or update operation.
• To determine the existence of a record in Random Access Mode.
4.3.1.5 Flush - See "Records Operations," Section 1.2.4, for a summary
description of the flush operation.

A flush operation does not affect the Current Context for the Record Access
Stream.
4.3.1.6 Get -

To perform a get operation on a Relative file, RMS-11:

1. determines the location of the record in the file according to the specified
access mode:
• In Sequential Access Mode, location is indicated by:

- the Next Record pointer, if get was not immediately preceded by a
successful find operation
- the Current Record pointer set by an immediately preceding find
operation
• Location is determined by the specified relative record number in Random Access Mode.
• Location is determined by specified RFA in Access by RFA.

4-10

Relative File Applications

-!1111:+,

2. reads the bucket containing the indicated cell from disk into the task's I/0
buffer, if it is not already in memory.
Example Your fixed-length records are 50 bytes long; bucket size is two blocks. When
you read sequentially through the file, RMS-11 must request a disk I/O
operation every twentieth get record operation your program executes.

NOTE
If you have opened a Relative file with read-only

access and allow write declarations, each get operation causes an I/O operation.

3. checks the contents of the cell:

'"-"

• If the cell contains a valid record, RMS-11 returns the RF A to the

program and moves the record from the I/O buffer to the specified user
buffer in the program - unless the program is operating in Locate
Record Transfer Mode.
• If the cell is empty or contains a deleted record, the response depends on

the access mode:
- In Sequential Access Mode, RMS-11 repeats steps 1 through 3.
- In Random Access Mode, RMS-11 reacts according to the specified
match criterion:

On equal match, RMS-11 returns the error code ER$RNF.
On greater-than or greater-than-or-equal match, RMS-11 adds one to
the relative record number and repeats steps 1 through 3.
- In Access by RFA, RMS-11 returns the appropriate error:

ER$RFA
No valid record has ever existed at the specified location.
ER$DEL
The overhead byte in the cell indicates that the record in
it was deleted.

A get operation affects the Current Context for the Record Access Stream as
follows:
• get in Sequential Access Mode not immediately preceded by a find
operation:
Current Record
Set to the relative record number of the record read. See "Find in Sequential Access Mode" for example.

Relative File Applications

4-11

Next Record
Set to relative record number one higher than relative record number for
Current Record. See "Find in Sequential Access Mode" for example.

• get in Sequential Access Mode immediately preceded by a successful find
operation:
Current Record
Unchanged (from Current Record set by find operation),
Next Record
Set to relative record number one higher than relative record number for
Current Record (possibly changing Next Record set by find operation),
• get in Random Access Mode or Access by RF A:
Current Record
Set to the relative record number of the record read.
Next Record
Set to relative record number one higher than relative record number for
Current Record. This differs from find by RFA which does not change
Next Record.

4.3.1.7 Put -To perform a put operation on a Relative file, RMS-11:
1. determines the destination of the record in the file according to the speci-

fied access mode:
• In Sequential Access Mode, the Next Record pointer indicates destination.

• In Random Access Mode, the specified relative record number indicates
destination.
2. determines if the bucket containing the indicated cell is in the file. If it is,
RMS-11 goes to the next step. If it is not, RMS-11 extends the file until it

has enough blocks for all buckets up to and including the required one.
Then, RMS-11 initializes all newly allocated buckets.
3. reads the bucket containing the indicated cell from disk into the task's 1/0
buffer, if it is not already in memory. The bucket may be in memory if it
was required by a previous operation.
4. checks the indicated cell: if it contains a valid record, returns error code

ER$REX; otherwise, goes to next step.
5. moves the record from the user buffer in the program to the task's 1/0
buffer.
6. writes the 1/0 buffer to disk, unless you have specified Deferred Write (see
"1/0 Techniques," Section 4.3.3).

4-12

Relative File Applications

""'

A put operation affects the Current Context for the Record Access Stream as
follows:
• put in Sequential Access Mode:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Cell with relative record number one higher than relative record number
of former Next Record (the cell where this put operation inserted a
record).
• put in Random Access Mode:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Unchanged.

4.3.1.8 Rewind - A Rewind operation sets the context of the Record Access
Stream to the beginning of the Relative file. In doing so, it affects the Current
Context for the stream as follows:

Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Set to first record cell in file.

4·.3.1.9 Update - In an update operation, RMS-11 moves the specified record
from the task's user buffer to the 1/0 buffer, replacing the Current Record set
by the prerequisite get or find operation. Then, RMS-11 writes the bucket
over its original location on the disk, unless you have specified Deferred Write
(see "I/0 Techniques," Section 4.3.3).

An update operation requires a valid Current Record. Therefore an update
should follow a successful get or find operation; otherwise, RMS-11 returns
the error code ER$CUR. This error does not affect the original record in the
file on disk.
An update operation affects the Current Context for the Record Access
Stream as follows:
Current Record
None. Any operation requiring a Current Record will fail at this point.
Next Record
Unchanged.

Relative File Applications

4-13

4.3.2 Record Transfer Modes
You can manipulate records either in the 1/0 buffer or in your program's user
buffer (see Figure 4-1). Each of these options is called a Record Transfer
Mode. You can change Record Transfer Mode at run time, even between
record operations.

.-.,,
---~!

Figure 4-1: RMS-11 Task Structure

jS1zE DEPENDS ON:- -

I• NUMBER OF FILES OPENED SIMULTANEOUSLY :
1 • BUCKET SIZES

~ NUMBE~i:.!IECrD~CCESS STREA~ _ _ J

USER BUFFERS

-===:t

(

1/0
.BUFFERS
VIRTUAL
MEMORY

RMS-11

PROGRAM

INTERNAL
CONTROL
STRUCTURES

fS1ZEDEPENDS ON:-

J:_ ~

I• RMS-11 FUNCTIONS USED I
~~VERLA ~TRUCTURE~~E~

H-M K-00069-00

4.3.2.1 Move Mode - Move Mode requires that each record be copied between the user and 1/0 buffers:

• On get operations, RMS-11 moves the record from the 1/0 buffer to the user
buffer before returning control to your program.
• On put and update operations, your program assembles the record to be
written into the file in the user buffer, and during the operation, RMS-11
moves the data into the I/O buffer before updating the file.
Move Mode is the default Record Transfer Mode for all programming languages on all file organizations.
4.3.2.2 Locate Mode - Locate Mode enables your program to manipulate
records in the 1/0 buffer, eliminating the data transfers between it and the
user buffer. However, when you specify Locate Mode, RMS-11 uses it only

4-14

Relative File Applications

when such usage does not compromise data integrity. Otherwise, RMS-11
uses Move Mode. Therefore, your program must still contain a user buffer.
Example RMS-11 uses Move Mode instead of Locate Mode when a Relative file is shared.
Example RMS-11 uses Move Mode instead of Locate Mode if you opened a file indicating you
were going to perform update operations on it.

RMS-ll's use of Move Mode instead of Locate Mode is transparent to your
program as long as you use RMS-11 facilities to access the record data.
For Relative files, your program can only get records in Locate Mode. See your
language documentation to determine if the language supports Locate Mode
and if it does, what the exact programming techniques are.

4.3.3 1/0 Techniques

,.._,..

You can use the following techniques to improve the performance of record
operations.
4.3.3.1 IAS/RSX-11 M Asynchronous Record Operations-Within each Record
Access Stream, your program can perform any record operation either synchronously or asynchronously. In synchronous operations, RMS-11 returns
control to your program after the operation ends, either successfully or with an
error.

NOTE
If you intend to use asynchronous RMS-11 record operations

and/or Asynchronous System Traps (ASTs) in other parts of
your program, see the section on your operating system in
Appendix A.

4.3.3.2 Deferred Write - Normally, each write-type record operation (delete,
update, and put) results in a bucket being written to disk. This convention
emphasizes data integrity: you know that when a write-type operation has
ended successfully, the file reflects that operation.

However, you can improve the performance of sequential write-type operations by using Deferred Write. Basically, Deferred Write directs RMS-11 to
write a bucket to disk only when RMS-11 must use the 1/0 buffer for some
other purpose.

Relative File Applications

4-15

NOTE
Deferred Write, although not illegal, is essentially invalidated
while a Relative file is being shared by multiple tasks or
streams. In that environment, every write-type operation results in an I/0 operation so that:
• The bucket locked by the prerequisite get or find (for update
and delete operations) or by the put operation can be
released.
• The new data is available to the other tasks or streams.

Therefore, if you perform sequential write-type operations on a nonshared
Relative file, Deferred Write improves performance. RMS-11 writes out the
buffer only when it must read another bucket to complete an operation.
Example Your records are 304 bytes long and the bucket size is three blocks. During sequential write-type operations, Deferred Write causes 1/0 operations per bucket to drop
from five to one.

Deferred Write offers little or no benefit to random write-type operations or
read-type operations of any mode.
Only your application program can specify Deferred Write:
MACR0-11
Use the initialization macro or $SET to set the value FB$DFW in the FOP
field of the F AB of the Relative file.
BASIC-PLUS-2
BASIC-PLUS-2 does not support Deferred Write.
PDP-11 COBOL
PDP-11 COBOL does not support Deferred Write.
RPG II
RPG II does not support Deferred Write.
DIBOL
DIBOL does not support Deferred Write.
4.3.3.3 Multiple Buffers - When you open a Relative file, normally RMS-11
allocates one bucket-sized I/0 buffer in your task's address space. RMS-11
uses this buffer during record operations. However, you can direct RMS-11 to
allocate more than the one buffer.

RMS-11 uses any extra buffers to keep, or cache, buckets in memory. When a
record operation requires that a bucket be read from disk, RMS-11 checks its
cache first. RMS-11 does not perform an 1/0 operation if both the following
are true:
• The requested bucket is already in memory.

4-16

Relative File Applications

• That bucket is still valid, that is, the file is not shared and/or the bucket has
been kept locked.
You do not benefit from multiple buffers during sequential operations. You
can improve performance with multiple buffers during random operations
only if your program accesses the same buckets often.
4.3.3.4 Multiple Record Access Streams - RMS-11 allows each program to
use from one to 255 streams on a Relative file.

4.3.4 File Operations
You can perform the following file operations on Relative files. File operations
do not involve records and can only perform synchronously.
4.3.4.1 Close - A close operation disconnects all Record Access Streams
connected to a file before it releases access.
4.3.4.2 Create - In addition to the file specification, RMS-11 passes the
following information to the file processor when it creates a file:

• An initial allocation of blocks for the file.
• The locations on a specific device where the processor should allocate those
blocks.
• The following file attributes:
File organization
Record format
Forms control
Record size
Number of virtual blocks in the file
Bucket size
Default extension quantity
The other file attributes, such as the Virtual Block Number of the first
block in the first bucket and the last initialized block, RMS-11 stores in the
Prologue of the file.
4.3.4.3 Open -

You can specify the file you want to open in two different

Relative File Applications

4-17

On IAS/RSX- llM systems, open by File ID is significantly faster than open
by filespec, because the process bypasses directory reads and other overhead.
However, on RSTS/E, open by File ID is no faster than open by filespec.
4.3.4.4 Erase -

"File Operations," section 1.2.5.3, introduces the concept of

erasing files.
4.3.4.5 Extend -

"File Operations," section 1.2.5.3, introduces the concept of

extending files.

4-18

Relative File Applications

Chapter 5
Indexed File Organization

DIGITAL designed the RMS·-11 Indexed file organization to achieve the
following goals:
Contents-addressable record access
Each record in the file can be located on the basis of the values in designated portions of the data, called key fields.
Uniform random access time
Each record in the file can be located with approximately the same number of I/O operations, regardless of when it was added to the file.
Alternate Key capabilities (comply with ANSI CQBOL Level 2)
Each record in the file can be located via more than one key field.
Very good performance on sequential access by Primary Key
A program can sequentially read a reasonably designed Indexed file by
Primary Key almost as fast as it can sequentially read a Sequential file.
Good performance on sequential access by Alternate Keys
Each record in the series can be accessed with (typically) one to three I/0
operations.
Unique record address for the life of the file (data base key concept)
A record in a file can be located via a unique identifier (Record's File
Address) established by the put operation. The record may be deleted, but
its unique identifier is never reused.
Preserve state of processing despite system failure
Normally, each logical write operation results in a physical transfer of
data from memory to disk. Therefore, the file reflects each record inserted.
However, you can override this mode with Deferred Write.

5-1

More importantly, RMS-11 performs record operations so that both of the
following are true:
• File corruption is avoided or minimized even if the system failure occurs
during a write-type record operation.
• Even if some corruption exists, user data can still be accessed.
NOTE
You should still reorganize your file if the system fails during
write-type processing on an RMS-11 Indexed file.

5.1

Physical Structure
On disk, an Indexed file consists of three kinds of blocks:
Prologue
RMS-11 information about the file, including attributes and key and area
descriptions
Index
Index records for Primary and Alternate Keys pointing the way to a data
record
Data
Your data records and index data records
The Prologue contains information about the keys and areas of the file.
RMS-11 allocates at least one block for the Key Descriptors and at least one
block for the Area Descriptors. RMS-11 uses more blocks as needed. Size
calculations are discussed in "Initial Allocation," Section 6.6.1.
Also, if both of the following are true:
• you have defined the same bucket size for all areas of the indexed file
• that bucket size is two, four, or eight blocks
RMS-11 extends the Prologue to an integral multiple of bucket size. This step
can improve performance by aligning buckets with file clusters. If the bucket
size does not meet these criteria, RMS-11 does not make the Prologue larger
than necessary. "Bucket Size," Section 6.5, contains more discussion.
The location of the index and data blocks is up to you (see Figure 5-1):
• If the file is a single area, RMS-11 allocates data and index blocks in

buckets as it needs them; they are therefore interspersed throughout the
file.
• If the index and data are set up in separate areas, RMS-11 allocates each

type of bucket from the appropriate area; the index is therefore set apart
physically from the data portion of the file.

5-2

Indexed File Organization

Figure 5-1: Indexed File with and without Areas
~··

PRIMARY
INDEX

DATA
RECORDS

WITHOUT AREAS

Pl = PRIMARY INDEX
DR = DAT A RECORDS
Al= ALTERNATE INDEX

WITH AREAS

F-M K-00099-00

Indexed File Organization

5-3

RMS-11 formats buckets in an Indexed file as it requires them for record
storage. The RMS-11 control bytes are set to their initial values (see Figure
5-2):
• Fourteen bytes beginning with byte 0 of the bucket contain bucket control
information.
• The last byte of the last block duplicates the first byte of the bucket for
checking 1/0 completion.

RMS-11 packs index or data records, including record format overhead, into
each bucket, beginning with byte 14, end-to-end and byte-aligned.

Figure 5-2: Formatted Bucket

~-----BUCKET BOUNDARIES---~

RECORDS

CONTROL FLAGS
LEVEL IN INDEX
- - - - VBN OF NEXT BUCKET IN LEVEL
- - - - - - R E C O R D IDENTIFIER INFORMATION
- - - - - - - - POINTER TO FREE SPACE IN BUCKET
- - - - - - - - - - BUCKET ADDRESS SAMPLE
- - - - - - - - - - - A R E A CONTAINING THIS BUCKET
-----------~CHECKBYTEEQUALSLASTBYTE-----

H-MK-00063-00

5.2 Conceptual Structure
No matter how it is laid out physically, the Indexed file is conceptually a
Prologue plus a group of indexes, one per key. Each index consists of horizontal chains of buckets called levels and can be illustrated as a pyramid (see
Figure 5-3).

5-4

Indexed File Organization

Figure 5-3: Index as a Pyramid

LEVEL 2

LEVEL 1

LEVEL 0

O-MK-00062-00

BUCKET

The lowest level of an index is Level 0. The level number is incremented for
each successive (and smaller) level, that is, Level 1, Level 2, and so on. The
highest level in an index is a single bucket called the Root; this bucket is the
entry point to the index for random accesses using this key. Each index has at
least two levels (0 and 1).
The depth of an index is equal to the level number of the Root. An index
depth relates to the time needed to randomly access any record in the file via
that index.

5.2.1

Data

Level 0 of each index is called the data level; it consists of data buckets. In the
Primary index, Level 0 contains buckets of your data records. In the Alternate
indexes, Level 0 buckets contain pointers to your data records.
5.2.1.1 Level 0 of the Primary lndex-RMS-11 physically orders data records
by increasing Primary Key value along the bucket chain. The records having
the lowest Primary Key value reside in the first bucket of the level and so
on until the records with the highest Primary Key values comprise the last
bucket. RMS-11 preserves this order regardless of the insertion sequence
of the records.

Each bucket in Level 0 shares the following properties:
• The last data record in a bucket has an equal or higher key value than any
other record in the bucket.
•• The last data record in a bucket has a lower key value than the first record
in the next bucket in the chain.
With these properties, each bucket has a High-Key value, located in the last
record of the bucket. This concept is the core of RMS-11 Index file structure.

Indexed File Organization

5-5

NOTE
RMS-11 places records with duplicate key values next to each
other on a first-in, first-out (FIFO) basis. If these duplicate
records cannot fit in the same bucket, RMS-11 stores the overflow in a Continuation Bucket. Continuation buckets are extensions of Level 0 buckets and as such, are not indexed. This
extension storage preserves the High-Key concept.
5.2.1.2 Level O of the Alternate Indexes - Levels 0, the data levels, of Alternate indexes contain Secondary Index Data Records (SIDRs). A SIDR
consists of (see Figure 5-4):

• an Alternate Key value from a data record stored in the Primary data level.
The SIDRs in the data level of each Alternate index are stored in ascending
order by this key value.
• one or more pointers to data records in the Primary data level. Multiple
pointers occur when you allow duplicates for the Alternate Key and records
with duplicate values for the key actually exist in the file.

Figure 5-4: Format for Secondary Index Data Record
A. DUPLICATES ALLOWED
DUPLICATE
COUNT

SIZE
OF
ARRAY

VALUE

SIDA IDENTIFIER
CONTROL

B. NO DUPLICATES ALLOWED
Sl7E
OF
SIDA

KEY

SIDA IDENTIFIER
CONTROL

POINTER ARRAY

KEY

VBN

72E

DATA RECORDjD
FLAG BYTE
POINTER TO PRIMARY

LEVEL 0
1-~~....--,----.~~~--[J
VALUE

VBN

DATA RECORD I D J
FLAG BYTE
POINTER TO PRIMARY
LEVEL 0

Q-MK-00061-00

5.2.2 Indexes
Levels 1 and above in an index are called the index levels; they consist of
index buckets. Index buckets contain index records that guide RMS-11
through the levels to the data records in Primary Level 0. An index record
consists of:
• the High-Key value of a bucket in the next lower level in the index. Since
RMS-11 arranges these values in ascending sequence, there is a High-Key
value for index buckets also. However, the last High-Key value in the last
index bucket of a level is set to the highest possible key value, rather than
the highest key value in the file. The associated pointer references the last
bucket in the next lower level.

5-6

Indexed File Organization

• a pointer to the bucket associated with the High-Key value
Example The buckets in Level 1 of the Primary index contain the High-Key values of the data
buckets in Level 0. Then Level 2 contains the High-Key values from Level 1 and so
on. Figure 5-5 shows a sample Primary index.

Figure 5-5: Example of a Primary Index

LEVEL 3
(ROOT)

MAXIMUM KEY VALUE

LEVEL 2

LEVEL 1

MAXIMUM KEY VALUE

In other words, each bucket on a given level is represented by an index record
in the next higher level. Thus the number of buckets required on each successive level decreases exponentially until the Root bucket is reached.
Example If an index bucket can hold ten index records, then:
• If Level 0 contains 2000 data buckets,

Level 1 contains 200 index buckets
Level 2 contains 20 index buckets
Level 3 contains 2 index buckets
Level 4 contains 1 index bucket
• If Level 0 contains 10,000 data buckets,

Level 1 contains 1000 index buckets
Level 2 contains 100 index buckets
Level 3 contains 10 index buckets
Level 4 contains
1 index bucket

Indexed File Organization

5-7

5.2.3 Random Access Using the RMS-11 Indexed File Structure
The following steps show how RMS-11 uses this Indexed file structure to
execute a random access operation. These steps comprise a process called
"follow the index."
1. RMS-11 examines memory-resident Index Descriptors to find the location
of the Root for the specified index.

NOTE
The Root can be cached (see "1/0 Techniques," Section
7.3), eliminating the 1/0 operation to read the Root in the
next step.
2. RMS-11 reads the Root and scans for the first value greater than or equal
to the key value specified when the operation was initiated. If all else fails,

the search will find the highest possible key value in the last index record.
3. RMS-11 reads the bucket indicated by the pointer associated with the
selected key value and scans for the first key value greater than or equal to
the value specified. RMS-11 repeats this step through the levels until
Level 0 is reached.
Example Refer to Figure 5-5 during this example.

The specified Primary Key value is "YOOS."
RMS-11 determines the Virtual Block Numbers (VBNs) of the Root bucket from the
memory-resident Index Descriptors and requests the file processor to read those
blocks into an 1/0 buffer. RMS-11 scans the index records in the Root. The first key
value equal to or greater than "YOOS" is the maximum key value in the last record.
RMS-11 uses the bucket pointer in this index record to request another 1/0 operation. The file processor reads the specified blocks into the 1/0 buffer, and RMS-11
scans them looking for a key value equal to or greater than "YOOS." Again, it finds
no qualifying key value until the last record in the bucket, containing the maximum
key value. This index record points to a Level l bucket.
Upon RMS-11 's request, the file processor brings the indicated bucket into memory.
RMS-11 searches the bucket, terminating with the last record in the bucket which
contains the maximum key value.
The file processor reads the indicated Level 0 bucket at RMS-ll's request.

5.2.4 Why this Structure?
Mechanical data storage devices make 1/0 operations the slowest part of file
processing. Ideally, a file is read to memory when it is opened and maintained
there, without additional 1/0 operations, until the file is closed. Some very
small files allow this approach and are handled most efficiently by your own
search techniques rather than the indexing facilities of RMS-11.
However, most files you require for indexed access are very much larger than
the memory available for data buffering. Such files are partitioned into pieces

5-8

Indexed File Organization

that can be read to memory. RMS-11 calls these pieces buckets. By definition, one 1/0 operation is required to access one bucket.
If no index to the data exists, a task must sequentially scan through the

buckets of a file to find a specific record. Such a search, on the average,
accesses half the buckets in the file (see Figure 5-6).

F'igure 5-6: Search Time Curves
RMS-11 PRIMARY INDEX SEARCH,
10 RECORDS PER INDEX BUCKET•
BINARY SEARCH FOR PRIMARY KEY,
F~ECORDS ORDERED BY PRIMARY KEY

NUMBER OF
PRIMARY DATA
BUCKETS

NUMBER OF BUCKETS READ IN SEARCH

*LINE SHIFTS TO LEFT AND BREAK RIGHT MOVES UP
AS NUMBER.OF INDEX RECORDS PER BUCKET GOES UP,
ASSUMING OPTIMAL PACKING

H-MK-00058-00

Indexed File Organization

5-9

You can optimize nonindexed access by:
• ordering the records by a key value
• using a binary search technique
Then, the number of accesses required to find a record approaches log2 of the
total number of buckets (see Figure 5-6). This better, but still mediocre,
speed is realized on one of perhaps many keys.
The RMS-11 indexed structure uses buckets so that your programs can handle files more efficiently. In most cases, RMS-11 uses n+l I/0 operations to
locate a record by Primary Key, where n is the depth of the file's Primary
index.
In a small file, this technique is not appreciably quicker than a sequential
scan. However, given typical key sizes, a Primary index of depth 3 can represent from 1,000 to 125,000 buckets of data records, using only single-block
buckets. Normally, four disk accesses are needed to get any record by Primary
Key value.
Example You want to search 50 buckets of data for records with specific primary key values.
The average number of buckets you read during each search depends on your search
technique (see Figure 5-6):

25.5 buckets for a nonindexed search of unsorted records
5+ buckets for a binary search of records sorted by primary key
2
buckets for an RMS-11 indexed search

5.3 Procedures for Performing Random Record Operations
The procedures for performing random record operations on Indexed files
depend on the circumstances for the individual operation, the file's design,
and whether Alternate indexes must be updated.

5.3.1

Putting a Record

When your program initiates a put operation, RMS-11 moves the data from
the task to the proper bucket in Level 0 of the Primary index and updates all
indexes involved with the record. This process can be simple, requiring minimum VO operations. It can also be complex, requiring more procedures and
data transfers. The complexity depends on whether there is enough room for
the new record in its data bucket.
5.3.1.1 Slmplest Case - In the simplest put operation, RMS-11 finds room in
the target data bucket to insert the record. To execute the operation, RMS-11
performs the following steps:
1.

5-10

Determines the value of the Primary Key field from the record.

Indexed File Organization

2. Follows the Primary index to the proper Level 0 bucket.
3. Reads the Level 0 bucket and sequentially scans for the first record with a
Primary Key value greater than the specified value. Establishes a position
before that record, or after the last existing record in the bucket if:
• key values are equal
• the first record in the next data bucket has a higher key value
4. Compresses deleted records. RMS-11 can reuse bytes in a deleted record
depending on the record format and whether you allow duplicates in the
Primary Key field. "Key Characteristics," Section 6.2.5, discusses reusing
space from deleted records.
5. Dete:rmines if the record to be inserted fits in the bucket (in this simplest
case, it does).
6. Inserts the record at the established position. No Primary index buckets
are updated since no High-Key value has changed.
7. If there are Alternate Keys, updates those indexes, using the following
sequence of steps for each one:
a. Follows the Alternate index to the proper Level 0 bucket.
b. Reads the Level 0 bucket and sequentially scans for the key value
specified in the record:
• If a value higher than the one specified is found, inserts SIDR for the
record before the SIDR for the higher value.
• If a match is found, determines if duplicates are allowed for the

Alternate Key:
- If duplicates are allowed, follows the duplicate pointer array in
the SIDR to the end, then inserts a pointer to the newly inserted
record. This procedure preserves the first-in, first-out c.~mvention.
After the last Alternate Key, returns a successful completion code
to the program.
- If duplicates are not allowed, returns to Level 0 of the Primary

index, flags the newly inserted record as deleted, logically removing it, and returns an error code to the program.
Example Refer to Figure 5-5 during this example.

RMS-11 examines the record in the user buffer of the Record Access Stream initiating the put operation. The value in the Primary Key field is "JACKSON." RMS-11
locates the Primary Root and requests the file processor to read the bucket into the
1/0 buffer associated with the stream. When that 1/0 operation ends, RMS-11 scans
the bucket, looking for a key value equal to or greater than "JACKSON." It finds
"JONES."

Indexed File Organization

5-11

RMS-11 requests the bucket indicated by the pointer in the "JONES" index record.
When RMS-11 scans this Level 2 bucket, it finds that "JONES" again ends the
search. Following the pointer in this index record, RMS-11 requests another bucket.
Its search of the Level 1 bucket ends in another "JONES" index r~cord.

-'"""'

RMS-11 requests the Level 0 bucket indicated by this last index record. It finds that
a data record with a Primary Key value of "JONES" is the only occupant of the
bucket. There are no deleted records to compress, so RMS-11 writes the
"JACKSON" record before the "JONES" data record, moving it down in the bucket.
There are no Alternate Keys. RMS-11 returns a successful completion code to the
program.

5.3.1.2

Bucket Spllttlng -

If there is not enough room in the target data

bucket for the record, RMS-11 allocates a new bucket and reorganizes the
records in the old one between the two buckets in a procedure called bucket
splitting.
This insertion process is identical with the simplest case (Section 5.3.1.1) to
step 5 where RMS-11 determines if the new record fits in the bucket. When
there is not enough room, RMS-11 does the following:

1. Reads the appropriate Area Descriptor from the file Prologue. If enough
blocks for a bucket are allocated for the area, formats them into a bucket
and updates the Area Descriptor to reflect the new bucket. Otherwise,
RMS-11 requests the operating system to allocate enough blocks, then
proceeds as described.
2.

Splits the target bucket at the point where the record should be inserted.
Moves the records in the high portion of the bucket into the new bucket;
these records have Primary Key values higher than that of the new record.

NOTE
When RMS-11 moves a record between buckets, it marks
the record's original location with a Record Reference Vector (RRV). An RRV is a copy of the record's header (both
contain seven bytes). RRV s preserve Alternate Key and
RF A access, holding the original location of the record and
pointing to its current location. Only one RRV is created for
a record: if it moves again, RMS-11 updates the RRV with
the record's new location.
Since the original location of a record is filled, either with
the record or a pointer to that record, RMS-11 does not
have to update Alternate indexes every time a record
moves. This convention means one extra 1/0 operation may
be needed to find or get a record via an Alternate Key, but
it prevents a complex and costly index update for each
bucket split.
3. Inserts the data record in the original target bucket. If it will not fit,
inserts it into the new bucket. If it will not fit there, creates another
bucket (see step 1) and puts the record there.

5-12

Indexed File Organization

4. Updates the Level 0 bucket chain to include the new bucket(s).

5. Returns to the Primary Root bucket and follows the index to the Level 1
index bucket that points to the data bucket that split.
6. Inserts index record(s) for the new data bucket(s). If the index bucket
splits, uses a procedure similar to this to move the index records and
update the next higher level of the index. Splitting can occur all the way
to the Root where a new Root is created and the file Prologue updated.
7. If there are Alternate Keys, update those indexes as described in step 7 of
the simplest case (Section 5.3.1.1). Bucket splitting can occur in Alternate
indexes also.
5.3.1.3 Incremental Reorganization - The process of inserting each data record where it belongs in Level 0 and updating the indexes when RMS-11
inserts the record is called incremental reorganization of the file.

Incremental reorganization has the following advantages:
• eliminates reorganization periods where special software incorporates overflow areas into the main file and that file is not available for processing
• ensures equal access time to old and new records
• enables performance on sequential access by Primary Key to approach the
speed of sequential access on a Sequential file
Of course, this process has its costs: additional 1/0 operations occur when
a bucket splits. But with good file design and file loading, bucket splitting
(and the time for each bucket split) is minimal. Chapter 6 discusses these
considerations in detail.

5 . 3.2 Getting and/or Finding a Record
To execute a random get or find operation, RMS-11 performs the following
steps:
1. Determines from the instruction initiating the opera ti on the following
criteria:
• key of reference, that indicates the index to search and the key field
within the data record to examine
• value to find
• value match (equal to, greater than, or either)
• number of characters for value match

2. Follows the index to the proper Level 0 bucket.

Indexed File Organization

5-13

3. Reads the Level 0 bucket. Sequentially scans for the first record with a
value in the specified key field that matches the specified value according
to the value match criteria. This search can continue into other buckets.
• If no such record is found, returns an error code.
• If such a record is found:

a. Determines which index has been read:
• If this is the Primary index, goes to the next step.
• If this is an Alternate index, the record located is a SIDR. Follows

the SIDR pointer to the Primary Level 0 data record.
b. For a get operation only, moves the record to the user buffer associated with the Record Access Stream performing the operation.
c. Sets the current context for the stream performing this operation.
The effect of each record operation on context is described in "Record
Operations," Section 7 .1.
d. Returns successful completion code.
Example Refer to Figure 5-5 during this example.

RMS-11 determines the key (and index) of reference and the value to find from the
instruction initiating the operation. In this case, they are the Primary Key (key 0)
and "ABI."
RMS-11 locates the Primary Root and requests the file processor to read the bucket
into an 1/0 buffer. RMS-11 sequentially scans the Root for an index record whose
key value is equal to or greater than "ABI." It finds "ABRAM."
RMS-11 requests the bucket indicated by the pointer in the "ABRAM" index record. When RMS-11 searches this Level 2 bucket, it finds an index record containing
the key value "ABNER." Following the pointer in this index record, RMS-11 requests another bucket. The search of the Level 1 bucket ends in the key value
"ABLE.''
RMS-11 requests the Level 0 bucket indicated by this last index record. RMS-11
changes its search criteria to that specified in the initiating instruction: it looks for a
record where the first three bytes of the Primary Key field equal "ABI." Since the
only record in the bucket contains "ABLE" in its Primary Key field, RMS-11
cannot satisfy the search requirements. It returns a "record not found" error code to
the program.

5.3.3 Updating a Record
RMS-11 requires an update to be preceded by a get or find operation,
although some higher level languages hide this prerequisite.
To execute an update operation, RMS-11 performs the following steps:

5-14

Indexed File Organization

1. Locates the key fields of the revised record in the user buffer associated
with the Record Access Stream performing the operation. Compares those
key values with the values in the Current Record:
• If the Primary Key value changed, returns an error code.
• If an Alternate Key value changed, checks if you allowed changes for

that key:
- If not, returns an error code.
- If so, continues processing.

2. For each Alternate Key where the key value changed, deletes the preupdate value from the Alternate index:
a. Reads the data bucket containing the Current Record, if it is not in
memory.
b. Saves the pre-update Alternate Key value from the Current Record.
c. Follows the index to the Level 0 bucket that should contain the SIDR
for the pre-update key value.
d. Reads the Level 0 bucket and sequentially scans for the pre-update key
value:
• If a value higher than the one specified is found, goes to the next

Alternate index.
Example RMS-11 is scanning a bucket, searching for a pre-update key value of
"D". It finds a record with a key value of "E". Since "E" is greater than
"D", RMS-11 ends the search and this step in the procedure.

• If a match is found, scans the duplicate pointer array in the SIDR to

the entry for the record being updated and flags it as deleted.

NOTE
To allow keys to change, RMS-11 requires that you also
allow duplicates. Therefore, if you allow Alternate Key values to change, there is a duplicate pointer array in the
SIDR for each key value.
However, PDP-11 COBOL enables you to specify changes,
but no duplicates for keys. In this situation, the PDP-11
COBOL Object Time System (OTS) uses a find operation
to check whether an update will cause a duplicate; if it
does, the OTS returns an error to your program.
3. Reads the data bucket containing the Current Record, if it is not in
memory. Replaces the Current Record in the 1/0 buffer with the updated
version in the user buffer.
4. Writes the bucket to the file.

Indexed File Organization

5-15

5. For each Alternate Key where the key value changed, inserts the postupdate value in the Alternate index:

_,....._
1

a. Reads the data bucket containing the Current Record, if it is not in
memory.
b. Follows the index to the Level 0 bucket that should contain the postupdate key value.
c. Reads the Level 0 bucke.t and sequentially scans for the post-update
key value:
• If a value higher than the one specified is found, inserts a SIDR for

the new record before the SIDR for the higher value.
• If a match is found, follows the duplicate pointer array in the SIDR

to the end, then inserts a pointer to the new record. After the last
Alternate Key is updated, returns a successful completion code to
the program.

5.3.4 Deleting a Record
RMS-11 requires a delete operation to be preceded by a get or find, although
some higher level languages hide this prerequisite.
To execute a delete operation, RMS-11 performs the fo,llowing steps:
1. If there are Alternate Keys, updates those indexes, using the same sequence of steps for each:

~
--,

-·

a. Reads the data bucket containing the Current Record, if it is not in
memory.
b. Follows the index to the Level 0 bucket that should contain the SIDR
for the key value in the deleted record.
c. Reads the Level 0 bucket and sequentially scans for the specified key
value:

"""

• If a value higher than the one specified is found, goes to the next

Alternate index, if any.
• If a match is found, determines if you allow duplicates:
- If so, follows the duplicate pointer array in the SIDR to the entry

for the record being deleted and flags it as deleted.
- If not, deletes the SIDR.
2. Reads the data bucket containing the Current Record, if it is not in memory.

3. Changes the flag byte in the header of the Current Record to indicate that
it is deleted.

5-16

.;a ea::

Indexed File Organization

I ..

4. Writes the bucket to the file.
5. If the record has moved, reads the Level 0 bucket containing the RRV.
Changes the flag byte in the RRV to indicate that it is deleted.
6. Writes the bucket to the file.

RMS 7 1:L does not compress a deleted record until it needs space to insert
anoth1er user data record into the bucket (see "Putting a Record," Section
5.3.1). RMS-11 does not compress deleted RRVs.

NOTE
RMS-11 does not modify or reduce any index structure or allocation during a delete operation.

5.4 Procedures for Performing Sequential Record Operations
Your program may use the Sequential Access Mode to perform the following
record operations:
•Find
•Get
•Put
During sequential get and find operations, RMS-11 does not usually read an
index to locate the specified record. Instead, RMS-11 uses the Next Record
pointer for the stream performing the operation to identify the proper data
bucket. Then, RMS-11 requests the file processor to move that bucket into
the 1/0 buffer, if it is not in memory. If it has requested a SIDR bucket,
HMS-11 then follows the appropriate pointer to the user data record.
During sequential put operations, RMS-11 compares the Primary Key value
of the specified record with the Primary Key value of the last record written:
• If the specified record's Primary Key value is equal to or greater than the

last record's Primary Key value, RMS-11 performs a random put operation
(described in Section 5.3.1).
• If the specified record's Primary Key value is less than the last record's

Primary Key value, RMS-11 returns an error code to the program.

5.5 1/0 Cost of Performing Record Operations
Table 5-1 provides simple algorithms for predicting the number of 1/0 operations any RMS-11 record operation requires:

• n =index depth of the indicated key; all indexes do not necessarily have the
same depth.
• Algorithms do not include 1/0 operations caused by program or RMS-11
overlays, operating system overhead, or by file extensions (discussed in
Chapter 8).

Indexed File Organization

5-17

Table 5-1: 1/0 Cost of Performing Record Operations

Record Operation

Primary
Key

Each
Alternate
Key

Getting or Finding a Record Randomly
Record in original location
RRV in original location (record moved)

n+l
n+l

n+2
n+2

Getting or Finding a Record Sequentially

0-2 1

1-4~

n+2
2n+6 3
(n**2+11n+20)/2 4

Updating a Record 5
Alternate Key value did not change
Alternate Key value changed

1
1

0
2(n+2) 6

Deleting a Record
Record in original location
RRV in original location (record moved)

1
3

n+2 7
n+2 7

Putting a Record
Simplest case
Split in data level
Bucket split up entire index

Breaks down to:
0 or 1 1/0 to position to Current Record
0 or 1 1/0 to locate Next Record

Breaks down to:
0 or 1 1/0 to position to SIDR for Current Record
0 or 1 1/0 to locate SIDR for Next Record
1 or 2 I/Os to retrieve user data record
Breaks down to:

n+2
n+l

3
4

I/Os to read and write the old bucket
I/Os to read and write the Level 1 index bucket
I/Os to write the new bucket and update the Area Descriptor in the Prologue

Breaks down to:
(n+2) + (n+l) + n + (n-1)+ ... + 3 + 2 I/Os to return to the Primary Root and read and
write updated buckets from Level 0 to the Root
3(n+l) I/Os for each bucket split (see footnote 3)
5 I/Os to create the new Root

Values assume record length does not change and cause bucket splitting.

n+2

Value is different if one of the following is true:

if either the old or new key value doesn't belong in the index; for example, the field
contains the null key value defined for the key, or a variable-length record does not
contain the whole key field.

• You specified the "fast delete" option (available in MACR0-11 only) when you initiated the delete operation. Then, RMS-11 does not update the Alternate indexes.
• RMS-11 has to scan a long duplicate array into one or more Continuation Buckets.
Then, one I/0 operation is needed for each additional bucket.

5-18

Indexed File Organization

Chapter 6
Indexed File Design

Indexed file design ranges from the basic elements of your application (record
definition and key selection) to the structure of the file to the methods used to
put the data into the file. This range includes:
1. Record definition

2. Key selection
3. Areas
4.

Placement control

5. Bucket size
6. Allocation

7. Population techniques

6.1

Record Definition
You can use only fixed- and variable-length records in RMS-11 Indexed files.
RMS-11 calculates their lengths as follows:
7

rfo

bytes for RMS-11 record header
bytes for record format overhead (0 for fixed; 2 for variable)

+ ds

bytes in the data itself

bytes for each valid data record

6-1

Set your record size to reflect application requirements; do not adjust it to fit
bucket size. For instance, if you are using one-block buckets, do not set a
record length so the records just fit into the buckets:
512
-15
497
- 7
490

bytes in a block
bytes Indexed file overhead per bucket
bytes left for records
bytes for the record header
bytes left for the data and record format overhead

This coordination seems ideal at first. However, when the record moves during a bucket split or RMS-11 deletes the record, and some RMS-11 overhead
is left in the bucket, a normal data record cannot fit: the bucket is essentially
useless, with up to 490 bytes of unused space.
Of course, if your application requires 490-byte records, use them, recognizing
the preceding limitation and perhaps, choosing a different bucket size. However, if you can avoid that or comparable lengths for larger buckets, do so.
NOTE

Records in an Indexed file cannot span buckets and bucket
sizes are limited by the operating systems. Therefore, the maximum record size, including overhead, is 16,369 or 7,665 bytes.

6.2 Key Selection
A file's keys cause a significant portion of an Indexed file and of the 1/0
operations needed to access the file. During key selection, you should consider
the following:
• N um her of keys
• Key size
• Key data type
• Position of key in record
• Key characteristics

6.2.1

Number of Keys

You can specify from 1 to 255 keys for an Indexed file:
• One Primary Key that RMS-11 requires for every file
• 254 Alternate Keys
Cost For each key specified in an Indexed file, RMS-11 builds an index.
Since RMS-11 requires a Primary Key, you must accept that key's

6-2

Indexed File Design

~""at4

index overhead, but consider the cost before specifying an Alternate
Key for the file:
• RMS-11 updates Alternate indexes when your program:
-

puts a new record into the file

updates a record in the file and the Alternate Key values change

deletes an existing record

The time required for this update relates to the number of I/0 operations needed to follow each Alternate index from the Root to Level 0,
to change or insert the SIDR, and to rewrite the bucket. RMS-11 can
require additional time if one or more buckets in the index split.

• An index takes room in the file. You can estimate the disk space for
an Alternate index (see "Initial Allocation," Section 6.6.1).
Whether the cost of each Alternate Key is bearable depends on your application. If its primary purpose is to put, update, or delete records, each Alternate
Key will noticeably burden the operations; therefore, the number of Alternate
Keys should be kept to the minimum. Rarely used alternate access paths call
for a separate program that sorts by the desired nonkey field and then processes the data.
However, if the primary purpose of your application is to get records from the
file, then Alternate Keys do not burden processing. In fact, Alternate Keys
give flexibility to information retrieval. However, the cost of the extra keys is
borne on those few occasions when records are added to the file.

6.2.2 Key Size
Keys for Indexed files have length restrictions according to their data types, as
shown in Table 6-1; see also Section 6.2.3.

'rable 6-1: Key Data Types
Data Type
String

Length (bytes)
1-255

15-bit signed integer

31-bit
signed integer
,

16-bit unsigned binary

32-bit unsigned binary

Packed Decimal

1-16

Cost The cost of each key's size is borne in the data record and in the index:
RMS-11 stores an entire key value in each index record. However, there
is only one index record for each bucket in the next level down.

Indexed File Design

6-3

6.2.3 Key Data Type
Each key in an Indexed file can be one of the following data types:
String
RMS-11 interprets each character of the key in a byte by its binary contents. Permissible values are not limited to valid ASCII codes.
Example The key value "RMS-11" is represented as follows:
7

Most significant byte = R

01010010
01001101
01010011
00101101
00110001

00110001

The first (lowest-addressed) byte of the key is the most significant byte of
a string key for collating purposes. RMS-11 compares Primary Keys byteby-byte first-to-last when it determines where the record should be placed
in the file.
The maximum key value is all bits in each byte set to 1 (377 J.
Cost The number of bytes specified as size. For example, if you specify a
key length of 12, each representation of the key in the data record
and in the index takes 12 bytes.

Two-Byte Signed Integer
Each key requires two bytes; RMS-11 interprets the data in the following
format:
7

Least significant byte
Most significant byte

Sign

NOTE
The least significant byte of an integer or binary key is the byte
with the lowest address. Significance increases with address.
Within a byte, the lowest significant bit is bit 0, and significance increases with position. See your PDP-11 Processor
Handbook.

6-4

Indexed File Design

A two-byte signed integer can represent the decimal values -32, 768
through +32, 767.
Maximum key value is +32,767.
Cost Two bytes per representation.

Four-Byte Signed Integer
Each key requires four bytes; RMS-11 interprets the data in the following
format:
7

~ Least significant byte

E==1

Most significant byte

Sign
A four-byte signed integer can represent the decimal values -2,147,483,648
through + 2, 147 ,483, 64 7.
Maximum key value is +2,147,483,647.
Cost Four bytes per representation.

Two-Byte Unsigned Binary
Each key requires two bytes; RMS-11 interprets the data in the following
format:
7

====1 Least significant byte

~ Most significant byte

A two-byte unsigned binary can represent the decimal values 0 through
+65,535.
Maximum key value is 65,535.
Cost Two bytes per representation.

Four-Byte Unsigned Binary
Each key requires four bytes; RMS-11 interprets the data in the following
format:
7

F==i

Least significant byte

E===3Most significant byte

Indexed File Design

6-5

A four-byte unsigned binary can represent the decimal values 0 through
+4,294,967,295.

Maximum key value is 4,294,967,295.
Cost Four bytes per representation.

Packed Decimal
RMS-11 recognizes two decimal digits of the key in each byte except the
last. The key format takes the following form:
7

~4 A
~

A+l

Idi Isign I A+N-1
where:

dl, d2, through di are decimal digits, and
dl is the most significant digit
di is the least significant digit
sign

has a value 10 through 15 where
+ is represented by a 10, 12, 14, or 15
- is represented by an 11 or 13

is the length of the key in bytes (maximum of 16)
is the length of the digit string, an odd number in the range 1
through 31, where

i= 2N-1
Maximum key value is 99 in each byte with the sign positive.

6.2.4 Position of Key In Record
You can locate any key anywhere in the record:
• Alternate Keys can precede the Primary Key.
• Keys can overlap each other 1 •
You benefit from careful placement of keys within the record:
1

6-6

PDP-11 COBOL, in keeping with the ANSI standard, does not permit more than one key to
start at the same position. The standard calls it leftmost correspondence.

Indexed File Design

Deleting a Record
When you allow duplicates in the Primary Key of variable records,
RMS-11 compresses a deleted record by removing all data except:
• the record header
• enough of the record to contain the Primary Key
Therefore, you can optimize record deletion if you place the Primary Key
at the beginning of the record. The closer the key is to the beginning of the
record (and the shorter the key), the less overhead remains in the file.
However, if you have fixed-length records or do not allow Primary Key
duplicates, the position of that key in the record is not significant.
Putting a Record
You can optimize put operations for variable-length records only, by placing Alternate Keys at the end of the record. Then, if no valid data is
present in an Alternate Key field, you can shorten the record to exclude
that field, thus reducing the record space in the data level as well as
eliminating a reference to that record in the Alternate index.
You can segment string keys; all other key data types must be contiguous
bytes. You can specify up to eight segments in one string key, each segment
with its own length; the total of the lengths cannot exceed 255 bytes.
RMS-11 concatenates the segments you specify before performing any operations requiring a value for the key. RMS-11 defines a segment by byte position within the record and length in bytes. Therefore, the key segments you
define with either a MACR0-11 program or the RMSDEF utility do not have
to align with the data fields you define within the records.
Example You have an inventory application with a master product file. Within the product
records, you have fields for vendor number, vendor's part code, and your part number, among others. You can define the following keys for the file regardless of the
placement of the fields.
Primary Key = vendor number + vendor part code
Alternate Key 1 =vendor number + your part number
Alternate Key 2 = vendor number
Alternate Key 3 = your part number

Cost

See the preceding considerations about the placement of keys within a
record, knowing that a key consists of all segments.

6.2.5 Key Characteristics
Key characteristics include:
• Duplicates
• Changes
•Null Key

Indexed File Design

6-7

Characteristics are restricted according to key number:
~

Characteristics

Primary(O)

Alternate(l +)

Duplicates

Yes or No

Changes

Yes or No

Null Key

Yes or No

The combination of changes and duplicates is also restricted by key number:
Combination
CHG

CHG

NO CHG

NOCHG
+

Key Type

DUP

NOD UP

DUP

NOD UP

Primary(O)

Error

Allowed

Alternate(l +)

Allowed

Error

Allowed

NOTE
PDP-11 COBOL allows the CHG+NODUP combination for Alternate Keys. To enable this option, the PDP-11 COBOL OTS
uses a hidden find operation to check on duplicates each time
an Alternate Key value changes on an update operation
(REWRITE in PDP-11 COBOL).
Duplicates - If duplicates are allowed in a key, more than one record can
have the same value in that key field.
Cost File Space
Duplicates have little affect on space usage as long as recorps are
not frequently updated with changing key values or deleted. If anything, records with duplicate key values are stored more efficiently
than nonduplicate values: fewer index records are required to cover
data records with duplicate Primary Keys. In Alternate indexes,
one SIDR with one representation of the key value is needed to
cover multiple data records with the same value in the key field.
However, this benefit becomes apparent only in large files with long
string keys.
Putting a Record
RMS-11 stores records with duplicate key values for first-in, firstout access. Putting (and updating) records containing duplicate key
values takes more time as the number of duplicates builds up.
A put operation can fail because duplicates are not allowed for one
of the keys. If this is the Primary Key, RMS-11 has wasted little
time since it has performed only the I/0 operations to find the
previous record with that value in the key field.

6-8

Indexed File Design

"'-"'

However, if you allowed no duplicates in one of the Alternate Keys,
RMS-11:
1. Updates the Primary index, including the data level.

2. Updates the preceding Alternate indexes.
3. Discovers that it cannot insert the record because a record already exists with that key value.
4. Reverses the actions it has taken, removing all updates from the
indexes it has already rewritten. Entries made in SIDR duplicate arrays are flagged as deleted and not compressed out of
existence. However, RMS-11 cannot reverse bucket splits.
5. Returns an error code.
Deleting a Record
If you do not allow duplicate values for the Primary Key, RMS-11
compresses a deleted record to a two-byte indicator. However, if you
allow duplicate values for the Primary Key, RMS-11 keeps enough
of the record to contain the entire Primary Key:
• If the format is fixed, the entire record remains in the file.
• If the format is variable, enough of a record remains in place to
hold the entire Primary Key.

'-"'

If you do not allow duplicate values for an Alternate Key, RMS-11
removes the SIDR when it deletes the data record. However, if
duplicates are allowed, the pointer remains in the SIDR array with
the delete flag set. RMS-11 never checks an array to see if all
records have been deleted.

Updating a Record
If you allow duplicate values on the Primary Key, the length of a
variable record cannot be changed during an update operation.
Also, updating records containing duplicate key values takes more
time as the number of duplicates builds up. Finally, the SIDR
pointers for deleted records are flagged as deleted, but not removed
from the duplicate array.
Summary
Duplicates are not costly for write-type operations unless there are
too many of them. Pick a key field that minimizes duplicates.
Example Fields where there are only two choices for entries, such as sex, are
definitely not good key fields.

Changes - The value of a Primary Key field cannot change during an update operation; however, you can allow the value in any Alternate Key field to
change if you are willing to allow duplicate values in that key.

During an update operation, RMS-11 checks the characteristics of all keys
and compares the new key values (in the record about to be rewritten) with

Indexed File Design

6-9

the old values: if you do not allow changes in a key field, but changes have
been made, RMS-11 immediately terminates the update operation with an
error code.

Cost If an Alternate Key value changes during an update operation, RMS-11
must trace the old SIDR and delete it, then insert the new one, starting
with the Root of the index for both processes. But if the data does not
change, RMS-11 does not touch the Alternate index.
Null Key - You can specify ·a null key value for any Alternate Key. If
RMS-11 finds that an Alternate Key field is filled with the null key value
specified for that key, it does not insert an entry into the index for the record
being written.
Zero (0) is the null key character for. the numeric key data types (integers,
binaries, and packed decimal). The null key character for string keys can be
any octal value (000 8 through 377 ~ including an ASCII character.

Cost The use of null key value can reduce the disk space an Alternate index
occupies, but it also precludes accessing those records not entered in
the index via that Alternate Key.

,.,~

6.3 Areas
You should divide an Indexed file into areas. An area is a portion of the file
that RMS-11 treats as an entity for:

~":

• Initial allocation
• Extensions
• Bucket size
• Placement on physical volume
Areas allow you to gather logical elements of the file into groups of continuous
ranges of Virtual Block Numbers (VBNs). These VBNs can be mapped onto a
contiguous set of logical blocks on disk. This tight sequence of VBNs is lost
when RMS-11 extends an area.

NOTE
Unless you completely allocate each area when you create the
Indexed file, the division of the file into areas does not improve
performance.
Areas can be set up for:
• Primary index Level 0 (the data records)
• Primary index Level 1 (the lowest index level)
• Primary index Levels 2 + (the rest of the index)
• Alternate index Level 0 (SIDRs)

6-10

Indexed File Design

;;~

• Alternate index Level 1 (the lowest index level)
• Alternate index Levels 2+ (the rest of the index)
Dividing a file into areas primarily saves I/0 time. As explained in Section
5.1, in a single-area file, RMS-11 intersperses index and data buckets: index
buckets are scattered among the data buckets. During each random record
access, RMS-11 consults the appropriate Index Descriptor in memory and
then directs (through the operating system) the disk head to read the Root
and Levels 2+, Level 1, then the appropriate Level 0 bucket. These buckets
can be anywhere in the file, and the disk head can travel large distances
several times to complete one access operation. Figure 6-1 shows an Indexed
file with one area.

Figure 6-1: Single-Area Indexed File

BUCKET
4

NUMBER

[D,_A_,T_A_P-~N-,~-~-;v-.---D-A~-A--.--D-A-TTI

·ROOT
OF
PRIMARY
INDEX

DATA

•• •

0-M K-00059-00

Example To randomly access a specific record in the single-area Indexed file illustrated in
Figure 6-2, RMS-11 makes the following 1/0 requests:
1. Read VBN 17933
2. Read VBN 305
3. Read VBN 14
4. Read VBN 20433

You can now realize how much the device has to move its read head to service one
ranqom access operation.

A multi-area file, on the other hand, can have all index buckets allocated
contiguously (if enough blocks were initially allocated): all index information
is available in one physical part of the disk. RMS-11 can then traverse an
index with no or little head movement until it reads the indicated data bucket. In addition, a sequential read of the file moves the head mechanism
smoothly through the physically contiguous area assigned to the Primary
index Level 0. Figure 6-3 shows an Indexed file with two areas.

Indexed File Design

6-11

Figure 6-2: Example of Single-Area Indexed File
ROO T
VBN

17933

~
LEVE L2

VBN 305

L
LEVE L 1

VBN 14

------

LEVE LO

VBN 20433

VBN =VIRTUAL

BLOCK NUMBER

H-MK-00057-00

Figure 6-3: Two-Area Indexed File
BUCKET
NUMBER

N-1

DATA

--~~---~~--~~---~~---..---1

ROOT
PRIMARY PRIMARY
OF
PRIMARY
INDEX
INDEX
PRIMARY
INDEX
INDEX

AREA 0

DATA

AREA 1

.._
I

O·MK-00056-00

To refine your file even more, place the lowest level of each index (Level 1) in
an area separate from the rest of the index (Levels 2+).
Example To randomly access a specific record in the multi-area Indexed file illustrated in
Figure 6-4, RMS-11 makes the following 1/0 requests:
1. Read VBN 418

2. Read VBN 423
3. Read VBN 1537
4. Read VBN 14703

You can now realize how much proper use of areas reduces disk head movement
during a random access operation.

6-12

Indexed File Design

·~

;

Figure 6-4: Example of Multi-Area Indexed File

VBN 418

ROOT

VBN 423

LEVEL 2

- - - - -i__::::~- - - VBN 1537

LEVEL 1

LEVEL 0

AREA1

AREA 2

VBN 14703

VBN = VIRTUAL BLOCK NUMBER

H-MK-00055-00

When you specify multiple areas, initially RMS-11 creates the file as Area 0.
After the file processor signals successful creation, RMS-11 explicitly extends
the file for each extra area specified. As a result, files with more than one area
cannot be created as contiguous files, though each area can be contiguous
within itself. However, this lack of contiguity may be in name only. Operating
systems usually try to allocate each extension of a file as near the end of that
file as possible. Therefore, if there are enough unused blocks after the end of
Area 0, all extra areas are allocated contiguously: the file comprises a consecutive series of logical blocks.
Contiguity is very important to performance. See further discussion in "Initial
Allocation," Section 6.6.1, and in "Virtual-to-Logical-Block Mapping," Section 8.3.

Indexed File Design

6-13

6.4 Placement Control
Placement control enables you to specify the location on a disk for a file or the
areas of a file. You use placement control for the following reasons:
• To start a file or area at the first block of a track or cylinder so that the file
or area can reside in one or contiguous tracks or cylinders. This effort minimizes head movement during file access.
• To place the files used by a single application together on a disk. This effort
reduces 1/0 time by minimizing head movement among the files.
Example You want to run a general ledger application that uses several files (an accounts
file, a transaction file, and so on). The application consists of several tasks. So,
you start with an initialized disk and copy the tasks onto it. Then, you create (and
populate) your data files, placing them near the tasks. This effort reduces the
distance the disk head moves to service the I/0 operations required by an RMS-11
program: disk-resident overlays (discussed in Chapter 8) and data file accesses.
Note, however, you gain more improvement if you eliminate head contention by
placing the individual files on separate disks.

You calculate track and cylinder starting block numbers as follows:
1. Read the documentation that came with your disk drive. Find and write
down the following numbers:

• Number of surfaces on a volume (or pack or disk)
• Number of tracks on a surface
• Number of sectors in a track

NOTE
On most DIGITAL disk drives, a sector equates to a
logical block.
Example The following decimal numbers apply to an RP06:
Number of cylinders per disk= 815
Number of tracks per cylinder = 19
Number of sectors per track = 22

2. Establish the starting Logical Block Number (LBN) for each track on the
disk by writing down the multiples of sectors-per-track. Since LBNs start
with 0, tracks start at multiples of track length.
Example From the RP06 specifications, the first 10 tracks start at LBNs:
0
110

22
132

44
154

176

198

3. Multiply sectors-per-track and tracks-per-cylinder to get sectors-per-cylinder. Establish the starting LBN for each cylinder on the disk by writing
down the multiples of sectors-per-cylinder.
Example For an RP06, the first 10 cylinders start at the following LBNs:
0
2090

6-14

Indexed File Design

418
2508

836

2926

1254
3344

1672
3762

NOTE
On RSTS/E, you specify Device Cluster Number (DCN) for
placement control. Therefore, you must make one more calculation in this procedure: divide the starting LBN by the device
clustersize to get the DCN containing that LBN. If you get a
nonzero remainder, use a DCN one higher than your result. See
the RSTS/E User's Guide for a table of device clustersizes.
After you decide where on the disk you want to place your file, you create the
file using RMS-11 placement control. In the process, you place Area 0 (which
may be the whole file, if you are not using multiple areas) at the location you
calculated.
Then, for Areas 1+, use VBN specifications to place them near the end of the
file as it exists at that time:
• Area 1 near the end of Area 0
• Area 2 near the end of Area 1
• and so on.
The VBNs specified are the running sum of the allocation quantities for the
areas.
Example You've divided an Indexed file as follows:

Primary data level in Area 0 (also includes the Prologue)
Primary index levels in Area 1
Alternate Key 1 data levels in Area 2
Alternate Key 1 index levels in Area 3
Using the following allocation quantities for those areas, you can calculate the last
VBN s for Areas 0 through 2, so you can start the next area near them:

Area

"-"'

Allocation
(in blocks)

Start near

15,467

NIA

15467

1,a42

15677

17019

VBN

RSTS/E automatically places Areas 1+ as near the end of Area 0 as possible.
If you are using a higher level language, you specify placement control via the
RMS DEF utility. If you are programming in MACR0-11, you specify placement control through the use of Allocation XABs.

Indexed File Design

6-15

6.5 Bucket Size
Buckets are the units of access for Indexed files. Bucket size is critical to the
virtual address space required by a task and to the speed with which it
performs. There is, of course, a trade-off: the larger a bucket, the larger the
task, but the faster it reads data:
• The speed of an RMS-11 operation is closely proportional to the number of
I/0 operations involved. For Indexed files, the number of data transfers
during a random retrieval operation is approximately equal to the depth of
the index (in most cases, one more than the depth). That number includes
only the I/O operations directly related to the record operation; other data
transfers can be required to service the operation, including overlays and
system overhead (discussed in Chapter 8).
Therefore, the larger the buckets, the shallower the index, and the faster the
random retrieval operation. Without other considerations, you should pick
the largest possible bucket. Your operating system limits bucket size as
follows:
·
Operating System
IAS

RSTS/E
RSX-11M

Maximum Bucket Size
32 blocks
15 blocks
32 blocks

• The larger the bucket, the more records fit in it, and sequential access can
require fewer I/0 operations.

However, there are other considerations. RMS-11 requires two I/0 buffers,
each the size of the largest bucket, when it opens an Indexed file. By making
bucket size smaller, you reduce the size of the buffers your task requires.
Depending on the record operations your program requires, that virtual address space may be better used in overlay structure optimization (discussed in
"Task Building with RMS-11 Routines," Section 8.1).
Therefore, you should set bucket size to some lower value that still allows
good performance; a reasonable goal is an index depth of three (Root at Level
3), though very large files can require four levels of index.
Each area can have its own bucket size, but normally you should use the
maximum size for all buckets:
• You should consider more than the size of your data record (plus the sevenbyte header) when you calculate Primary data bucket size:
-

Records that move from one bucket to another leave a seven-byte pointer.

Deleted records leave from two bytes to enough to hold the Primary Key
to the whole record.

Therefore, you should consider the predominant activity in the file:
-

6-16

If you intend to populate the file and then only read from it, you do not
consider activity overhead. You must populate the file with records in

Indexed File Design

.~.

ascending order by Primary Key value (discussed in "Population
Techniques," Section 6. 7).
If you intend to populate the file and then insert and/or delete a lot of
records, you should allow for those activities in your bucket size calculations.

• In choosing a bucket size, you should consider the file clustersize. Since file
clustersize governs file extension quantities and the way the RSTS/E file
processor handles disk read operations, sometimes you can improve performance by aligning buckets with clusters. RMS-11 makes this step more
reasonable by increasing the size of the file Prologue to an integral multiple
of bucket size when:

- you define the same bucket size for all areas
-

that bucket size is two, four, or eight blocks

6.5.1

Bucket Size for Primary Index

You can calculate bucket sizes with the following steps:
1. Calculate the following quantities for different bucket sizes (1, 2, 3, and so
on):

NIRBK = (512*BKS)-15
PKL + BPL

(Equation la)

and
NDRBK = ((512*BKS)-15) - AO
RSZ + RFO

..,

(Equation lb)

where:
NIRBK

is the number of index records per Level l+ index buckets

NDRBK

is the number of data records per Level 0 bucket

BKS

is the bucket size as number of blocks

PKL

is the Primary Key length in bytes

BPL

is the bucket pointer length:
BPL = 3 for pointers to the first 65,535 blocks in file
BPL = 4 for pointers to the blocks in file numbered between
65,536 and 2 24-1
BPL = 5 for pointers to the blocks in file numbered between
2 24 and 2 32-1

Indexed File Design

6-17

RSZ

is the size of the record:
• data size for fixed-length records
• average record length for variable-length records

RFO

is the record format overhead:
RFO = 7 bytes for fixed-length records
RFO = 9 bytes for variable-length records

is activity overhead:

where:
NRI is the number of data records you estimate will be inserted randomly over the life of the file
DO is the deletion overhead, the number of bytes not reusable in a bucket after RMS-11 deletes a record.

When you allow no duplicates for the Primary Key:

When you allow duplicate values for the Primary Key:
• For fixed-length records, DO = RSZ
~i

• For variable-length records,
DO = 2 + position of last byte of Primary Key + 1
ND is the number of data records you estimate will be deleted over the life of the file.
If you cannot predict values for the components of AO, use

a value at least 5% of bucket size.
2. Select bucket size for data and index areas where the following equation is
true:
NIRBK**n 2: NRF/NDRBK

(Equation 2)

where:
NRF

is the number of data records in the file

NRF/NDRBK is the number of Primary index data buckets
n

is the depth of the index.

This equation portrays the exponential relationship between the number
of data records in a file and the depth of its index. You use the values for
NIRBK and NDRBK you calculated in step 1.
a. Set up a grid (see the example after step 2e).

6-18

Indexed File Design

b. For each value of NIRBK, calculate the left side of Equation 2, for n =
2, 3, and for very large files, 4.
c. For each value of NDRBK, calculate the right side of Equation 2.
d. Where the equation is true, that is, the left side is greater than or equal
to the right side, you have a valid combination of bucket sizes. The
bucket size used to calculate the left side may be equal to the size used
to calculate the right side, but it does not have to be.

NOTE
You gain no advantage using different index and data
bucket sizes. RMS-11 requires two I/0 buffers, both the
size of the largest bucket defined for the file.
In fact, PDP-11 COBOL users must not choose different
index and data bucket sizes.
e. Select one of the valid combinations according to your application's
requirements.

NOTE
• Equation 2 is true only for files where records are inserted in order by ascending value of the Primary Key.
See also "Population Techniques," Section 6.7, in this
chapter.
• Bucket size is a step function of index depth. Therefore intermediate bucket sizes generally waste address
space.
Example Given a file in where:

BKS = 4 BKS = 8 -

index depth of 3
index depth of 2

then bucket sizes of 5, 6, and 7 blocks would not normally be used.
Example A file of 50,000 200-byte fixed-length records with a 15-byte Primary Key. We use
the equations in steps 1 and 2 to fill in the following grid:
1. Calculate values for Number oflndex Records per BucKet (NIRBK) using Equa-

tion la and bucket sizes 1 through 6. Drop the remainder; use only the integer
part of the result.
2. Calculate values for Number of Data Records per BucKet (NDRBK) using
Equation lb and bucket sizes 1 through 6. Drop the remainder; use only the
integer part of the result.
3. Calculate the number of data buckets in Level 0 (NRF/NDRBK) corresponding
to bucket sizes 1 through 6. Round the result up to the nearest integer.
4. Calculate NIRBK squared (NIRBK**2) for the values of NIRBK corresponding
to bucket sizes 1 through 6. Round the result up to the nearest integer.

Indexed File Design

6-19

5. Calculate NIRBK cubed (NIRBK**3) for the values of NIRBK corresponding to
bucket sizes 1 through 6. Round the result up to the nearest integer.
BKS

NIRBK

112

141

169

NDRBK

NRF/NDRBK 25000

12500

7143

5556

4167

3572

NIRBK**3

19683

175616

592704

1014049

2803220

4826810

NIRBK**2

729

3136

7056

12544'

19881

28561

Now we determine the combinations of bucket sizes where Equation 2 is true:
6. Compare the values in the NIRBK**3 row one at a time to each of the values in
the NRF/NDRBK row. Where the NIRBK**3 value is greater than or equal to
NRF/NDRBK, we have a valid bucket size combination.
Example The first NIRBK**3 value is 19683. This is less than 25000, the first
NRF/NDRBK value, but it is greater than 12500, the second
NRF/NDRBK value. Therefore, index bucket size of 1 (from
NIRBK**3 row) and data bucket size of 2 (from the NRF/NDRBK
row) is a valid combination.

7. Compare the values in the NIRBK**2 row one at a time to each of the values in
the NRF/NDRBK row. Where the NIRBK**2 value is greater than or equal to
NRF/NDRBK, we have a valid bucket size combination.
Example The first NIRBK**2 value is 729. This is too small to use, as is the
second value in the row. However, the third value is 7056. This is less
than 25000 (the first NRF/NDRBK value) as well as the next two
values, but greater than 5556, the fourth NRF/NDRBK value. Therefore, index bucket size of 3 (from NIRBK**2 row) and data bucket
size of 4 (from the NRF/NDRBK row) is a valid combination.

As a result of these comparisons, Equation 2 was true in the following cases:
NIRBK**3

NIRBK**2

DBKS

IBKS

IOB (bytes)

DBKS

IBKS

IOB (bytes)

1
2

2
1

2048
2048

6
4
3

6144
4096
4096

where:

6-20

DBKS

is the data bucket size from the NRF/NDRBK row

IBKS

is the index bucket size from the NIRBK**n rows

Indexed File Design

IOB

is the maximum J/0 buffer space required by the largest bucket size of the
pair

The choice of bucket size pairs depends on what you need to optimize most in the
application: task size or access time. After you choose, make data and index bucket
sizes equal to the larger size selected.

6.5.2 BucketSlzes for Alternate Indexes
The selection of bucket sizes for Alternate indexes follows the same procedure
as that of Primary Key bucket sizes:
1.

The records-per-bucket equations for Alternate indexes are:
NIRBK _ (512*BKS)-15
- AKL + BPL

'-"'

and
NDRBK =

((512*BKS)-15)*DF
AKL + (DBPL*DF) + 4 + DO

where:
AKL

is the Alternate Key length in bytes

DBPL

is the data bucket pointer length where no duplicates are
allowed:
DBPL = 4 for pointers to the first 65,535 blocks in file
DBPL = 5 for pointers to the blocks in file numbered between
65,536 and 2 24-1
DBPL = 6 for pointers to the blocks in file numbered between
2 24 and 2 32-1

is the duplicate factor:
DF = 1 if you allow no duplicates
DF = average number of records with same key values for
any key value present in the file

NOTE
The DF factor does not compensate enough if DF
is greater than the number of data records that
fit in a bucket. RMS-11 must then use Continuation Buckets to store the records with duplicate
values.

Indexed File Design

6-21

is the duplicate overhead:
DO = 0 if you allow no duplicates
DO = 4 if you allow duplicates

No record movement or space/deletion overhead occurs in index buckets.
2. RMS-11 cannot load buckets in Alternate indexes as efficiently as in the
Primary index because Alternate Key values inevitably fall in random
order (unless you use the RMSIFL utility described in Chapter 9). The
ideal values resulting from the equations in "Bucket Sizes for Primary
Index," Section 6.5.1, must be reduced by a packing efficiency factor.
Studies have shown that the factor for Alternate Keys is normally about
0.5. However, this factor applies only to the lower levels of the index, and
not to the Root. The packing efficiency of any index's Root is always one.
Therefore, the index depth equation for Alternate indexes is:
(0.5**n)*(NIRBK**n) ~ NRF/NDRBK
Example Using the file in the Primary Key example and adding a ten-byte First Alternate
Key, allowing no duplicates, we fill in the following grid (NRF = 50,000 since there is
one SIDR per data record):
BKS

NIRBK

117

156

195

235

NDRBK

113

142

170

NRF/NDRBK

1811

892

592

443

354

295

0.125*NIRBK**3

6859

57067

200202

474552

926860

1622240

0.250*NIRBK**2

361

1483

3423

6084

9507

13807

The index depth equation for Alternate indexes is true in the following cases:
NIRBK**3

NIRBK**2

DBKS

IBKS

IOB (bytes)

DBKS

IBKS

IOB (bytes)

1024

1
2

3
2

3072
2048

Do not choose a bucket size smaller than that selected for the Primary index (Section 6.5.1).

6.5.3 Program Syntax
RMS-11 requires bucket size as a whole number of blocks. However, some
higher level language compilers require or allow you to specify the bucket size
in number of records. This syntax can lead to a different number of records
per bucket than you are counting on.

6-22

Indexed File Design

~•,~,

Example A BASIC-PLUS-2 program contains the following clause in an OPEN statement
that creates an Indexed file:

BUCKETSIZE 5%
The record format is fixed; record length is 100 bytes. The compiler makes
the following calculation:
100 bytes for the data
+ 7 bytes for the record header
107 bytes for each record
x 5 records specified in a bucket
535 bytes for the records in a bucket
+ 15 bytes for the bucket overhead
550 bytes required to be in the bucket
A bucket must be a whole number of blocks long, so the compiler rounds the bucket
size to two blocks and passes that to RMS-11 to create the file.
However, two blocks contain 1024 bytes; that leaves 1009 bytes for record storage
after the bucket overhead is subtracted.
1009/107 = 9 records per data bucket
Therefore, the buckets which were originally supposed to contain only five records
now can contain nine.

Bucket size is set by RMSDEF or by your application program:
MACR0-11
• Single-area files
Use the initialization macro or $STORE to set the BKS field in the FAB
of the file to the chosen number of blocks before issuing $CREATE.
• Multi-area files
Use the initialization macro or $STORE to set the BKZ field in the
Allocation XAB for each area to the chosen number of blocks before
issuing $CREATE.
NOTE: The default value is one block per bucket.
BASIC-PLUS-2
Use the BUCKETSIZE clause in the OPEN statement that creates the
file.
PDP-11 COBOL
In the file-description-entry (FD), use the BLOCK CONTAINS clause.
RPG II
Use the RPGASN utility to override the default value set by the compiler.
DIBOL
Use RMSDEF; DIBOL does not create Indexed file.

Indexed File Design

6-23

6. 6 Allocatlon
RMS-11 requests the file processor to allocate blocks to a file at three different points in the file's life:
• when the file is created
• when RMS-11 must dynamically extend the file to complete an operation
• when you explicitly instruct RMS-11 to extend the file
The allocation of blocks to a file takes time, mainly I/O time as the operating
system performs its function. If RMS-11 has to request an allocation every
time it requires a new bucket, this time can be a significant factor in an
application's performance, especially during file population.
You can help optimize performance by minimizing allocation overhead in the
following areas:
• Initial allocation
• Default extension quantity

6.6.1

lnltlal Allocation

Total allocation of an Indexed file when you create it is most efficient.
The total allocation for a file is the sum of the Prologue and the allocations for
the different indexes that make up the file; an index's allocation is the sum of
the allocations for all levels in the index. You should start with the Primary
Level 0 and "build" each level of each index on paper, as shown in the
following steps.
1. Calculate the number of buckets in Level 0:
NBK@O = NRF/NDRBK
where:
NRF

is the total number of records in the file

NDRBK is the number of data records in a bucket in Level 0 (see
"Bucket Size for Primary Index," Section 6.5.1 for the method
of determining this value), and
NBK@O

is the number of buckets in Level 0

6-24

Indexed File Design

2. Calculate the number of buckets in Level 1:
NBK@l = NBK@O/NIRBK
where:
NBK@l

is the number of buckets in Level 1, and

NIRBK

is the number of index records per bucket in the index (see
"Bucket Size for Primary Index," Section 6.5.1).

NOTE
The method described in Section 6.5.1 assumes that you will
put records into the file in order by ascending Primary Key
value. However, if you will be loading the file in random Primary Key value order, you should divide the Section 6.5.1
NIRBK value by two for every index level but the Root. You
will need twice as many index buckets.
3. Calculate the number of buckets in Level 2:
NBK@2 = NBK@l/NIRBK
4. Continue this sequence of calculations until you reach the Root level, that

is:

NBK@n = 1 = NBK@(n-1)/NIRBK
where:

NBK@n is the number of buckets in the Root, which is 1 by definition,
n

is the index depth.

5. Calculate the allocation in blocks for each level:
AQ@O = NBK@O * DBKS
AQ@l = NBK@l * IBKS
AQ@n = IBKS
where:
AQ@O

is the allocation quantity in blocks for Level 0

DBKS

is the data bucket size in blocks

IBKS

is the index bucket size

6. Calculate the allocation for each Alternate index as shown in steps 1
through 5; see "Bucket Sizes for Alternate Indexes," Section 6.5.2, for
equations.

Indexed File Design

6-25

NOTE
Alternate indexes are normally populated in random key value
order. Therefore, you should divide the Section 6.5.2 NDRBK
and NKDBK values by two except for the Root level.
7. The total allocation quantity for the file (ALQ) is the sum of the index
allocation quantities plus the Prologue:
ALQ = PLG + AQPK + AQAKl + ... + AQAKn
where:
is the last Alternate Key defined for the file.

NOTE
Alternate indexes are normally populated in random key
value order. Therefore, you should divide the Section 6.5.2
NDRBK and NIRBK values by two except for the Root
level.

The Prologue of an Indexed file can be from two to 84 blocks long. The size is
the sum of the Key Descriptor blocks and the Area Descriptor blocks:
• VBN 1 describes the Primary Key (and contains other attribute information).
• Each Key Descriptor block covers up to five Alternate Keys.
• Each Area Descriptor block covers up to eight areas.
Finally, RMS-11 extends the Prologue to an integral multiple of bucket size if
the criteria described in Section 6.5 are met.
Example Given an Indexed file of 100,000 fixed-length user data records with the following
attributes, calculate a reasonable initial allocation size in blocks:

Data size = 200 bytes
Primary Key = 20-byte string; no duplicates allowed
Alternate Key = 8--byte packed decimal; no duplicates allowed
Data bucket size = Index bucket size = 3 blocks
Calculate the Primary Index first:
1.

AO= 0, so
NDRBK = ((512*3)-15)/(200 + 7) = 7 data records per bucket
NBK@O = NRF/NDRBK = 100000/7 = 14,286 buckets in Level 0

6-26

Indexed File Design

2. NIRBK = ((512*3)-15)/(20 + 3) = 66 index records per bucket
NBK@l = NBK@O/NIRBK = 14286/66 = 217 buckets in Level 1
3. NBK@2 = NBK@l/NIRBK = 217/66 = 4 buckets in Level 2
NOTE
If the number of buckets in the level under the Root is very
much less than the number of index records that fit in a bucket, you may be able to use a smaller bucket size without increasing the index depth.

4. NBK@3 = NBK@2/NIRBK = 4/66 = 1 bucket in Level 3, the Root
5. AQ@O = NBK®O*DBKS = 14286*3 = 42,858 blocks in Level 0
AQ@l = NBK@1*IBKS = 217*3 =
648 blocks in Level 1
AQ@2 = NBK®2*IBKS = 4*3
12 blocks in Level 2
AQ@3 = NBK®3*IBKS = 1*3
3 blocks in Level 3
AQPK = 43,521 blocks iin the Primary index
Now calculate the Alternate index:
1. NDRBK = ((512*3)-15)/(8 + (4*1) + 4)
=

89 data records per bucket

NBK@O = NRF/NDRBK = 100000/89 = 1124*2 = 2,248 buckets in Level 0
The doubling compensates for a packing efficiency of 0.5.
2. NIRBK = ((512*3)-15)/(8 + 3) = 138 index records per bucket
NBK@l = NBK@O/NIRBK = 17 *2 = 34 buckets in Level 1
3. NBK@2 = NBK@l/NIRBK = 1 bucket in Level 2, the Root
4. AQ@O = NBK®O*BKS = 2248*3 = 6,744 blocks in Level 0
AQ@l = NBK®l*BKS = 34*3 =
102 blocks in Level 1
AQ@2 = NBK@2*IBKS = 1*3 =
3 blocks in Level 2
AQAK = 6,849 blocks in Alternate index
Finally:
ALQ = PLG + AQPK + M~AKl = 3 + 43,518 + 6,849
= 50,370 blocks for the whole file

This allocation can be done by the RMSDEF utility or by your application
program as follows:
MACR0-11
• Single-area files
Use the initialization macro or $STORE to set the ALQ field in the FAB
of the file to the chosen number of blocks before issuing $CREATE.
• Multi-area files
Use the initialization macro or $STORE to set the ALQ field in the
Allocation XAB for each area to the chosen number of blocks before
issuing $CREATE.

Indexed File Design

6-27

BASIC-PLUS-2
Use the FILESIZE clause in the OPEN statement that creates the file.
PDP-11 COBOL
Use the /AL:n switch on the file specification in the ASSIGN clause or the
VALUE OF ID.
RPG II
Use the RPGASN utility to override the default value set by the compiler.
DIBOL
Use RMSDEF; DIBOL does not create Indexed files.

NOTE
If possible, you should allocate an Indexed file contiguously. To
ensure that the entire file is logically contiguous, allocate it all
when you create it and then copy it (if you used multiple areas)
into a contiguous file with PIP.

On IAS/RSX-llM, you can achieve the same result by:
• calculating the starting LBNs for each area (see "Placement
Control," Section 6.4)
• requiring the operating system to place those areas as directed. The RMSDEF utility enables you to specify starting
LBNs and that you want the creation to fail if the system
cannot allocate the blocks in the places you indicated.
See "Virtual-to-Logical-Block Mapping," Section 8.3, for the
impact of extending a contiguous file on RSTS/E.
MACR0-11
Use the initialization macro or $SET to set the FOP field in
the FAB of the file to include FB$CTG before issuing
$CREA TE for a single-area file. However, if you are using
an Allocation XAB to place one area, you set the XB$CTG
value in the AOP field of the Allocation XAB.
BASIC-PLUS-2
Specify CONTIGUOUS in the OPEN statement that creates the file.
PDP-11 COBOL
Use the /CO switch on the File ID.
RPG II
Use RMSDEF; RPG II does not create contiguous files.
DIBOL
Use RMSDEF; DIBOL does not create Indexed files.

6-28

Indexed File Design

6.6.2 Default Extension Quantity
If the file cannot be totally allocated at creation, you should establish a

reasonable Default Extension Quantity (DEQ) to minimize the number of
(and the time spent on) file extensions. Even if the file is totally allocated
when it is created, you should establish a reasonable DEQ in case the file gets
bigger than planned.
A good basis for calculation is the number of records that are added to the file
in a given period of time, such as a day; use the formula for allocation quantity in Section 6.6.1.
The DEQ should equal a multiple of the bucket size.
If you do not specify a DEQ, it defaults to zero whether you create the file

with RMSDEF or a higher level language. RMS-11 responds to a DEQ of zero
by requesting four times bucket size in blocks from the file processor each
time it automatically extends the file.
The DEQ for the file can be set by the RMSDEF utility or by your application
program as follows:
MACR0-11
• Single-area files
Use the initialization macro or $STORE to set the DEQ field in the F AB
of the file to the chosen number of blocks before issuing $CREATE.
• Multi-area files
Use the initialization macro or $STORE to set the DEQ field in the
Allocation XAB for each area to the chosen number of blocks before
issuing $CREATE.
The DEQ field in the FAB serves as run-time file DEQ; that is, RMS-11
uses it for any area whose DEQ is zero.
BASIC-PLUS-2
Use RMSDEF; BASIC-PLUS-2 does not support DEQ specifications.
PDP-11 COBOL
Use the /EX:n switch on the file specification in the ASSIGN clause or the
VALUE OF ID.
RPG II
Use the RPGASN utility to override the default value set by the compiler.
DIBOL
Use RMSDEF; DIBOL does not create Indexed files.

6.7 Population Techniques
File population entails a large burst of records written into the file after it has
been created and before it is made available for normal processing. You can

Indexed File Design

6-29

populate a file with RMSIFL, RMSCNV, or an application task. The aim of
populating an RMS-11 Indexed file is to avoid bucket splits and record movement during the population and during later use of the file. The techniques to
achieve this goal are:
• ascending order by Primary Key
• fill number

6. 7 .1 Ascending Order by· Primary Key
The best way to populate an indexed file is to insert the records in ascending
Primacy Key value order. You do not need to insert the records all at once.
This technique:
• minimizes population time
• avoids the creation of RRV records, allowing RMS-11 to fill buckets with
data records and thereby find records with the least access time.
Contrast this technique with records loaded in descending order by Primary
Key value. In that case, you introduce the packing efficiency factor p to the
Primary Key equations. Normally, pis 1, when you insert records in ascending order and the factor drops out of the equation, as shown here:
NIRBK**n~ NRF/NDRBK

But when p < 1, the equation becomes:
(p**n)(NIRBK**n) ;;::: NRF/NDRBK
Since p is a fraction, the introduction of this factor reduces the left side of the
equation, at times dramatically, thereby potentially increasing:
• the index depth needed to cover a specific number of data records
• frequency of bucket splitting (an important factor in the time required to
populate an Indexed file)
As mentioned in "Bucket Sizes for Alternate Indexes," Section 6.5.2, Alternate indexes are a prime example of packing inefficiency, a situation avoided
only with the RMSIFL utility. The best general approximation for p in the
case of Alternate indexes is 0.5, the value used in Section 6.5.2.
You can populate a file with records in ascending order by Primary Key as
follows:
• Use the RMSIFL utility. The utility:
-

sorts your input file into ascending order by the output file's Primary
Key, if it is not already sorted that way

transfers the records from the input file to the output file

RMSIFL uses techniques not available to you to further improve the population of an Indexed file.

6-30

Indexed File Design

• Use the RMSCNV utility, specifying the Mass Insert (MA) switch.

''-"'

• Write a MACR0-11 program to populate the file and specify:
- in the FAB, Deferred Write (FB$DFW in FOP field) when you open the
file
- in the RAB when you connect to the file:
• Mass Insert (RB$MAS in ROP field)
• Sequential Access Mode (RB$SEQ in RAC field)
Be sure to sort your input records into ascending order by the Indexed file's
Primary Key before you run the program.

6.7.2 Random Insertions after File Population
If you will be inserting records into an Indexed file after it is populated, you
should consider ways to optimize these operations:
• If the inserted records have the full range of Primary Key values, you should
use fill numbers.
• If the inserted records are sorted into ascending Primary Key value and
added at the logical end-of-file, use Mass Insert.
\
6.7.2.1 Fiii Number - You can optimize for evenly distributed random insertions by leaving free space in buckets during the initial population of the file.
To do this, you specify a fill number as a set amount of bytes for each area in
your file. Normally, RMS-11 ignores this number, but you can direct RMS-11
to obey it: RMS-11 then fills each bucket in the file to the level specified by
the number.
Example Your bucket size is two blocks; you set the fill number to 768 bytes. When you tell
RMS-11 to obey the fill number, it only uses 768 out of 1024 bytes in each
bucket-the buckets are logically three-quarters size.

You use the fill number when you populate a file to improve its performance
during normal operations: if free space is available in every bucket in the file,
any record randomly inserted into the file is likely to fit without causing a
bucket split.
The size of a fill number depends on:
The amount of insertion activity you expect
Allow room (including record header) for the number of records you will
add to each bucket during normal operations. Occasional inserts might
not warrant the use of fill numbers, whereas heavy insertion can require
room for multiple additional records in each bucket to optimize, but not
eliminate, bucket splitting activity.
The type of bucket (data or index) involved
Because of the difference in record sizes and frequency of insertion, data
and index buckets should normally have different fill numbers.

Indexed File Design

6-31

Example The file contains 240-byte fixed-length records with a Primary Key field 24
bytes long. To optimize random insertions, the fill number for data buckets
should therefore be at most:
bucket-length
less bucket overhead (15)
less 247
This number leaves room for one data record.
This same fill number for index buckets leaves room for nine index records. A
more reasonable fill number for index buckets is:
bucket-length
less bucket over head
less 2 times 27 bytes
This number leaves room for two index records, where:
Index record = Primary Key Length + Bucket Pointer Length
= 24 + 3 = 27

See "Bucket Sizes for Alternate Indexes," Section 6.5.2, for a more complete discussion.
NOTE
RMS-11 ignores a fill number less than 50% of the bucket
length and uses the 50% figure.
The fill number for a file can be set by the RMSDEF utility or by your
application program as follows:
MACR0-11
Use the initialization macro or $STORE to set the DFL and IFL fields in
the key XABs to the chosen fill number before issuing $CREATE.
BASIC-PLUS-2
BASIC-PLUS-2 does not support fill number specifications.
PDP-11 COBOL
PDP-11 COBOL does not support fill number specifications.
DIBOL
DIBOL does not support fill number specifications.

RPG II
RPG II does not support fill number specifications.
You use fill numbers when you populate a file by:
• the RMSCNV or RMSIFL utility with the /LO switch specified.
• a MACR0-11 program where the value RB$LOA is set into the ROP field of
the active Record Access Block.
Most higher level language tasks cannot cause RMS-11 to obey fill numbers.

6-32

Indexed File Design

You use Mass Insert when you have a series of records
to add to an Indexed file and:
6. 7.2.2 Mass Insert -

• You have sorted the records into ascending order by the file's Primary Key.
• The lowest key value in the records is greater than the highest key value in
the file; that is, the records will be inserted at the logical end-of-file.
While the Mass bit is on, RMS-11 performs a put operation normally (see
"Putting a Record," Section 5.3.1) except that it:
• does not unlock the Primary Level 0 data bucket
• keeps a pointer to the Primary Level 1 bucket that pointed to the proper
Level 0 bucket
These extra steps enable RMS--11 to:
• write the next record without following the Primary index (if the Mass bit is
still on)
• rapidly split the Primary Level 0 bucket when it is full: since RMS-11 has a
pointer to the Primary Level 1 bucket that will contain the index record for
the new bucket, it can update that bucket without following the index.
By using these techniques, RMS-11 can extend the Primary Level 0 bucket by
bucket, packing records into the buckets in the order they are written. As each
bucket gets full, RMS-11 creates a new one, beginning with the next record
inserted, and notes its existence in the Primary Level 1 index bucket.

NOTE
Mass Insert significantly improves performance for single-key
Indexed files. The percentage of improvement lessens with each
additional key defined in the file.
You can enhance Mass Insert performance by using Deferred Write (see "1/0
Techniques," Section 7 .1.3).

Indexed File Design

6-33

Chapter 7
Indexed Task Design

The record and file processing capabilities described in Chapter 1 are available for Indexed files. This chapter discusses the operations and their implementation and restrictions with Indexed files.

7.1 Record Operations
RMS-11 performs a record operation at the request of a program. See also the
discussions of read- and write-type record operations in Chapter 5. The available operations include:
Connect
Delete
Disconnect
Find
Flush
Get
Put
Rewind
Update
In all record operations, RMS-11 establishes Current Record (if any) and
Next Record (if applicable). If any record operation fails, RMS-11 normally
sets Current Record to NONE and does not change Next Record. "Record
Access Stream," Section 1.3.1.5, introduces the concepts of Current Record
and Next Record.

7.1.1

Connect

A connect operation affects the context for the Record Access Stream as
follows:
Current Record
There is no Current Record. Any operation requiring Current Record fails
at this point.

7-1

Next Record
The Next Record is the first record in the file according to the collating
sequence of the specified key of reference.
Example In an Indexed file with multiple keys, the Next Record varies by the key specified in the instruction initiating the connect operation:
• If the Primary Key is specified, the Next Record is the first record in Primary

Level 0, the one with the lowest Primary Key value in the file.
• If an Alternate Key is specified, the Next Record is indicated by the first

SIDR in the Alternate index's Level O; the record itself can be located anywhere in the Primary Level 0.

7.1.2 Delete
In a delete operation, RMS-11 flags the header of the Current Record to
indicate that it is a deleted record. The prerequisite get ~r find operation
brought the bucket containing the record into the 1/0 buffer.
Then, RMS-11 writes the bucket over its original location on the disk, unless
you specified Deferred Write (discussed in "1/0 Techniques," Section 7.2).
A delete operation affects the context for the Record Access Stream as follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record.
Unchanged.

7.1.3 Disconnect
A disconnect operation destroys the context for the Record Access Stream.
You cannot resume this context by reconnecting the stream.

7.1.4 Find
"Getting and/or Finding a Record," Section 5.3.2, describes how RMS-11
performs a find operation.
If the record does not exist or has been deleted, RMS-11 returns an error code
depending on the access mode:

- In Sequential Access Mode, the error code is ER$EOF.
- In Random Access Mode, the error code is ER$RNF.
- In Access by RFA, the error code is:

ER$RFA
No valid record has ever existed at the specified location.
ER$DEL
The record header indicates that the record was deleted.

7-2

Indexed Task Design

A find operation affects the context for the Record Access Stream as follows:
• find in Sequential Access Mode:
Current Record
Set to value of the record found.
Example You've connected a stream to an Indexed file, specifying 0 as the key of
reference. There is no Current Record, but the Next Record is the first record
in Primary Level 0. If you execute a sequential find operation, the Current
Record is set to this record.

Next Record
Set to record logically following Current Record in index of reference.
NOTE
RMS-11 enacts this logical sequence only when it actually
accesses the Next Record:
1. RMS-11 locates the Current Record, reading a bucket if
necessary.
2. RMS-11 locates the record logically following the Current
Record, reading another bucket if necessary.
If the Indexed file is shared, the actual record in the Next
Record position can change between the operation that
accesses the Current Hecord and the one that finds the Next
Record.
Example From the previous example, Next Record is the record in the file with the
next higher Primary Key value.

• find in Random Access Mode or Access by RF A:
Current Record
Set to the record found, that is, the record identified by the Record's
File Address.
Next Record
Unchanged.
Example In the previous examples, you did a sequential find after connecting the
stream to the file. You now execute a find by RF A. The Current Record is set
to the record specified, but the Next Record is not changed. Therefore, if you
do another sequential find, Current Record will be set to the second record in
Primary Level 0, not the record following the one found by RF A.

You use find instead of a get operation because:
o it is quicker because the record is not moved to the user buffer. Although the

time required to move a record from one part of memory to another is very
short, do not expend it if you do not need to.
ct

it does not change Next Record in Random Access Mode or Access by RF A.
This convention allows you to branch off sequential processing for purpose
of updating or deleting, and yet keep your place.

Indexed Task Design

7-3

You can use a find operation in the following ways:
• To skip records in Sequential Access Mode by initiating successive find
operations.
• To establish a Current Record for a delete or update operation.
• To determine the existence of a record in Random Access Mode.

7.1.5 Flush
"Records Operations," Section 1.2.4, summarizes the flush operation.
A flush operation does not affect the context for the Record Access Stream.

7.1.6 Get
"Getting and/or Finding a Record," Section 5.3.2, describes how RMS-11
performs a get operation.
If the record does not exist or has been deleted, RMS-11 returns an error code

depending on the access mode:

- In Sequential Access Mode, the error code is ER$EOF.

- In Random Access Mode, the error code is ER$RNF.
- In Access by RFA, the error code is:
ER$RFA
No valid record has ever existed at the specified location.
ER$DEL
The record header indicates that the record was deleted.
A get operation affects the context for the Record Access Stream as follows:
• get in Sequential Access Mode not immediately preceded by a successful
find operation:
Current Record
Set to value of the record read. See "Find in Sequential Access Mode"
for example.
Next Record
Set to record logically following Current Record in index of reference.
See note and example in "Find in Sequential Access Mode."
• get in Sequential Access Mode immediately preceded by a successful find
operation:
Current Record
Unchanged (from Current Record set by find operation).

7-4

Indexed Task Design

Next Record
Set to record logically following Current Record in index of reference
(possibly changing Next Hecord set by find operation).

NOTE
A find in Access by RF A changes the key of reference to
the Primary Key. But the change does not become apparent unless you follow the find by a sequential get operation. In this case, the Current Context for the stream is
affected as follows:
Current Record
Unchanged from Current Record set by find operation.
Next Record
Set to record logically following Current Record in the
Primary Key sequence.
A get by RFA has the same affect.
• get in Random Access Mode
Current Record
Set to record specified, that is, the record read.
Next Record
Set to record logically following Current Record in index of reference.
• get in Access by RF A:
Current Record
Set to record specified by Record's File Address.
Next Record
Set to record logically following Current Record in the Primary index.
This differs from find by RFA which does not change Next Record.

'1.1.7 Put
"Putting a Record," Section 5.3.1 describes how RMS-11 performs a put
operation.
A put operation affects the context for the Record Access Stream as follows:
• put in Sequential Access Mode:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Undefined. The record retrieved by a sequential find or get at this point
is not specified.

Indexed Task Design

7-5

• put in Random Access Mode:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Unchanged.

7 .1.8 Rewind
A Rewind operation sets the context of the Record Access Stream to a logical
beginning of the Indexed file. In doing so, the operation affects the context for
the stream as follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Set to first record in file according to the specified key of reference.

7 .1.9 Update
In an update operation, RMS-11 moves the specified record from the task's
user buffer to the I/0 buffer, replacing the Current Record set by the prerequisite get or find operation. Then, RMS-11 writes the bucket over its original
location on the disk, unless you specified Deferred Write (described in "I/O
Techniques," Section 7.2). "Updating A Record, Section 5.3.3," describes the
update operation in detail.
An update operation requires a valid Current Record. Therefore an update
should follow a successful get or find operation; otherwise, RMS-11 returns
the error code ER$CUR. This error does not affect the original record in the
file on disk.
An update operation affects the context for the Record Access Stream as
follows:
Current Record
None. Any operation requiring a Current Record fails at this point.
Next Record
Unchanged.

7.2 Record Transfer Modes
You can manipulate records either in the I/0 buffer or in your program's user
buffer (see Figure 7-1). Each of these options is called a Record Transfer
Mode. You can change Record Transfer Mode at run time, even between
record operations.

7-6

Indexed Task Design

Figure 7-1:

RMS-11 Task Structure
j81zE DEPENDS ON:- -

;

I• NUMBER OF FILES OPENED SIMULTANEOUSLY :

I • BUCKET SIZES
~ NUMBE~~RECrD~CCESS STREAMS _ _

U ER BU FFERS

'
1/0
BUFFERS

VIRTUAL
MEMORY

RMS-11

PROGRAM

INTERNAL
CONTROL
STRUCTURES

fSizEDEPENDS ON:-

j_---:

I• RMS-11 FUNCTIONS USED I
~ ~VERLA0TRUCTURE~~E~

H-M K-00069-00

7,.2.1

Move Mode

Move Mode is the default Record Transfer Mode for all programming
languages on all file organizations.
• On get operations, RMS-11 moves the record from the 1/0 buffer to the user
buffer before returning control to your program.
• On put and update operations, your program assembles the record to be
written into the file in the user buffer, and during the operation, RMS-11
moves the data into the I/O buffer before updating the file.
Move Mode is the default Record Transfer Mode for all programming languages on all file organizations.

7.2.2 Locate Mode
Locate Mode enables your program to manipulate records in the 1/0 buffer,
eliminating the data transfers between it and the user buffer. However, when
you specify Locate Mode, RMS-11 uses it only when such usage does not
compromise data integrity. Otherwise, RMS-11 uses Move Mode. Therefore,
your program must still contain a user buffer.
Example RMS-11 uses Move Mode instead of Locate Mode when an Indexed file is shared.
Example RMS-11 uses Move Mode instead of Locate Mode if you opened the file indicating
that you were going to perform update operations on it.

Indexed Task Design

7-7

RMS-ll's use of Move Mode instead of Locate Mode is transparent to your
program as long as you use RMS-11 facilities to access the record data.
For Indexed files, your program can only get records in Locate Mode. See your
language documentation to determine if the language supports Locate Mode
and if it does, what the exact programming techniques are.

7.3 1/0 Techniques
You can use the following techniques to improve the performance of record
operations.

7 .3.1

IAS/RSX-11 M Asynchronous Record Operations

Within each Record Access Stream, your program can perform any record
operation either synchronously or asynchronously. In synchronous operations,
RMS-11 returns control to your program after the operation ends, either
successfully or with an error.
When you execute an asynchronous operation, RMS-11 may return control to
your program before the operation is complete. The program continues processing while the physical transfer of data between disk and memory is carried
out. However, you must not initiate another record operation on that stream
until the first operation ends; otherwise, RMS-11 returns the error code
ER$ACT. See your language documentation for asynchronous techniques.

NOTE
If you intend to use asynchronous RMS-11 record operations
and/or Asynchronous System Traps (ASTs) in other parts of
your program, see the section on your operating system in
Appendix A.

7.3.2 Deferred Write
Normally, each write-type record operation (delete, put, and update) results
in a bucket being written to disk. This convention emphasizes data integrity:
you know that when a write-type operation ends successfully, the file reflects
that operation.
However, you may improve the performance of sequential (by Primary Key)
write-type operations by using Deferred Write. Basically, Deferred Write
directs RMS-11 to write a bucket out to disk only when RMS-11 must use the
1/0 buffer for some other purpose.

NOTE
Deferred Write, although not illegal, is essentially invalidated
while an Indexed file is shared by multiple tasks or
streams-except when you are also using Mass Insert. In the
non-Mass-Insert, write-shared environment, every write-type
operation results in an I/O operation so that:

7-8

Indexed Task Design

"-"'

• The bucket locked by the prerequisite get or find (for update
and delete operations) or by the put operation can be
released.
• The new data is available to the other tasks or streams.
Example Your records are 114 bytes long and the bucket size is two blocks. During sequential
write-type operations, Deferred Write could cause I/0 operations per bucket to drop
from nine to one.

Deferred Write offers little or no benefit to random write-type operations or
read-type operations of any mode.
Only your application program can specify Deferred Write:
MACR0-11
Use the initialization macro or $SET to set the value FB$DFW in the FOP
field of the FAB of the Indexed file before initiating the $CREATE or
$OPEN operation.
BASIC-PLUS-2
BASIC-PLUS-2 does not support Deferred Write.
PDP-11 COBOL
PDP-11 COBOL does not support Deferred Write.
RPG II
RPG II does not support Deferred Write.
DIBOL
DIBOL does not support Deferred Write.

7.3.3 Multiple Buffers
When you open an Indexed file, RMS-11 normally sets up two bucket-sized

VO buffers in your task's address space. RMS-11 uses both buffers for record
operations. However, you can direct RMS-11 to use more than the two
buffers.
RMS-1 Jl uses any extra buffers to keep, or cache, index Root buckets if either
of the following is true:
• The file is shared only by tasks with read-only access.
• The file is not shared.
RMS-11 caches the Roots as it uses them. Therefore, only keys specified or
implicit in record operations have their index Root buckets cached:
• During normal put operations, RMS-11 typically accesses all indexes in a
file. You benefit from root caching only when the number of extra buffers
equals or exceeds the num her of indexes.

Indexed Task Design

7-9

• During Mass Insert put operations, one extra buffer provides some benefit,
regardless of sharing and number of indexes. If the file is not being shared,
you benefit from Root caching only when you provide one more extra buffer
than indexes.

• During get operations, RMS-11 accesses one index (associated with the key
of reference). You benefit from Root caching when you provide an extra
buffer for each different key you reference.
• During update and delete operations, RMS-11 accesses the Alternate
indexes where a SIDR must be inserted or deleted. You benefit from Root
caching when you provide an extra buffer for each Alternate index affected.
While Root caching saves one disk read per index accessed, you may be able
to employ the address space used for the extra buffers more profitably to
optimize RMS-11 overlays (see Chapter 8).

7 .3.4 Multiple Record Access Streams
RMS-11 allows each program to use from one to 255 streams on an Indexed
file.

7.3.5 Sequentlally Reading Write-Shared Files
If your task is trying to read sequentially by Primary Key an Indexed file that
is write-shared, you can improve performance by specifying write-access as
well.

Example Include in your BASIC-PLUS-2 OPEN statement the clauses ACCESS MODIFY
and ALLOW MODIFY.

When there is a possibility that your task will update a record (established
when it opened the file), RMS-11 locks the bucket when your task gets a
record and holds the bucket in the task's 1/0 buffer. If your task then gets
records sequentially, RMS-11 finds them in memory. When a record in a
different bucket is specified, RMS-11 unlocks the previous bucket and repeats
the procedure with the new one.

However, if your task opens a file in a read-only and write-sharing mode,
RMS-11 does not retain the lock on the buckets read; RMS-11 reaccesses the
file for each following get operation, though it does not start at the Root and
go down the index again.

7. 4 File Operations
You can perform the following file operations on Indexed files. File operations
do not involve records and can only perform synchronously.

7.4.1 Close
A close operation disconnects all Record Access Streams connected to a file
before it releases access.

7-10

Indexed Task Design

7 .4.2 Create
In addition to the file specification, RMS-11 passes the following information
to the file processor when it creates a file:

• An initial allocation of blocks for each area in the file. You specify both the
areas and their allocations in your instructions to RMS-11.
• The location on a specific device where the processor should allocate those
blocks. You also supply this information.
• The following file attributes:
File organization
Record format
Forms control
Record size
Number of virtual blocks in the file
Bucket size
Default extension quantity
The other file attributes, such as key and area descriptions for Indexed files,
in the Prologue of the file.

7.4.3 Open
You can specify the file you want to open in two different ways:
By filespec
The first time you open a file, you must use the file specification (see
Appendix A).
By File ID
When you create or open a file by filespec, RMS-11 returns an identifying
notation to your program. You can store this File ID, either in memory or
in a file, and use it to open the file from that point on.
On IAS/RSX-llM systems, open by File ID is significantly faster than open
by filespec, because the process bypasses directory reads and other overhead.
However, on RSTS/E, open by File ID is no faster than open by filespec.

7.4.4 Erase
"File Operations," Section 1.2.fi.3, introduces the concept of erasing files.

7 .4. 5 Extend
"File Operations," Section 1.2.5.3, introduces the concept of extending files.

Indexed Task Design

7-11

Chapter 8
Co1mmon Optimization Techniques
Chapter 2 introduced four application design considerations. Two, sharing
and ease of design, were discussed there. The others, speed and space, were
the underlying concepts for the file and task design discussions in Chapters 4
through 7. They are also the prime considerations for the use of the techniques
discussed in this chapter.
You can optimize the speed of and the space used by your application by:
• improving the structure of each task; this includes:
- the method of combining your program with RMS-11 routines (discussed
in Section 8.1)
- program development, including sequence of operations (discussed in
Section 8.2)
• using all features of the environment where the task runs. Especially important is optimization of virtual-to-logical-block mapping (discussed in
Section 8.3), but there are other factors (discussed in Section 8.4).

8.1 Task Building with RMS-11 Routines
The software routines that perform the RMS-11 functions are distinct from
your programming language. These routines must be combined with your
program with the following steps (see Figure 8-1):
1. A compiler or the assembler converts your program to object code. In the
process, the RMS-11 routines that your program uses are listed as unresolved global references.

2. The Task Builder utility, provided with your operating system, combines
object modules into an executable task. It resolves the RMS-11 global
references with the RMS-11 routines in either:
• an object module library named RMSLIB.OLB
• an RMS-11 Resident Library
You must select the form of RMS-11 that is joined with your program to
make a task. This section should guide your choice.
3. When the Task Builder is finished, your task is ready to run.

8-1

Figure 8-1: Source-to-Task Sequence
SOURCE
PROGRAM

COMPILER
or
ASSEMBLER

TASK
BUILDER

H-MK-00074-00

The RMS-11 routines that become part of your task can be overlaid or not.
Overlays are task segments that can run independently. Therefore, they do
not have to be available to the task at the same time and can share address
space. When a segment is needed, the operating system makes it available,
replacing (overlaying) a segment no longer being used. By interchanging its
parts, a task can run even though it is too large to be executed as one piece.
Nonoverlaid RMS-11 - The Task Builder concatenates the RMS-11
routines with your program, that is, without overlays (see Figure 8-2A), if you
add the following term to the command line:
tLB: [ 1, 1 JRMSLIB/LB 1

The Task Builder extracts from RMSLIB.OLB only those routines required
by your program. These routines contribute from BKB to 44KB to the task
size.
Overlaid RMS-11 - If the sum of your program, including Run-Time System, and RMS-11 code is greater than 64KB, there is not enough user address
space on the PDP-11 for your task to run without overlays.
LValid for synchronous operations only. If you are using IAS/RSX-11 M asynchronous 1/0 operations, you must use the following terms:

1LB:Cl 1lJRMSLIB/LB:ROEXEC:RORSET:ROWATB1RMSLIB/LB

8-2

Common Optimization Techniques

Figure 8-2: RMS-11 in Tasks
A. NONOVERLAID RMS-11

RMS-11
PROGRAM

.__________________________

___._______________________....J

~------..._-.

BUFFERS _ _.,;,-

FROM
8KB

TO
44KB

B. RMS-11 IN DISK-RESIDENT OVERLAYS

RMS-11 (ROOT)
RMSFAB

RMS RAB

PROGRAM

BUFFERS ----::s-

-1•r--------10KB-----•-1
(CAN BE LARGER WITH
FEWER OVERLAYS)

C. RMS-11 IN MEMORY-RESIDENT OVERLAYS

RMS-11
RESIDENT

PROGRAM

LIBRARY

BUFFERS-~·

UNSHARED RMS-11-----'

ADDRESS PAGE REGISTERS
F-MK-00054-00

Common Optimization Techniques

8-3

NOTE
Although you can overlay segments of your program, this section is devoted to the best use of RMS-11 overlays. Therefore,
all references to overlays mean overlays in RMS-11 routines.
Overlays can take one of two forms:
Disk-Resident Overlays
The overlay segments are part of the task image (see Figure 8-2B), and
they remain on disk until they are needed. When a routine is required, the
operating system reads the overlay segment containing that routine into
the task's address space, replacing a segment no longer needed. Section
8.1.1 discusses disk-resident overlays.
Memory-Resident Overlays
The overlay segments are part of a task image (see Figure 8-2C) maintained separately in memory. When a routine is needed, the operating
system maps the segment into the task's address space with two of its
Active Page Registers (APRs). Section 8.1.2 discusses memory-resident
overlays.

8.1.1

Disk-Resident Overlays

One disk-resident overlay can address others which can address others and so
on. This chain of calls defines the overlay structure of a task. You describe
this structure in a file with Overlay Description Language (ODL) statements
described in your Task Builder manual. See Figure 8-3 for a sample overlay
structure and an ODL file that describes it.
You must generate an ODL file for each overlaid task and supply it to the
Task Builder. However, you do not create ODL statements for the RMS-11
portion of your task. The RMS-11 installation process provides overlay
descriptions in two forms:

• A series of standard ODL files describing disk-resident RMS-11 overlay
structures that require differing amounts of task address space. The larger
structures may run faster; you should use the best one for your application.
• A prototype ODL file you can modify, making overlay segments larger if
there is room in your address space or eliminating them if your program
does not use those functions.
The installation process places these files in account [1,1] on logical device
LB:.

8-4

Common Optimization Techniques

(

Figure 8-3: Sample Overlay Structure and ODL File

-I

J
B

(CALLED BY A)

l
0

(CALLED BY A OR B)

!3
!3
0
::s

9·
.....

(CALLED BY A OR C)

o·

::s
~
~

t:J"'

::s

l
.ROOT A-(X,Y,D)
X: .FCTR B-(E,F)

s::

Y: .FCTR C-(G,H,Z)

rJl

Z: .FCTR 1-(J,K,L)

(CALLED BY A,C,OR I)

T-MK-00053-00

er'
Qt

Each higher level programming language has its method of generating the
ODL file for your program and referencing the RMS-11 ODL files. They
normally generate the following hierarchy of files:
program-name.CMD

You supply this indirect file to the Task Builder utility. The file contains
the appropriate command line(s) for the utility and references a primary
ODL file.
program-name. ODL

This primary ODL file determines the general structure of the task and
references secondary ODL files, including RMS-11 ODL files, such as a
standard file or your modification of the prototype.
See your language documentation for more details.
However, if you are a MACR0-11 programmer, you must write your own ODL
file. Make sure the file contains the following terms, if you want to use
RMS-11 disk-resident overlays:
• The factor names RMSROT and RMSALL in the .ROOT statement.
RMSROT represents a set of concatenated modules that perform functions
common to multiple RMS-11 operations. Normally, you should concatenate
RMSROT with your program's root so that it is memory-resident while the
task runs. However, you can concatenate RMSROT with one of your own
overlay segments if both of the following are true:
- RMSROT is memory-resident before your program opens any RMS-11
files and RMSROT remains memory-resident until all RMS-11 files are
closed.
- Each time RMSROT is brought into memory, your program initializes the
RMS-11 control areas with the $INIT macro.
RMSALL represents the overlay structure of modules that perform
RMS-11 operations. RMSALL can be:
- concatenated with your nonoverlaid program. The RMS-11 overlay structure then becomes the only tree in the task.
- concatenated with one of your program's overlay segments. RMSROT
must be memory-resident also before your segment initiates any RMS-11
operation. This approach avoids the division of your task's address space
into separate co-tree areas.
- set up as a co-tree. Both trees, yours and RMS-ll's, have a longest series
of overlay segments that must be resident in memory at the same time.
This series determines the maximum amount of memory the overlay
structure requires. The concatenation of these co-tree maximums must be
less than 64KB.
Example

• ROOT USER-RMSROT-RMSALL

RMSROTand RMSALL are concatenated with a nonoverlaid user root.

8-6

Common Optimization Techniques

.ROOT USER-RMSROT-*<RMSALL>

Example

RMSROT and RMSALL are concatenated with a nonoverlaid user root. An
autoload indicator is included for RMSALL, but this indicator is not necessary
because the secondary RMS-11 ODL file includes all necessary autoload indicators.
Example

.ROOT USRROT-RMSROT-USRSEGtRMSALL
RMSROT is concatenated with the user root. RMSALL is defined as a co-tree
with the user overlay structure as the main tree. Each tree is segregated in the
task's virtual address space.

Example

.ROOT USRROT-RMSROT-CUSRSEGtRMSALL)
RMSROT is concatenated with the user root. RMSALL is overlaid with the
user overlay structure. Each tree can use all of the task's virtual address space,
less the root, but modules in one tree cannot call modules in the other tree.
Therefore, no segment in USRSEG can perform RMS-11 operations .

., An indirect reference to an RMS-11 ODL file, either a standard file or your
customized version of the prototype, in the form:
@file-· name
This RMS-11 ODL file resolves the references to RMSROT and RMSALL.
Example

.ROOT
USRROT-RMSROT-USRSEGtRMSALL
USRSEG: • FCTR
<USR 1 tUSR2 tUSR3)
@LB:[1 t1JRMS11H
.END

8.1.1.1 Standard ODL Flies-DIGITAL provides the following standard ODL

files. Do not change these files.
RMSllS.ODL
Structured to add a little more than 8KB to task size, this file features
only Sequential and Relative file organization routines in nineteen overlay
segments.
RMSllX.ODL
Structured to add a little more than 9KB to task size, this file features
Sequential, Relative, and Indexed file organization routines in 57 overlay
segments.
HMS12X.ODL
Structured to add no more than 12KB to task size, this file features
Sequential, Relative, and Indexed file organization routines in 17 overlay
segments.
The size of the RMS-11 overlay structure and the number of overlay
segments were changed by modifying the RMSllX.ODL overlay structure
as follows:
• File open and Record Access Stream connect routines combined in one
overlay segment.
Common Optimization Techniques

8-7

• File close and Record Access Stream disconnect routines combined in
one overlay.
• Sequential file record operation routines combined in one overlay.
• Relative file record operation routines combined in one overlay.
• Routines that insert records into Primary Level 0 combined in one overlay.
• Primary Level 0 bucket splitting routines combined into two overlay
segments separate from record insertion routines.
• Routines that insert Secondary Index Data Records (SIDRs) into
Alternate Levels 0, including bucket splitting, combined into one overlay.
• Routines that insert index records into Levels 1+, including bucket splitting, combined into one overlay.
This modification reduced overlays for Indexed organization routines from
32 to 8.
Appendix C contains copies of ODL files for RMS-11 overlay structures that
occupy 16KB and 20KB. You may enter one or both of these files on your
system and use them while building a task. Appendix C also contains a
graphic illustration of the overlay structures created by RMSllS.ODL and
RMSllX.ODL.
8.1.1.2 Prototype ODL Optimization - The prototype 9KB ODL file is named
RMSll. ODL. This file is similar to the standard RMS11X.ODL file, except
that it contains comments and instructions in addition to ODL statements.
You can optimize this overlay structure to accommodate task requirements.

You change an ODL file as follows:
• Combine segments that overlay each other into a single overlay; this change
reduces the number of overlays and increases task speed at the cost of task
size (the task gets bigger). For example, some RMS-11 record operations,
notably the Indexed file put and update operations, require a series of
overlay segments in the 9KB overlay structure. You could make these operations run faster by combining some or all of these segments.
• Eliminate modules because your task does not use those functions, and the
task's disk file gets smaller though the task image in memory may not
change size.

NOTE
Know what you can eliminate, particularly in the ODL files
for higher-level-language programs: the compiler can call
routines that you are not aware of. For example,
BASIC-PLUS-2 tasks use Block I/O for virtual arrays. If you
eliminate something the task needs, the Task Builder prints
an UNDEFINED SYMBOLS error message.

8-8

Common Optimization Techniques

When you are going to modify the prototype, first copy the file into another
file with a different name and/or account: not everyone on your system wants
the same optimizations you do, so make them specific to your application.
8.1.1.2:1 Techniques - While you are changing the prototype, obey all Task
Builder syntax rules and conventions (see the Task Builder manual for your
operating system) and observe the following methods:

NOTE

Under no circumstances should you change or move the modules included in any factor.
Concatenate overlaid modules within a factor one at a time.
Example The factor
Fl:

.FCTR A-B-(C,D,E,F)

can be optimized incrementally for performance, with one overlay made memoryresident with A and Bin each of the following steps (except the last); see also Figure
8-4.
Fl:

.FCTR A-B-C-(D,E,F)

2. Fl:

.FCTR A-B-C-D-(E,F)

3. Fl:

.FCTR A-B-C-D-E-F

Each of these factors is valid and represents a level in the size and speed of the task.

Figure 8-4:

,..,.,.

Incremental Optimization Example

COMBINATION STEPS:

CD F1

c
D
E

H-MK-00052-00

Common Optimization Techniques

8-9

Combine factors with caution.
• Do not combine around overlays: either concatenate the overlaid modules or
put them at the end of the module chain.
Example The factors:

Fl:
F2:

.FCTR A-B-(C,D)
.FCTR X-Y-(Z,W,R)

cannot be combined like this (see also Figure 8-5):
Wrong Fl:
F2:

.FCTR A-B-(C,D)-F2
.FCTR X-Y-(Z,W,R)

because this implies:
Fl:

.FCTR A-B-(C,D)-X-Y-(Z,W,R)

The Task Builder cannot successfully build this task because it would have to
combine modules X and Y with modules A and B allowing room for the variable
lengths of C and D in between.
Right Fl:
F2:

.FCTR A-B-F2
.FCTR X-Y-(C,D,Z,W,R)

Note that each original root (A-B and X-Y) is still the root for its associated
overlays (C,D and Z,W,R, respectively).

Figure 8-5: Concatenation around Overlays Example
SOURCE:

F1~

x
y

WRONG CONCATENATION:
F1

c
F2

x
y

1 1
w

RIGHT CONCATENATION:
F1

8
F2

x
y

R
H-M K-00051-00

8-10

Common Optimization Techniques

• Remove duplicate module names. Duplicates will arise because some factors
use common modules to preserve overlay tree linkages.
Example The factors

Fl:
F2:

.FCTR A-B-(C,D)
.FCTR X-Y-B-(Z,W,R)

are combined in this way (see also Figure 8-6):
Fl:
F2:

.FCTR A-B-F2
.FCTR X-Y-(Z,W,R,C,D)

Figure 8-6: Duplicate Mod.ule Example No. 1

F1FR

SOURCE:

x
y

G_B

l 1
w

CONCATENATION:
F1

A
B

F2L-~~~~~~~~~~~---i

x
y

D
H-M K-00049-00

Example The factors

Fl:
F2:
F3:

.FCTR A-(F2,F3)
.FCTR B-C-(X,Y,Z)
.FCTR D-(X, Y,Z,R)

can be combined in this way (see also Figure 8-7) :

Common Optimization Techniques

8-11

Fl:
F2:
F3:

.FCTR A-F3-F2
.FCTR B-C-(X,Y,Z)
.FCTR D-R

or in this way:
Fl:
F2:
F3:

A-F2-F3
B-C
D-(X, Y,Z,R)

This example simulates the optimization of Indexed file put and update operations discussed in Examples of ODL Optimization later in this section.

Figure 8-7: Duplicate Module Example No. 2

SOURCE:
F1

I I
y

CONCATENATIONS:

I I
y

x
z

I I l
y

R
H-M K-00050-00

• Preserve the route down the overlay tree to each module, that is, all modules above the one in question must be in memory before it is brought in.
See Figure 8-8 for examples.
Example Before RMS-11 calls the R3IKYI module to insert index records into Levels 1 and
higher for a put operation, it must have in memory the overlay segments
RMSRAB, RMSIDX, R3PUT, and R3UIDX.

8-12

Common Optimization Techniques

F:igure 8-8: Overlay Structure Diagram
RMS11

·~·

RM SF AB
ROPRFN
RODPYC
R1CLOS
R2CLOS
R3CLOS
ROOPFL
RMSCRE
R1CRCK
R2CRCK
R3CRCK
ROCRFL
R2WPLG
R3WPLG
RMOPIN
RM SR AB
RMSCD
RMS REL
R2GUPD
R3PUT
RMS ID}<
R3GET
R3PUT
R3P I }-{C
R3IUDR
R3IUDC
R3BSPL
R3BRR'.'
R3ALOC
R3 IS ID
R3ALOC
R3ISDI
R3U ID}<
R3IKYI
R3ROOT
R3ALOC
R3MK ID
R3UPDA
R3DELE
R3IUDR
R3IUDC
R3BSPL
R3BRRl.i
R3ALOC
R3ISID
R3ALOC
R3ISDI
R3U I 0}-{
R3IKYI
R3ROOT
R3ALOC
R3MK ID
RMSMIS
RMS SEQ
R1GET
R1PUT
R1UPDA

Common Optimization Techniques

8-13

You can receive the following error
messages from the Task Builder after you modify an ODL file:
8.1.1.2.2 Possible Task Bullder Errors -

n UNDEFINED SYMBOLS SEGMENT $name
Put the specified module names back in the appropriate factors.
MODULE $name MULTI PL y DEFINES SYMBOL $name

Delete the specified module name from the last factor in the overlay
segment containing it.
DIGITAL established a convendon for RMS-11 module names. Therefore, if
the Task Builder prints the form $name, look in your ODL file for the module
name:
Rx name

where:

x = 0 for routines common to multiple file organizations
x = 1 for Sequential organization routines
x = 2 for Relative organization routines
x = 3 for Indexed organization routines
8.1.1.2.3 Calculating Changes In Task Size -As you change the overlay struc-

ture, you may be changing the task size:
• on disk, if you combine segments and eliminate duplicate modules
• in memory, if you affect the largest overlay segment
To determine your affect on these sizes, you must know the sizes of the
RMS-11 factors and within the factors, the modules. There are two sources:
1. Your best source is a task-build map. A map not only shows overlay
segment sizes, but their relative levels. Figure 8-8 is derived from a taskbuild map. You generate a map by including a second output file specification in your Task Builder command line.
2. Use the Librarian utility provided by your operating system to list the
modules in LB: [1,lJRMSLIB.OLB. Add module sizes to get factor sizes.
Source ODL statements in this
section were taken from the prototype RMSll.ODL file supplied with each
RMS-11 kit.
8.1.1.2.4 Examples of ODL Optimization -

NOTE
• Some overlay segments for Indexed file routines are larger.
than any of the segments containing Sequential and Relative
file routines. Therefore, changes in Sequential and Relative
overlay segments impact task size only if you have eliminated
these large Indexed operations. However, th~ changes do decrease overlay 1/0 operations and possibly, the size of the
task's disk file.

8-14

Common Optimization Techniques

• The term LB: in this section has the same meaning on
IAS/RSX-llM systems as LB:[l,1].

Sequential File Record Operations - The prototype has Sequential file record operations in four overlay segments:

• Factor RMSQOP contains the root for the record operations.
• Factor RMINlO contains get and find operations.
• Factor RMOUlP contains the put operation.
• Factor RMOUlU contains the update operation.
To reduce the number of overlay segments, modify the RMSQOR factor,
referenced by the root factor RMSQOP:
RMSQOR: .FCTR CRMIN10tRMOU1PtRMOU1U>

as follows:
• To combine the get and find operations with the root factor, change
RMSQOR to:
RMSQOR: .FCTR RMIN10-CRMOU1PtRMOU1U)

'-''

GAIN 1 less overlay segment (plus an increase in segment size)
• To combine all operations with the root factor, change RMSQOR to:
RMSQOR:

.FCTR RMIN10-RMOU1P-RMOU1U

The RMOUlP and RMOUlU factors duplicate the module RlPSET.
Therefore you must change RMOUlU from:
RMOU1U: .FCTR LB:RMSLIB/LB:R1UPDA:R1UBLD:R1PSET

to:
RMOU1U: .FCTR LB:RMSLI6/L6:R1UPDA:R1UBLD

GAIN 3 less overlay segments (plus an increase in segment size)
Relative File Record Operations operation in three segments:

The prototype has Relative file record

• Factors RMOX26, RMI02C, and RMI02A contain the root for the record
operations.
• Factor RMI02G contains the get, update, and delete operations.
• Factor RMI02P contains the put operation.

Common Optimization Techniques

8-15

To put all record operations in a single overlay, make the following changes:
1. Change RMI02A from:
RMI02A:

.FCTR LB:RMSLIB/LB:R2FIND:R2GSET-CRMI02GtRMI02P>

to:
RMI02A: .FCTR LB:RMSLIB/LB:R2FIND:R2GSET-RMI02G-RMI02P

2. Change the RMI02U factor referenced by RMI02G from:
RMI02U:

.FCTR LB:RMSLIB/LB:R2UPDA:R2PSET

to:
RMID2U:

.FCTR LB:RMSLIB/LB:R2UPDA

because both RMI02P and RMI02U use the module R2PSET.
GAIN 2 less overlay segments (plus an increase in segment size)
Indexed File Record Insertion Operations - During both put and update
operations, RMS-11 may insert records into buckets of all levels of the Primary and Alternate indexes. The routines that perform these record insertions
are located in module sets, one set for each type of insert (Primary Level 0,
Alternate Level 0, and index levels). One of the modules is a control module;
the other(s) perform the insert. In the prototype overlay structure, each module of each set is a separate overlay.

The simplest improvement of Indexed file record operations combines the
control and execution modules of each set into one overlay segment. Use any
or all of the following steps:
• To combine the Primary Level 0 insertion pair, change:
RMOU34: .FCTR LB:RMSLIB/LB:R3IUDR-CRMOU3FtRMOU3G1RMOU3A)

to:
RMOU34:

.FCTR LB:RMSLIB/LB:R3IUDR-RMOU3F-CRMOU3GtRMOU3A>

The bucket splitting routines in RMOU3G and RMOU3A stay overlaid. The
usefulness of this change depends on file activity. If you populate the file
using a fill num her so that the file has room for random record insertions,
thus avoiding bucket splits, then this optimization correlates with that
savings.
GAIN 1 less overlay segment (plus an increase in segment size)

8-16

Common Optimization Techniques

To combine the Alternate Level 0 insertion pair, change:
RMOU35:

.FCTR LB:RMSLIB/LB:R31SID-CRMOU3HtRMOU3A>

to:
RMOU35:

.FCTR LB:RMSLIB/LB:R31SID-RMOU3H-RMOU3A

There is no gain in leaving RMOU3A overlaid; the Task Builder allocates
the space in the task anyway.
GAIN 2 less overlay segments (plus an increase in segment size)
• To combine the index bucket insertion pair, change:
RMOU3G:

.FCTR LB:RMSLIB/LB:R3UIDX:R31KEY-CRMOU3MtRMOU3A>

to:
RMOU3G:

.FCTR LB:RMSLIB/LB:R3UIDX:R31KEY-RMOU3M-RMOU3A

There is no gain in leaving RMOU3A overlaid; the Task Builder allocates
the space in the task anyway.
GAIN 2 less overlay segments (plus an increase in segment size)
TOTAL GAIN 7 less overlay segments (plus an increase in segment size)
Indexed Get and Update Record Operations date operations, change:
RMI031:

.FC~R

To combine the get and up-

CRMI03GtRMI03UtRMI03P>

to:
RMI031:

.FCTR CRMI03G-RMI03U tRMID3P>

'rhis change combines the branch factors for the get and update operations
and leaves them overlaying the put operation branch.
The factor RMI03G contains no overlays, but the factor RMI03U has several
layers of overlays. You can further optimize update operations with one or
both of the following steps:
•Change:
RMI03U:

.FCTR LB:RMSLIB/LB:R3UPDA:R3USET:R3RPLC-<RMOU3QtRMI03D>

to:
RMI03U:

.FCTR LB:RMSLIB/LB:R3UPDA:R3USET:R3RPLC-RMI03D-RMOU3Q

Common Optimization Techniques

8-17

• Make all of the following changes:
1. Change:
RMOU3a:

.FCTR LB:RMSLIB/LB:R31UDR-CRMOU3Ft RMOU3Gt RMOU3A)

to:
RMOU3a:

.FCTR LB:RMSLIB/LB:R31UDR-RMOU3F-RMOU3G-RMOU3A

2. Change:
RMOU35:

• FCTR LB: RMSLI B/LB: R31 SID- CRMOU3H, RMOU3A >

to:
RMOU35:

.FCTR LB:RMSLIB/LB:R31SID-RMOU3H-RMOU3A

3. Change:
RMOU3Q:

.FCTR CRMOU3a1RMOU351RMOU3G1RMOU3P>

to:
RMOU3Q: .FCTR RMOU3a-RMOU35-CRMOU3G1RMQU3P>

These changes concatenate the routines that insert records into index
and data levels, while it leaves the routines that handle bucket splitting
overlaid.
Combinations of File Organizations - Some applications use more than one
type of file, getting from one and putting to the other (or some other combination of record operations). Optimization of ODL files for these applications
requires the combination of modules from different branches of the overlay
structure.

The overlay structure for record operations starts with the factor:
RMSREC:

.FCTR RMSRAB-CRMSSQ01RMOXZG1RMOX3G1RMSBLK 1RMMISC>

Below this factor, the routines for a record operation are included in your task
depending on whether you select a valid or a dummy factor.
You can combine Sequential file get and Indexed file put operations with the
following steps:
1.

Select valid RMSQOP factor for Sequential get operations (RMSQOP is
referenced by RMSSQO):
RMSQOP:

.FCTR LB:RMSLIB/LB:R1WTLS:R1CKEF:R1DELE-RMSQOR

2. Select valid RMI03P factor for Indexed put operations (RMI03P is referenced by RMI031 which is included in RMOX36):
RMI03P:

8-18

.FCTR LB:RMSLIB/LB:R3PUT:R3PSET-CRMOU3Q1RMOU33)

Common Optimization Techniques

3. Select all factors associated with RMI03P:
RMOU3CJ
RMOU3ll
RMOU3F
RMOU3G
RMOU3A
RME}<TD
RMOU35
RMOU3H
RMOU3A
RMEHTD
RMOU36
RMOU3M
RMOU3A
RMEHTD
RMOU3P,
RMOU3A
RMEHTD
RMOU33

Select dummy factors for RMOX26 (Relative file record operations),
RMSBLK (Block VO), and RMMISC (flush, free, truncate, and rewind
operations):

RMOH26: • FCTR RMDDSCJ
RMSBLK: • FCTR RMDBLK
RMM I SC: • FC TR RMDDMC

You can accomplish the same result by modifying the RMSREC factor as
follows:
Change:
r<MSREC: • FCTR RMSRAB- ( RMSSCJO , RM0}-{26, RM0}-{36 , RMSBLK , RMM I SC)

to:
r<MSREC: • FCTR RMSRAB- <RMSSCJO, RMOH36)

GAIN 7 less overlay segments (plus an increase in segment size)
The basic optimization is complete: the overlay structure reflects only those
:record operations used by the task.
A further improvement combines modules that remain overlaid. Each of the
following steps is an optimization that you can perform without regard to the
others.
1
•

To optimize the RMSREC factor, change it from:
RMSREC: .FCTR RMSRAB-<RMSSQO,RMOH26,RMOX3G,RMSBLK,RMMISC)

to:
RMSREC:

+FCTR RMSRAB-<RMSSCJO-RMOH3G,RMOX2G,RMSBLK,RMMISC)

Common Optimization Techniques

8-19

This change combines the branch overlay segments for Sequential file record operations and for Indexed file record operations into a single overlay.
GAIN 1 less overlay segment (plus an increase in segment size)
If you changed the RMSREC factor according to the previous step, this step
becomes:

Change:
RMS REC: • FCTR RM SR AB-< RMSSQO, RM0}-{36 >

to:
RMS REC:

• FCTR RMSRAB-RMSSQO-RMOH36

• To optimize the RMI03P and other factors, combine the modules associated with a normal put operation, while leaving the bucket splitting
modules overlaid. Change them from:
RMI03P: .FCTR LB:RMSLIB/LB:R3PUT:R3PSET-CRMOU3Q,RMOU33)
RMOU3Q: • FCTR <RMOU3l1, RMOU35 tRMOU36 tRMOU3P >
RMOU3l1: .FCTR LB:RMSLIB/LB:R3IUDR-CRMOU3FtRMOU3GtRMOU3A)

to:
RMI03P: .FCTR LB:RMSLIB/LB:R3PUT:R3PSET-RMOU34-<RMOU3Q,RMOU33)
RMOU3Q: .FCTR CRMOU3S,RMOU3GtRMOU3PtRMOU3GtRMOU3A>
RMOU34: .FCTR LB:RMSLIB/LB:R3IUDR-RMOU3F

GAIN 2 less overlay segments (plus an increase in segment size)

8.1.2 Memory-Resident Overlays
The RMS-11 Resident Libraries contain RMS-11 routines in re-entrant executable code. Tasks that use RMS-11 can be built with global references
resolved in one of the Resident Libraries. The associated Resident Library
must be resident in memory before the tasks can be executed.
While it is executing one of these tasks, the operating system uses one or two
of the task APRs to map references from the task to the Resident Library.
Therefore, any time the task requires an RMS-11 routine, the operating system changes the APRs to point at the segments of the Resident Library that
contain the routines for the operation.
This mapping is called memory-resident overlaying. Since the overlay segments are in memory, the operating system does not perform an 1/0 operation
to provide the routines as it does with disk-resident overlays.

8-20

Common Optimization Techniques

8.1.2.1

Bulldlng the RMS-11 Resident Libraries -

The RMS-11 installation

process provides two RMS-11 Resident Libraries:
RMS RES
This 48KB Library consists of a root segment and five segments containing the RMS-11 routines that support all file organizations. While a task
using the Library is running, one of its APRs is constantly pointing at the
root segment while the other points to the appropriate routines (see Figure
8-2C).
RMSSEQ
This 8KB Library consists of one segment of routines that support
RMS-11 Sequential files only. The operating system uses one APR to map
this Library.
Each RMS-11 Resident Library is position independent within the virtual
address space of your task. You can assign the APRs for a Resident Library or
allow the Task ·Builder to use default assignments: see your Task Builder
manual.

lAS/RSX-llM - The RMS-11 installation process provides the following
files:
RMSRES.TSK and RMSSEQ.TSK
The task images of the Resident Libraries. This file must be resident in
memory before any task using the Library can run (see ''Installing an
RMS-11 Resident Library," Section 8.1.2.2).
HMSRES.STB and RMSSEQ.STB
The symbol table files for the Resident Libraries. When the Task Builder
links a task to a Library, it uses the appropriate STB file to resolve global
references (see "Task Building against an RMS-11 Resident Library,"
Section 8.1.2.2).

RSTS/E - The RMS-11 installation process provides the following files:
RMSRES.TSK and RMSSEQ.TSK
The task images of the Resident Libraries. The appropriate file is used by
the Task Builder when you build against a Resident Library (see "Task
Building against an RMS-11 Resident Library," Section 8.1.2.2).
RMSRES.STB and RMSSEQ.STB
The symbol table files for the Resident Libraries. When the Task Builder
links a task to a Library, it uses the appropriate file to resolve global
references (see "Task Building against an RMS-11 Resident Library,"
Section 8.1.2.2).
HMSRES.LIB and RMSSEQ.LIB
Specially formatted files containing the task images of the Resident
Libraries. This file must be resident in memory before any task using the
Library can run (see "Installing the RMS-11 Resident Library," Section
8.1.2.3).

Common Optimization Techniques

8-21

After the Resident Libraries are built, you can build tasks, directing the Task Builder to
resolve global references with a Library, with one of the following sequences of
commands:
8.1.2.2 Task Bulldlng against an RMS-11 Resident Library -

TKB command string
or
TKB>/
ENTER OPTIONS:
TKB>LIBR=RMSRES:RO
TKB>//

TKB command string
TKB>/
ENTER OPTIONS:
TKB> LIBR=RMSSEQ:RO

TKB">!!

where:
command string is described in the Task Builder manual for your operating
system.
You must also do one of the following:
• Specify RMSRLX.ODL for RMSRES and RMSRLS.ODL for RMSSEQ as
the RMS-11 secondary ODL file-name in your primary ODL file.
• Add the following term to the Task Builder command string if you are using
synchronous RMS-11 operations:
,LB:[l, lJ RMSLIB/LB:ROA UTO,RMSLIB/LB
If you are using IAS/RSX-llM asynchronous 1/0 operations, you must use

the following terms:
,LB: [l,lJRMSLIB/LB:ROEXEC:ROAUTO,RMSLIB/LB
The Task Builder looks for the appropriate .TSK and .STB files on the library
device LB: in account [1,1). If you have moved the RMS-11 Resident Libraries
to another account, you must use the RESLIB option so you can specify an
account number. See your Task Builder manual for an explanation of this
option.
When you use RMSRES, the Task Builder adds some RMS-11 code to your
program to form the task. The utility then uses the RMSRES .STB file to
resolve the rest of the global references.

NOTE
• A Library does not have to be installed or resident for you to
task build against it.
• The time required to build a task using a Resident Library is
significantly less than the time required when you use diskresident overlays.
• If you are not using Indexed file put and update operations,
you can eliminate approximately lK of the RMS-11 code
RMSRES adds to your task with the following steps.

1. Copy the RMS-11 standard ODL file RMSRLX.ODL
(used when task building against RMS RES) into a file of a
different name.

8-22

Common Optimization Techniques

2. In the renamed version of RMSRLX.ODL, change the factor:
'

"-"'

RMSALL:

.FCTR *<RM><PUT tRMHUPD)

to
RM SALL:

• FCTR name
• NAME name

where:

name is a six-character string of your choosing.

3. Substitute the name of the modified ODL file for
RMSRLX.ODL in your primary ODL file and build the
task again.
• If you use an RMS-11 secondary ODL file-name other than

RMSRLX.ODL or RMSRLS.ODL, the Task Builder prints
the following error message:
MODULE $name AMBIGUOUSLY DEFINES SYMBOL $name

8.1.2.3 Installing an RMS-11 Resident Library - After a Resident Library is
built and before any task using the Library can run, the Library must be
resident in memory. The installation process depends on the operating system; see the following paragraph that applies to your operating system.

IAS - The operating system automatically installs the Resident Library
when you run a task using it. See the !AS Operator's Procedure Manual.
RSTS/E - Use the UTILTY ADD LIBRARY command. See the RSTS/E
System Manager's Guide for a description of this command.RSX-UM - Use the SET /MAIN command to declare a partition 24KW in
size. Then use the INS command to load the Resident Library into that space.
See the RSX-llM/M-PLUS MCR Operations Reference Manual for a description of the commands.

8.1.3 Deciding between Types of Overlays
The method you use to link RMS-11 routines with your programs depends on
three things:
• amount of physical memory available on your computer system
• amount of virtual address space in your task exclusive of RMS-11 routines
• number of jobs that must be run simultaneously
RMS-11 routines require memory (see Figure 8-2):
•• nonoverlaid RMS-11 routines are linked into a task when they are referenced; the virtual address space required varies from 8KB to 44KB, depend-

Common Optimization Techniques

8-23

ing on the file organizations supported and the record operations used. This
RMS-11 code cannot be shared with other tasks.
• disk-resident overlays reside in the user address space. The smallest overlay
structure requires about BKB of the space; you can use larger structures.
Each task using disk-resident overlays requires physical memory for its own
overlay structure.
• RMSRES uses 48KB of physical memory that can be shared among all
tasks using that Resident Library. A task using the Library reduces its
available address space by 16KB, because two APRs are addressing the
Library. The task also contains approximately lKB of RMS-11 routines.
Physical memory is a secondary consideration when you are dealing with
Resident Libraries. Primarily, a Library saves time because it eliminates the
1/0 operations associated with disk-resident overlays.
The Resident Library approach becomes cost-effective in terms of physical
memory when you have five or more RMS-11 tasks running simultaneously:
1. Assume the 9KB overlay structure. Five tasks running in memory at the
same time need a little more than 45KB of physical memory for RMS-11
alone.

2. Assume the Resident Library. The tasks require about lKB of their own
space for RMS-11, but the system must use 48KB for the Library.
The difference in memory requirements is more than compensated for by
speed.
However, by reducing the memory available for jobs, a Resident Library can
restrict the number of jobs that can run at the same time and thereby increase
swapping or other overhead.

8.2 Program Development
You should consider performance while you are writing an application
program:
• Your program's flow of operations can either cooperate with or fight against
the RMS-11 overlay structure.
• Task-building consumes a significant portion of your machine resources.
Minimize that time when you can.

8.2.1

Flow of Operations Should Reflect Overlay Structure

The overlay process causes a significant portion of the 1/0 performed for a
program with disk-resident overlays. You should structure the task to
maximize the time each segment stays in memory and minimize the number

8-24

Common Optimization Techniques

of overlay operations. You do this by placing similar RMS-11 operations
together in your program. This process also makes you aware of the nature of
the operations your program is performing:
File-related operations
File-related operations are generally required at the beginning and end of
processing. Therefore, they are fairly easy to group.
Example Open all files that the program uses and simultaneously set up all Record
Access Streams at the beginning of the program.
Example Disconnect Record Access Streams and close all the files at one time, probably
at the end of the program.

NOTE
Most higher level languages automatically perform connect and
disconnect operations during the execution of file open and close
statements.

Record operations
The primary overlay burden of your task comes from record operations.
However, the nature of processing often dictates the placement of record
operations in your program. Therefore, the type and sequence of these
operations direct your optimization of the ODL files (see Section 8.1.1).
Example If your task gets from a Sequential file, then puts to an Indexed file, you could
reduce the number of overlays required for those specific operations (see Section
8.1.1.2).
Example If your task gets from an Indexed file, then updates the record, you should
optimize those operations (see Section 8.1.1.2).

Where possible, though, sequence operations to minimize overlays.

8.2.2 Task Bullder Considerations
The Task Builder utility constructs a task and ensures that its overlays, if
any, work properly. To do this, the Task Builder must know the task's overlay
structure if you use disk-resident overlays: you supply this information via an
ODL file.
You want to reduce the time the Task Builder runs while it builds your task.
The first thing you can do is reduce the number of overlays in the task. See
"Disk-Resident Overlays," Section 8.1.1. Each overlay adds time to task
building because it requires a symbol table to be built and then resolved.

NOTE
If you use memory-resident overlays (Resident Library), you
reduce Task Builder overhead needed to process overlay
segments.

You can also reduce task building time by not requesting a map. If you really
need a map for debugging, specify a short one (which is the default anyway).

Common Optimization Techniques

8-25

Example On a PDP-nno using the RSX-UM operating system, task builds were timed:

Map Selection

Time for Task Build

No Map
Short Map
Long Map

x
1.18.x to 1.49x
l.51x to l.91x

8.3 Vlrtual-to-Loglcal-Block Mapping
When RMS-11 issues a data transfer request, it specifies a starting Virtual
Block Number (VBN) and the size of the request in bytes to the operating
system. The system maps the VBN into a Logical Block Number (LBN) it
must use to find the block on disk. To do this, the system uses a set of
retrieval pointers called a window (to the file). The operating system creates a
window in its part of memory by reading the first set of pointers from disk
when a task opens a file. These pointers specify blocks on disk, and from the
structure and content of the pointers for a file, the system equates virtual
blocks to logical blocks.

CONVENTION
The cover term file directory in this manual has
the same meaning as the following systemspecific terms:

System
IAS
RSTS/E
RSX-UM

Term
directory entry and file header(s)
User File Directory entries
directory entry and file header(s)

8.3.1 Retrieval Pointers on Disk
The file directory contains the retrieval pointers for a file. The representation
depends on your operating system.

The file processor stores retrieval pointers in a file
header, using enough file headers to cover the file. A file header can contain
up to 102 pointers. Each pointer consists of:
8.3.1.1

IAS/RSX-11 M -

• the number of blocks the pointer maps
• the Logical Block Number where the group of blocks starts
The largest group of blocks that can be covered by one pointer is 256 blocks.
Therefore, one file header can map a maximum of 26,112 logical blocks.
8.3.1.2 RSTS/E -The file directory includes file retrieval blocks. Each block
contains seven retrieval pointers. A pointer consists of the Logical Block
Number of the first block in a file cluster. Since clusters are groups of logically

8-26

Common Optimization Techniques

contiguous blocks and the number of blocks in a cluster (clustersize) is a file
attribute, the RSTS/E file processor can calculate which logical blocks reside
in a cluster.
However, if the file is contiguous, the file processor bypasses the file retrieval
blocks and uses:
• the LBN of the first block in the file
• the requested VBN as an index into the file

8.3.2 Retrieval Pointers In Memory
An operating system keeps one window in memory for each file. If that window does not contain the retrieval pointer that covers the virtual block requested by RMS-11, the system must bring more pointers into memory in a
process called window turning. Window turning normally requires an 1/0
operation:
• Since the RSTS/E operating system stores windows in linked blocks, the
system may request several I/0 operations before the correct window is
found.
• The IAS and RSX-llM operating systems build File Control Block (FCBs)
in memory when a task opens a file. An FCB contains information about one
file header, including the range of virtual blocks covered by the header's
retrieval pointers. Whenever the system has to turn a window, it consults
the FCBs for the file to determine which file header contains the appropriate retrieval pointer. The file processor then reads that block from disk,
requiring only one 1/0 operation (unless the software needs one or possibly
two overlays).
Example An evaluation of one application revealed that window turning during record operations accounted for nearly 30% of the 1/0 operations.

8.3.3 Optimizing Window Turning
When you reduce window turning, you improve performance. The methods for
doing this are specific to the operating systems.
8.3.3.1 IAS/RSX-11 M - You can reduce the I/0 operations associated with
window turning as follows:

Increase Window Size - Use one of the following methods to increase the
number of retrieval pointers the system keeps in memory for the file:

• Initialize the disk volume that will contain the RMS-11 file with a window
size greater than the default seven pointers per window. See your system
documentation for initialization procedures.
• Mount the volume containing the RMS-11 file using the /WIN switch to
specify a window size greater than the volume default. See your system
documentation for volume mounting procedures.

Common Optimization Techniques

8-27

• Use a MACR0-11 subroutine that sets the RTV field in the FAB for the file.
See the RMS-11 MACR0-11 Reference Manual.

In each of these methods, you can specify an amount as follows:
• If you specify a -1, the system tries to make the window large enough to

map the entire file, using up to 81 retrieval pointers.
• If you specify a positive number of pointers, the system allows a maximum

of 127 pointers in each window.

NOTE
You can also set window size with a BASIC-PLUS-2 program, using the WINDOWSIZE keyword.

The initialization and mount methods apply to all files on a disk. These
methods cause the system to use more executive memory than when you set
window size for an individual file in a program.

..,._

'~~

Maximize Contiguity - Either make the file contiguous, or if that's not possible, reduce the number of extents in the file, making each extent as large as
possible.

One retrieval pointer can map up to 65,536 logical blocks. When the file
processor reads a pointer from the header, the software checks if the extent
mapped by the pointer is logically contiguous with the extent covered by the
preceding pointer in the window. If it is, the file processor adds the extent size
to the size field of the pointer in the window, then it reads the next pointer. If
the two extents are not contiguous, it adds the new pointer to the window.
This compaction extends across file headers.

In this way, IAS and RSX-llM can map any file with a default window if the
file is sufficiently contiguous.
Areas in Indexed Files window turning.

Areas localize successive block requests and reduce

Size of Fl lA CP - The large version of the FllACP does not require overlays
of its own routines to perform window turning, where the small version does.
8.3.3.2 RSTS/E - There are two ways to optimize the I/0 operations associated with window turning:

• Directory caching involves your system configuration.
• Contiguity involves the file's attributes.
Directory Caching - RSTS/E has a Directory Caching feature (also called
FIP Buffering) you can select when you generate your system. This feature
uses the Extended Buffer Pool (X BUF) to save (cache) directo:ry blocks in
memory. If the system needs a certain directory block, to do a window turn,

8-28

Common Optimization Techniques

for instance, and that block is already cached, the block is retrieved from the
cache without a disk VO operation. See the RSTS/E System Manager's Guide
for more explanation of this feature.

NOTE
In RSTS/E Version 7.0, directory caching is related to the data
caching feature (discussed in "Other Optimizations," Section
8.4).

Contiguity - Use one of the following methods to increase the number of
contiguous blocks in the file:
• Make all blocks in the file contiguous and there is no window turning.
Cost When RMS-11 detects that there is not enough room in a file to
complete an operation, it automatically requests the file processor to
extend the file by the Default Extension Quantity. However, if the
file is contiguous, RSTS/E returns a privilege violation. RMS-11
aborts the current record operation, passing to the program the error
code ER$PRV. All put and update operations that require an extension of the file will fail after that.
At this point, the system manager or a privileged user can force the
file to become noncontiguous. The file ·assumes all the window turning problems of noncontiguous files described previously.

~·

Therefore, you should avoid extending (or trying to extend) contiguous files
by:
• completely allocating the file when it is created
• converting the file into a larger one before its space is exhausted (use the
RMSIFL or RMSCNV utility)
• Increase the file clustersize: the fewer the clusters, the fewer the retrieval
pointers, and the less the operating system must turn windows to cover a
virtual block.
Cost Each allocation for the file must contain a whole number of clusters.
As clustersize increases, so does the chance that enough contiguous
blocks cannot be found to allocate or extend the file.
This effort is probably worthwhile only if sequential access has high
priority.
• Divide an Indexed file into areas, segregating the upper index levels into
physically compact and contiguous sets of blocks.

8.4 Other Optimizations
You can improve the environment where your RMS-11 task runs with data
caching and by:
• allocating more resources to the task
• improving disk usage

Common Optimization Techniques

8-29

8.4.1 Data Caching
Data Caching, an option during RSTS/E system generation, uses the Extended Buffer Pool (XBUF) to save file blocks in memory. This facility should
provide:
• Noticeable performance improvement for applications sequentially accessing Sequential and Relative files.
• Some performance improvement for applications randomly accessing indexed files. A critical factor is the relationship between bucket sizes and
cache clustersizes.
See the RSTS/E System Manager's Guide for more details.

8.4.2 Allocating More Resources to the Task

'~;

You can improve the performance of a task by giving it more of the system to
use, more CPU time, more memory, and so on. Of course, you take those
resources away from other jobs, unless the system is not used to capacity.
The techniques for allocating system resources vary by operating system.
Each of the following techniques affects system throughput by changing the
number of 1/0 operations your task requires to complete its work.
• IAS and RSX- llM
Priorities
Checkpointing
Round-Robin Scheduling
Swapping
• RSTS/E
Swapping
Priorities

8.4.3 Disk Usage
You should consider the devices that store your data and task images when
you are optimizing the performance of an application. Efforts at improving
disk usage often result in significant increases in performance.
• Use the fastest disk drives available because the physical I/O operation
causes the most significant portion of 1/0 time.

8-30

Common Optimization Techniques

Minimize VO request overhead:
- Reduce VO request queues, using private packs if necessary.
- Assign exclusive use of the disk driver to your RMS-11 task.
If you must share the driver, use one that overlaps disk seeks .

., If your system has multiple disk drives which are not heavily used by other

people, spread an application's files, including disk-overlaid tasks, across
the devices. Thus, while a job runs, one disk device does not access more
than one file. At least, put data files on a disk device other than the one
containing a disk-overlaid task image.
Of course, if you are using the Resident Library, and not overlays, you do
not consider the task file, unless your code or your language run-time facilities are overlaid.
o

Combine free blocks on a disk into one contiguous group using the DSC
utility. By eliminating fragmentation, you are increasing the chances that
file extents are contiguous, even if they are not requested that way. The
more contiguous the file, the less the disk head moves to access it.
However, this procedure changes file IDs.

Common Optimization Techniques

8-31

Chapter 9
RMIS-11 Utilities

TO IAS USERS

~·

You do not have all the functions described in this chapter
available via the DIGITAL Command Language interface.
However, you can use these functions if you:
• install RMS-11 Version 1.8 on your computer system
• use a terminal in M C:R mode
• employ the syntax shown in this manual
The RMS-11 utilities are provided as independent tasks to those who do not
have full access to RMS-11 functionality, either because they are not programmers or are programming in a higher level language (other than
MACR0-11).
The utilities normally installed are:
RMSBCK
The RMS-11 file back-up utility, RMSBCK, transfers any RMS-11 disk
file to another disk or to magnetic tape in a special format that cannot be
read and/or modified by a user program. These RMS-11 back-up files can
only be accessed by the restoring utility RMSRST. You may use
RMSBCK and RMSRST to guard against hardware failure or software
error by having your data backed up on another volume.
RMSCNV
The RMS-11 file conversion utility, RMSCNV, moves records between
two RMS-11 files of any organization or record format.

9-1

RMS DEF
The RMS-11 file definition utility, RMSDEF, creates RMS-11 files, defining their attributes in an interactive sequence of questions and requests
for data.

NOTE
RMSDFN, the first file definition utility, is still available, but
is considered to be less useful than RMSDEF. See Appendix D
for a discussion of the RMSDFN utility.
RM SD SP
The RMS-11 display utility, RMSDSP, lists the RMS-11 file attributes
and structural data or the names of RMS-11 back-up files on magnetic
tape.
RMSIFL
The RMS-11 Indexed file load utility, RMSIFL, builds an RMS-11 Indexed file using records from another RMS-11 file of any organization.
The utility uses techniques derived from the basic structure of Indexed
files, rather than the standard RMS-11 file and record operations used by
RMSCNV.
RMSRST
The RMS-11 file restoration utility, RMSRST, reads the back-up files
created by RMSBCK and restores them with form, structure, content, and
attributes identical to the original files.

Each utility is described in a section, from Purpose to Call and Termination
to Cautions. However, common items, such as command lines and error messages, appear later in this section. But first a discussion on the uses of these
utilities.

NOTE
• The RMS-11 utilities require ANSI-standard labels on magnetic tapes. You must mount a tape volume to inform your
operating system's file processor that you are accessing an
ANSI-standard tape. See Appendix F for more details on
RMS-ll's magnetic tape handling.
• The RMS-11 utilities do not support the use of RSTS/E
special account characters ($, !, %, _, #, @) in file
specifications.

9.0.1 Using The RMS-11 Utilities
This section describes some services you can perform with the RMS-11
utilities. Once you are familiar with these tasks, you will find other ways in
which they can help you.

9-2

RMS-11 Utilities

RMSBCK and RMSRST - Your data is important: it is the prime, if not the
only, reason you are using a computer system; it costs money to collect,
maintain, and process; it makes you money in many ways, both direct and
indirect. And it is important that you never lose this data for all of these
reasons.
Your data cannot be locked away all the time: like money, it must be active to
be useful and growing, and in this state of activity, it is vulnerable to the
chance of damage, by hardware,. software, or operators. Therefore you make
your data secure by making a copy of it and storing the copy away from the
day-to-day environment.
The RMSBCK and RMSRST utilities are the two aspects of data security.
You use RMSBCK on a daily, weekly, or monthly basis, depending on the
volatility of your data. And you use RMSRST when you replace lost or corrupted data; the utility brings your data back to the same integrity it had
when the file was backed up.
Example You are a wholesale distributor of auto parts. Everyday your operators process
around 600 orders, involving order entry, invoice processing, and accounts receivable, plus your purchasing transactions, including inventory control, order processing, and accounts receivable.
Your data changes hourly, but you decide that a daily back up is sufficient. Therefore, each night, your second-shift system operator initiates the back-up procedure;
some operating systems can do this automatically. Your data is copied to tape (or
disk) by RMSBCK. Each day you are assured that if something goes wrong, you can
restore the files to last night's status, using RMSRST.

RMS CNV -

The RMSCNV utility has a variety of uses:

• You can back up any RMS-11 files by converting them to magnetic tape
Sequential files; you restore the files by running RMSCNV or for Indexed
files, RMSIFL, in the other direction.
• You can print any RMS-11 file endowed with carriage return control by
converting it to a line printer or a Sequential Stream file that can be queued
to a printer.
NOTE
RMS-11 does not convert numeric or binary data to ASCII
re pre sen tati on.
• You can complete the restoration process begun by RMSRST, which cannot
re-establish areas or contiguousness:
-

Create the file with the attributes you want using RMSDEF.

Populate the file using R.M:SCNV (or RMSIFL, for an Indexed file) with
the single-area, noncontiguous file as input.

RMS-11 Utilities

9-3

• You can perform daily processing using a small file designed for that purpose. Then, at the end of the day, you can use RMSCNV to add those
records to a larger master file.

• You can enhance your transition to a total RMS-11 environment by converting ASCil Stream files to RMS-11 file organizations:
1.

Create the new file with RMSDEF.

2. If you are building a Relative or Indexed file, sort the non-RMS-11 file
by relative record number, if possible, or Primary Key, respectively.
3. Build the new file with RMSCNV.
Example You have an ASCII Stream file named BPLUS.DAT. You want to build a Relative file named NEWREL.DAT. Create the Relative file with RMSDEF, then call
RMSCNV, using this command string:

NEWREL.DAT/FO:REL=BPLUS.DAT

RMSDEF -

The RMSDEF utility has a variety of uses:

• Used with RMSCNV or RMSIFL, RMSDEF creates any target file you
want.
• Used with higher level language applications 1 , RMSDEF extends the file
creation capabilities, enabling you to use features that are not explicitly
defined in the languages.
Example

Null key values.

Example Areas.

RMSDSP - Use RMSDSP to see an RMS-11 file's attributes and structural
data. Then you can:
• create another file just like it with RMSDEF.
• see the kind of file your higher level language task created so that you can
match and enhance it with RMSDEF.
Example You built an application of BASIC-PLUS-2 tasks, but you want to define file
attributes not supported by the language. First, you run the task that creates your
file(s), and then you display the attributes of these files with RMSDSP, preferably using a summary list file. You then know the minimum attributes your
RMSDEFined file(s) must have to fit with your BASIC-PLUS-2 application. If
those attributes do not match, your tasks fail with a "File Attributes Not
Matched" error when they try to open the file(s).

9.0.2 Utility Conventions
Even though each of the RMS-11 utilities is a separate task, they share
conventions, formats, and techniques. This section covers those common
items.
1

9-4

Especially DIBOL since a DIBOL task cannot create an RMS-11 Relative or Indexed file.

RMS-11 Utilities

NOTE
The IAS and RSX-llM operating systems produce and support different versions of a file-name/file-type combination.
These versions are totally separate files and can have different
HMS-11 attributes.
Example A file named PAYROLL.DAT has five versions in account
[303,351] on device DBO:
• DB0:[303,351JPAYROLL.DAT; 0 is a Sequential file with
fixed-length records.
• DB0:[303,351JPAYROLL.DAT; 1 is a Relative file with
variable-length records.
Example You are trying to convert one Relative file into another. You use
the command line:

CNV REL2.DAT = REL1.DAT
Because you did not include the FO switch, RMSCNV creates a
Sequential file named REL2.DAT with the next higher version
number. The utility then runs successfully, apparently obeying
your command string. The result is two versions of REL2.DAT,
one a Relative file, the other a Sequential file.

9.0.2.1 Command vs. Interactive -

Most of the RMS-11 utilities operate via
command string interpretation while another interacts with you, performing
its function in a sequence of questions and requests for data:

~·

Command

Interactive

RMSBCK
RMSCNV
RMSDSP
RMSRST
RMSIFL

RMS DEF

Beyond their user interface, the command and interactive utilities differ in:
HELP Messages
Command utilities each have one HELP message. Interactive utilities
have a unique HELP message for each question or request for data.
Error Messages
Command utilities share a set of error messages, some common to all,
some unique to a particular utility. These error messages are generated in
a standard format (discussed in Section 9.0.2.6). Interactive utilities, however, have separate messages for each question or request for data. All
error messages are listed in Appendix E.

RMS-11 Utilities

9-5

9.0.2.2 Installed vs. Unlnstalled

CONVENTION
The cover term installed in this manual has the
same meaning as the following system-specific
terms:

System

Term

IAS
RSTS/E

installed
invocation by Concise Command
Language (CCL)
installed

RSX-llM

RMS-11 utilities can be run whether or not they are installed. Only the
manner of invocation changes and both calls are described in each utility's
section. Your operating system user's guide explains the installation
procedure.
However, one invocation capability is not described in each section because it
is unique to the RSX-llM/IAS operating systems. Whether the utility is
installed or not, the command string may include the switch:
/UIC=[act nbr]
With this switch in its command string, each RMS-11 utility runs under the
account specified by [act nbr] until the task is terminated.
9.0.2.3 Indirect Flies -An indirect file contains a sequence of commands that
can be interpreted by a task, usually a system-supplied task such as a utility.
These commands appear in the indirect file exactly as they are entered from
your terminal. As such, they can be command strings or answers to questions.

The commands in an indirect file are executed when the file's name ils provided to a utility preceded by an at sign (@). RMS-11 utilities assume an
extension of . CMD if none is included in the filespec.
Example An indirect file contains a series of RMSBCK command strings. To invoke such an
indirect file, you enter the command:

BCK @BCKCMDS.CMD or BCK @BCKCMDS

As an installed task, RMSBCK accesses the file BCKCMDS.CMD, exiecutes
the commands, and returns control to the operating system.
RSX-llM allows you to use indirect files directly from the MCR environment
without first invoking a utility. Such an indirect file may contain command
lines for more than one utility and is called by entering only the file specification preceded by an at sign (@):
@INDIRECT.CMD

9-6

RMS-11 Utilities

For complete information on how to use indirect files, refer to your system
documentation.
You should type lengthy command
strings in segments, each on a separate line, all but the last ending in a
hyphen (-).The hyphen causes the utility to reprompt without attempting to
execute the command string.
9.0.2.4 Command String Continuation -

In particular, RSX-llM automatically terminates a command line at the end
of the terminal's input buffer. The utility then attempts to execute a command you may not have completed.
Example

>BC K DK 0 : * , *IQ U =FI L 1 , DAT t DK 1 : [ 5 0 t 1 FI L 2 , DAT t BCK>FIL3,*/CD:17-SEP-78

>
Example BCK
BCK> DKO:*+*/QU=FIL1 +DAT tDK1: [50 t1BCK>JFIL2+DAT tFIL3+*/CD: 17-SEP-78
B CK>·'· Z

Read>'

You can continue a command line at any point within the line; however, no
command line, regardless of continuations, may contain more than 158 characters, not including continuation hyphens.
9.0.2.5 Patch Level - Each RMS-11 utility contains an identifier that specifies software version number and the patch level of the utility itself. The
utility prints this identifier on your terminal in response to the /ID switch, in
the form:
~'ERS ION

1, Bnn

where nn indicates the patch level of the utility you are running.
Immediately after the RMS-11 installation process, whether it is included in
the system installation or not, each utility has a patch level of 00. Thereafter,
DIGITAL notifies you if a patch is necessary. Included in the notice are:
• the level of the utility before the patch is applied
• the patch procedure
• the patch level of the utility after the process is complete
9.0.2.6 Command Utility Error Messages - When RMS-11 command utilities

encounter an error condition during execution, they print error messages at
your terminal.
The two primary types of errors are fatal and nonfat al. Appendix E lists all
error messages.

RMS-11 Utilities

9-7

Fatal Error Messages - When an error requires the termination of processing, RMS-11 command utilities print a fatal error message, in the form:
?ut l -

message

where:
?

indicates a fatal error occurred

utl

is the three-character utility name

message

briefly describes the error

Example ?CNV -- ILLEGAL NUMBER OF INPUT FILES
Example ?DFN - ILLEGAL RMS RECORD FORMAT

Nonfatal Error Messages - When a nonfatal error occurs, RMS-11 command utilities print an error message and continue processing. The nonfatal
error message takes the form:
utl -

message

where:
utl

is the three-character utility name

message

briefly describes the error

Example The following command string is issued to RMSRST:

DKO:*+*=FIL1+DAT/QU,DK1:[50t1JFIL2+DAT1FIL3+*/CD:17-SEP-78
If the utility cannot find the file FILl.DAT in the default account, it prints the
nonfatal error message:

RST -- FILE NOT FOUND - SYO: C200tl1 JFIL1. DAT; 1
Since additional input files were specified, RMSRST continues processing.

Crash Dump - If RMS-11 command utilities encounter a situation they
cannot recover from, they produce crash dump information in the following
form:
************RMS-11 UTILITIES DAMAGE ASSESSMENT ROUTINE***

text

PLEASE DETACH AND SUBMIT WITH YOUR SPR

If you are using a CRT terminal, do not copy the information. If you are using
a hard-copy terminal, do include the output with a Software Performance
Report (SPR).

9-8

RMS-11 Utilities

Submit an SPR to the proper DIGITAL office. Put in the following information:
l.

The command line that led to the crash dump.

2. An attribute listing of the file(s) involved. Use RMSDSP with the FU
switch.

:3. A copy of the file(s) involved. If you cannot send a tape, use RMSCNV to
print the file on a line printer.
4. Your operating system and version.
5. Any programming languages used to process the file(s) involved and the
version you are using.
6. Other information pertinent to the situation.

9.0.3 Documentation Conventions
The following conventions adopted in this chapter make documenting and
understanding the RMS-11 utilities easier.
Description of Syntax
Convention

Definition

REQUIRED WORDS

All words that are in uppercase must be in the command in the
places shown.

user-defined words

When you use the command, you supply actual names for all words
that are in lowercase. See the next section for specific user-defined
words.

[may be used]

You may, but do not have to use words included within brackets. The
characters .. ] indicate that the series continues to include all
possibilities.

Punctuation

You must use all punctuation marks included in the command.

The following symbols for user-defined
words have the specific meanings indicated:
Specific User-Defined Words -

Convention

Definition

nnnnn

a decimal number; the number of ns indicates the maximum number of digits
that can be entered.

value

an argument selected from a specific and limited set of arguments; the number
of letters that appear from the word value indicate the maximum number of
characters that can be entered. For example, v, val, and value mean that one,
three, and five characters are allowed.

dd-mmm-yy

a date; for example, 17-0CT-78

RMS-11 Utilities

9-9

Symbols
Definition

Convention
CR

the Carriage Return character (ASCII 015 8)

00)

the RETURN key on your terminal; you press the RETURN key at this point
in the operation you are performing.

the Line Feed character (ASCII 012 8 )

(IT)

the LINE FEED key on your terminal; you press the LINE FEED key at this
point in the operation you are performing.

tTRL/O

the combination of the CTRL key and the C key. You press both keys at the
same time, and "· C appears on the terminal.

@'!im

the combination of the CTRL key and the Z key. You press both keys at the
same time, and "· Z appears on the terminal.

Prints vs. Types - When the software outputs characters on a CRT or hardcopy device, the software is said to print.
When you must input characters on a terminal, you are asked to type those
characters.

9.1

RMSBCK Command Utility

9.1.1 Purpose
You ensure your data from hardware failure or software error with the
RMSBCK command utility. Each time you invoke this utility, specially formatted back-up copies of the specified files are placed on a back-up medium.
Thereafter, if the original files are lost or damaged, you can use the RMSRST
utility and these back-up copies to replace the files.

9.1.2 Effect
RMSBCK copies standard RMS-11 files from one medium to another (disk to
disk or disk to tape), translating the data into a special back-up format. The
back-up copy contains the· source file's attributes: account number, creation
and revision dates, protection code, and so on. However, they do not include
placement control instructions; a file cannot be restored to the same physical
place on a device that it was backed up from. See also "RMSRST," Section
9.7.
Back-up files can only be accessed properly by the RMSRST utility. User
programs therefore cannot change back-up data.
RMSBCK only uses magnetic tapes with ANSI-standard labels. However, the
back-up data written by the utility between the labels does not comply with
ANSI standards.

9-10

RMS-11 Utilities

RMSBCK provides four processing features:
Explicit and Implicit File Selection
The RMSBCK command string permits one or more input filespecs with
or without wild card characters:
• Filespecs without wild card characters explicitly identify input files.
• Filespecs with wild card characters implicitly identify a collection of
files. When using wild card characters, you can restrict the number of
files selected for processing by including a date switch. A date switch
causes RMSBCK to examine either the creation date or the revision
date of each file selected as a result of the wild cards: the utility backs
up a file only if the internal file date conforms to the date you specified.
Data Integrity Checks
RMSBCK can check data integrity extensively as each back-up file is
created. You can direct the utility either to read back file contents after
they are written into the back-up file or to both read back and check
file contents on a byte-by-byte basis after they are written. RMSBCK
automatically retries read errors if processing continues to a normal
termination.
These integrity checks allow you to choose how reliable the back-up files
are:
• You can just rely on software and hardware accuracy and back up your
files in minimum time.
• You can verify that you can read the back-up files after they are created,
adding the read time to the minimum time.
• You can guarantee the back-up files can be read and that they exactly
match the source files, adding both read and compare times to the
minimum.
Extended Diagnostic Messages
During file processing, either with or without data integrity checking,
RMSBCK provides an extended diagnostic message capability known as
Query mode. When the utility encounters an error condition described in
Section 9.1.4.2, the utility prints a diagnostic message at your terminal.
Immediately following the message, RMSBCK prints the query CONTINUE
< Y '"' >. Each query requires you to type a response (Y or N) indicating
whether the utility is to continue or discontinue processing. With this
procedure, you can ensure that RMSBCK does not terminate the processing of a collection of files when errors are encountered in the processing
of any one file. See the QU switch in Section 9.1.4.2 for examples of
diagnostic messages.
Optionally, you can disable the Query mode. The utility prints a diagnostic message if it encounters an error condition, but does not give you the option
of continuing processing. Rather, the utility terminates automatically.

RMS-11 Utilities

9-11

Summary Listing
You can specify that RMSBCK summarize file processing, data integrity
checking, and error messages. RMSBCK lists this summary on your terminal or writes it into a file, creating a record of the utility's processing.
See the SL switch in Section 9.1.4.2 for a sample of the summary.

9.1.3 Call and Termination
9.1.3.1

Installed Utility

BCK [command string]
If you include a command string, the utility attempts to execute it and then
returns control to the current keyboard monitor or command interpreter.
If you do not specify a command string, RMSBCK assumes control of the user
interface and prints the prompt:
BCK>

You may type a command string or (CTRL/z) to terminate the utility. When
RMSBCK has executed a command string, it reprints the prompt.
9.1.3.2 Unlnstalled Utlllty

RUN $RMSBCK
RMSBCK assumes control of the user interface and prints the prompt:
BCK>

You may type a command string or (CTRL/ Z) to terminate the utility. When
RMSBCK has executed a command string, it reprints the prompt.

9.1.4 Command String
9.1.4.1

General Form

outfile[/switch] .. ]=infile[/switch] .. ][,infile[/switch] .. ]... ]
where:

out file

is the filespec of a back-up file to be created by the RMSBCK
utility:
• If the back-up medium is a disk, both the file-name and extension must be wild card characters(*.*): the file-names and extensions of the back-up files will be the same as the associated input
files. The version can be omitted or can be a wild card. In both
instances, RMSBCK uses the version of the associated input file.

9-12

RMS-11 Utilities

J_~

If you do not use wild card characters, RMSBCK prints the following error message and terminates execution of the command
string:
?BCK -- NO RENAME ALLOWED

• If the back-up medium is magnetic tape, the back-up files produced by a command string are contained within a single file on
the tape. This file is known as a container file since it contains
back-up files.

You must name the output container file in the command string,
but the name need not correspond to the name of any input file.
The version can be omitted: a value of 0 is used.
If you do not name the container file, RMSBCK prints the following error message and terminates execution of the command
string:
?BCK -- EHPLICIT CONTAINER FILE NAME NECESSARY

NOTE
You must use the container file-name in the infile
specification for RMSRST.
RMSBCK does not prevent multiple container files with the
same name on the same volume. Therefore, you should specify
unique file-names, extensions, and version numbers for output
container files written on the same volume.
infile

is the filespec of a disk file of which a back-up copy is to be created.
You can use wild card characters to specify implicitly a collection of
files.
If you specify a nondisk device in the infile specification, RMSBCK
prints the following error message and terminates execution of the
command string:
?BCK

-- ILLEGAL DEl.JICE - dun:

where: dun
switch

is the device name and number you used.

may be one code shown in Table 9-1 and described in Sections
9.1.4.2 through 9.1.4.4.

NOTE
A command string may also consist of the word "HELP" or a
question mark(?). RMSBCK responds with a HELP message.

RMS-11 Utilities

9-13

Table 9-1: RMSBCK Utility Switches
Type

Switch

Description

Default
Process

String

? or HELP

Print HELP message.

No help.

Global

/ID

Identity current version of
the RMSBCK utility.

No id.

/QU

Enable Query mode.

/SL[:filespec l

Provide summary listing
(in file, if specified).

No summary.

/RA*

Read after writing.

NORA

/RC*

Check after writing.

NORC

/RW

Rewind magnetic tape before writing.

NORW

/SU

Supersede old files.

NOSU

/CD:dd-mon-yy[:u]

Back-up files based on ereation date.

No date checking.

/RD:dd-mon-yy[:u]

Back-up files based on revision date.

No date checking.

Outfile

Infile

* The RA and RC switches are not available on RSTS/E. That operating system does not allow
the RMSBCK task to rewind a magnetic tape device to beginning of file after the utility
accesses that file for writing.

9. 1.4. 2 Global Switches

• ID causes RMSBCK to print its current version number, in the form:
BCK - - l.IERSIDN 1. Bnn

where:

nn is the patch level of the utility (see "Patch Level," Section 9.0.2).
This switch may appear alone as a command string.
• QU enables the Query mode. When the Query mode is enabled, the
RMSBCK utility allows you to continue or terminate processing when one
of the following occurs:
- A read error on an input file.
- A read-after-write error on an output file when the RA switch is specified.
- A check-after-write error when the RC switch is specified.
- The table allocated internally for data integrity checks or automatic retry
of read errors is full.
When one of these errors occurs, the utility prints a diagnostic message
specifying the type of error and the name of the file being processed. If you
answer Y (yes), RMSBCK continues processing the current file and com-

9-14

RMS-11 Utilities

mand string. If you answer N (no), the utility terminates processing immediately, bypassing the rest of the command string.
Example BC K - - CHECK AFT ER WRITE ERROR ON 0 UTPUT FI LE - filespec
t,IBN vbnl TO vbn2.

Example

CONTINUE <Y tN)?

tlCK -- INTEGRITY CHECK TABLE FULL.

CONTINUE <YtN>?

You can disable the Query mode through the NOQU switch. The RMSBCK
utility prints a diagnostic message when it encounters an error condition,
but does not give you the option of continuing processing.

Default If you specify no version of the QU switch, Query mode is enabled.
• SL[:filespec] causes RMSBCK to summarize activity during the execution
of the command string. This summary contains:
- the command string
- the names of files successfully backed up
- the names of files not backed up, with associated error messages
- diagnostic messages produced in Query mode
- a summary of any input errors and of errors remaining in output back-up
files following automatic retry of data integrity checks (when you specify
RA or RC).
If you do not specify an argument (filespec) with the SL switch, RMSBCK
prints the summary listing on your terminal. However, if you supply a
filespec, the utility creates the specified file and writes the summary listing
as the contents of the file.
Example
RMSBCK - t,IERSION 1.800 27-JUN-187816:55:14
DB1:*•*/SL:BCKUP.LOG=DBO:*.MAC
BCK
BCK
BCK
BCK
BCK
BCK
BCK
BCK

FILE PROCESSING COMPLETE - DBO:C303t351JRMSPRE.MAC; 74
FILE PROCESSING COMPLETE - DBO:C303t351JUTLMLB.MAC; 146
FILE ALREADY E}<ISTS - DB1:C303t351JDSRMS.MAC; 1
FI LE ALREADY E}< I STS - DB 1: [ 303 t351 J t,IEHT. MAC; 1
FILE PROCESSING COMPLETE - DBO:C303t351JBLKIO.MAC; 12
FI LE ACCESS ERROR ON DBO: [303 t351 JB .MAC; 2 tERROR CODE= 177734
FILE ALREADY EHISTS - DBi: [303 t351JWR.; 10
FILE PROCESSING COMPLETE - DBO: [303 t351 JSWTABL •MAC; 55

Default If you specify no version of the SL switch, RMSBCK prints messages on your terminal only if it encounters a fatal or nonfatal
error condition.
9.1.4.3 Outflle Switches

• RA directs RMSBCK to read the back-up file after writing it. The utility
reads back each block of the back-up file: if the device hardware detects a
read error, RMSBCK's response depends on whether Query mode is enabled
or not. See QU under "Global Switches," Section 9.1.4.2.

RMS-11 Utilities

9-15

NOTE
You cannot use the RA and RC switches in the same command string. RMSBCK prints the following error message
and terminates execution of the command string.
?BCK -- CONFLICTING OPTION - lsw

where:
is RA or RC depending on which is second in the command string.

Default If you do not include the RA switch, RMSBCK does no read-afterwri te data integrity checking.
• RC directs RMSBCK to check the back-up file after writing it. The utility
reads each virtual or tape block of the back-up file back into memory. At
the same time, it reads into memory the corresponding virtual block(s) of
the source file. RMSBCK then compares the contents of the two buffers. If
an error occurs during this process (a hardware read error or a mismatch
between the contents of the two buffers), RMSBCK's response depends on
whether the Query mode is enabled. See QU under "Global Switches,"
Section 9.1.4.2.
See note under RA in this section.

Default If you do not include the RC switch, RMSBCK does not check the
back-up file after writing it.
• RW directs RMSBCK to rewind a magnetic tape before writing back-up
container files. Rewinding logically deletes files existing on the tape.
If you specify RW when the back-up medium is a disk, RMSBCK prints the
following error message and terminates execution of the command string:
?BCK--ILLOGICALUSEOFOPTION -

/RW

Default If you do not include the RW switch, RMSBCK adds new container files to the logical end of a tape.
• SU causes RMSBCK to supersede files in the output account with the same
file-name, extension, and version as a new back-up file.
This switch applies only if the back-up medium is a disk. If SU is specified
for magnetic tape, RMSBCK prints the following error message and terminates execution of the command string:
?BCK -- ILLOGICAL USE OF OPTION -

/SU

Default If you do not include the SU switch, RMSBCK does not supersede
files in the output account. If the utility encounters a file with the
same file-name, extension, and version as an input file, it prints
the following nonfatal error message and continues processing.
BCK - - F I L E ALR E ADY E >{ I S TS -

9-16

RMS-11 Utilities

f ilespec

""·

9.1.4.4 lnflle Switches
• CD:dd-mon-yy[:v] causes RMSBCK to back up files based on their creation

dates. You can combine this switch with wild card characters to identify a
group of files, depending on the value of the switch argument v:
Valuu

Selection
Those files that satisfy the wild card specification and that were:

None
A
B

created on the specified date
created after the specified date
created before the specified date

RMSBCK treats a file with no or a null creation date as though that file
were created before January, 1900.

Default If you do not include the CD switch, RMSBCK applies no date
criterion when it selects files for back up (unless you specified the
RD switch).
ct

RD:dd-mon-yy[:v] causes RMSBCK to back up files based on their revision

dates, that is, the dates they were last accessed. You can combine this_
switch with wild card characters to identify a group of files, depending on
the value of the switch argument v:
Value

Selection
Those files that satisfy the wild card specification and that were:

None
A
B

revised on the specified date
revised after the specified date
revised before the specified date

RMSBCK treats a file with no or a null revision date as though that file
were revised before January, 1900.

Default If you do not include the RD switch, RMSBCK applies no date
criterion when it selects files for back up (unless you have specified the CD switch).
9.1.4.5 Command String Examples
Ct

MTO:ALPHA.BKP;l/RC=ALPHA.DAT
RMSBCK writes a back-up copy of the file ALPHA.DAT on the ANSIlabeled magnetic tape mounted on device MTO: The container file created
on the tape is named ALPHA.BKP ;1. The utility checks each block of the
output back-up file after it is written and since Query mode is enabled by
default, reports a qualifying error condition via a diagnostic message and a
query. You must indicate with a Y or N whether or not processing should
continue.

MTl:SAFE.BKP=*.*/QU/SL
RMSBCK backs up on ANSI-labeled magnetic tape the highest version of
all files in the default account on SY: into a container file named

RMS-11 Utilities

9-17

SAFE.BKP; O. Query mode is explicitly enabled; therefore, a qualifying
error condition causes RMSBCK to output a diagnostic message and wait
for your response whether to continue or terminate. While RMSBCK
processes the input files, it prints a summary listing on your terminal,
specifying the names of all files.
• MTO:MASTER.BKP=*.DAT/CD:31-DEC-77:B/SL:BACKUP.LST
RMSBCK backs up onto ANSI-labeled magnetic tape the highest versions
of all files in the default account on SY: that were created before December
31, 1977 and have an extension of DAT. Query mode is enabled by default.
The utility also creates a summary listing file named BACKUP .LST in the
default account on SY:. The file contains a listing of the names of the
files processed along with copies of diagnostic messages and queries output
during processing.
• DKl:[l00,101* .*; * /SU=DKl:[l00,20]*.*; * /RD:Ol-JUN-78:A/NOQU
RMSBCK backs up all files under account number [100,20] on DKl: that
were accessed after June 1, 1978. The back-up copies are written into account [100,10] on the same disk (DKl:), superseding files with the same filenames, extensions, and versions. Query mode is explicitly disabled; therefore, a qualifying error causes RMSBCK to print a diagnostic message and
terminate processing.

~':

9.1.5 Cautions
• RMSBCK cannot back up a system disk as a bootable volume.
• The back-up medium must be on-line and mounted before you issue a
command string.
• When the back-up medium is a disk, you cannot rename output files. That
is, the name of each back-up file will be the same as the name of the input
file.

• When the back-up medium is magnetic tape, you can use only ANSIlabeled tapes.
• When the back-up medium is magnetic tape, RMSBCK uses 2048-character
blocks. You cannot change this. If you want to use larger blocks, use
RMSCNV with the BL switch to convert the input file to a Sequential tape
file.
Example You have an 86,000-block Indexed file (400,000 82-byte records). The back-up
copy of this file requires at least two magnetic tape volumes. However, if you
convert the file to tape with 4096-character blocks, you need only one 2400-foot
tape volume. You can restore the file quickly with RMSIFL.

• If you include in a command string one name of a file listed in several
accounts via the PIP /ENTER switch, RMSBCK copies the entire file even

9-18

RMS-11 Utilities

though it exists in another account. But when you restore the file with
RMSRST, the link established by PIP is broken.
Example You have a file of customer names--and -a-dtlresses stored in account [30,1], but it
has different names in different accounts:

[10,lJCLIENTS.DAT
[20,lJCUSTOMERS.DAT
[30, lJACCTSRECV .DAT
These filespecs point to the one file in account [30,1].
If you back up [20,lJCUSTOMERS.DAT, RMSBCK copies the file out of [30,lJ.

However, when you restore the file with RMSRST, it is written into account
[20,1], and the restored file is no longer linked with the other names, that is,
[10,lJCLIENTS.DAT and [30,lJACCTSRCV.DAT.

• RMSBCK handles unused blocks in a file according to the file organization:
-

If the file is Sequential or Relative, RMSBCK does not copy the unused

blocks.
-

If the file is Indexed, RMSBCK copies the unused blocks.

• Do not use either the infile or outfile specification as an argument in the SL
switch.

9.2 RMSCNV Command Utility
9.2.1

Purpose

The RMSCNV command utility provides a versatile mechanism for performing any of the following functions:
• creating a new Sequential file from the records of an existing Sequential,
Relative, or Indexed file.
• superseding an existing Sequential file with the records of another Sequential file or the records of a Relative or Indexed file.
• appending to a Sequential file the records of another Sequential file or the
records of a Relative or Indexed file.
• writing into an existing Relative or Indexed file the records of another file of
any organization.

9.2.2 Effect
All the functions performed by RMSCNV involve the movement of records
from one file to another file. The utility reads records from the specified input
file and writes them into the specified output file. The manner in which
records are read and written depends on the file organizations of both the
input and output files and on switches you specify in the command string.

RMS-11 Utilities

9-19

Output File
Organization

Processing

SEQUENTIAL

RMSCNV:
• creates the output file with the attributes of the input file if the
append (AP) switch is not specified. The utility then reads records
from the input file and writes them sequentially into the new output
file. If the output file already exists:
- RMSCNV creates the next higher version of the file.
- RMSCNV terminates with the following error message:
?CNl.,I -- FILE ALREADY E><ISTS

• supersedes an existing file if you specify the SU switch. The utility
reads records from the input file and writes them sequentially over
the records already in the output file, starting with the first record
position in the file. RMSCNV creates the file with the attributes of
the input file if it does not exist.
• appends records onto an existing file if the AP switch is used. The
utility reads records from the input file and writes them sequentially
into the output file, starting with the record position following the
last record already in the file. RMSCNV terminates with the following error message if the output file does not exist:
?CNl.1 -- FILE NOT FOUND -

RELATIVE

filespec

RMSCNV reads records from the input file and writes them into successive record cells of the output file, beginning with cell 1. If the utility
encounters a cell containing a record, it terminates with a FA TA L RMS
ERROR message (STS = 1752203). All records written to that point are
still in the file. You should examine the input and output files to determine the extent of processing.
RMSCNV terminates with the following error message if the output file
does not exist:
?CNl.1 - - FI LE NOT FOUND

INDEXED

9-20

filespec

RMSCNV reads each record from the input file, then applies the output file's record structure, that is, key placement within the record, to
the data; this structure is an attribute of the output file and is not
dependent on the input file organization. The record is then inserted
into the output file on the basis of the value found in the Primary Key
field. Finally, RMSCNV updates the output file's Primary and any
Alternate indexes to reflect the presence of the record. RM SC NV terminates with the following error message if the output file does not exist:
?CNl.1 -- FILE NOT FOUND

Input File
Organization

filespec

Processing

SEQUENTIAL

RMSCNV reads the records in the Sequential Access Mode, starting
with the first record in the file.

RELATIVE

RMSCNV reads the records in the Sequential Access Mode, starting
with record cell 1.

INDEXED

RMSCNV reads the records in the Sequential Access Mode, following
the key of reference specified in the command string (the Primary Key
is default).

RMS-11 Utilities

9.2.3 Call and Termination
9.2.3.1

Permanently Installed Utillty

CNV [command string]
If you include a command string, the utility attempts to execute it and then

returns control to the current keyboard monitor or command interpreter.
If you do not specify a command string, RMSCNV assumes control of the user

interface and prints the prompt:
CNt.J>

You may type a command string or (CTPL/ ZJ to terminate the utility. When
RMSCNV has executed a command string, it reprints the prompt.
9.2.3.2 Uninstalled Utility

RUN $RMSCNV
RMSCNV assumes control of the user interface and prints the prompt:
CNt.J>

You may type a command string or (CTPL/ZJ to terminate the utility. When
RMSCNV has executed a command string, it reprints the prompt.
I~

9.2.4 Command String
9.2.4.1

General Form

outfile[/switch] .. ]=infile[/switch]
where:

outfile

is the filespec of the output file that is to receive the records of the
input file. Wild card characters are not permitted in any field of the
specification. The default version for Sequential files with the AP
or SU switch, Relative, and Indexed files is the highest version.
When creating a Sequential file, RMSCNV uses the highest version
number plus 1.

infile

is the filespec of the input file that is the source of records to be
written to the output file. Wild card characters cannot appear in
any field of this filespec.

switch

may be a code shown in Table 9-2 and described in Sections 9.2.4.2
through 9.2.4.4.

NOTE

I'-'

A command string may also consist of the word "HELP" or a
question mark(?). RMSCNV responds with a HELP message.

RMS-11 Utilities

9-21

Table 9-2: RMSCNV Utility Switches
.~,

Type

Switch

String

? or HELP

Print HELP message.

Global

/ID

Identify current version
RMSCNV utility.

/SL[:filespec]

Provide summary listing (in file, if
specified).

No summary.

/AP

Append records to Sequential file.

No append.

/BL[:nnnn]

Set magnetic tape block size.

512 bytes.

!FO:val

File organization.

Sequential file.

/LO

Follow fill numbers when writing Indexed file.

Fill buckets.

/MA

Use Mass Insert and sequential put
operations to optimize performance.

No Mass Insert; random put operations.

/PD[:[#Jxl

Pad input records to output record
length

Abort if different
lengths.

/SU

Supersede existing Sequential file.

No supersede.

trR

Truncate input records to output record length

Abort if
lengths.

/WF

Write or read fixed control area.

Ignore fixed control
area.

/KR:n

Key of reference number.

Primary key (n=O).

Outfile

lnfile

Description

Default
Process

No help.
of

the

No id.

different

9.2.4.2 Global Switches

• ID causes RMSCNV to print its current version number, in the form:
CNIJ - - IJERS ION - - 1 • Bnn

where:
nn

is the patch level of the utility (see "Patch Level," Section 9.0.2).

This switch may appear alone as a command line.
• SL [:filespec] directs RMSCNV to summarize processing. If a file specification is not included in the switch, RMSCNV prints the summary on your
terminal. If you specify a file, the utility creates it and writes the summary
listing into it.
The summary includes:

9-22

the command string

copies of error messages produced during execution of the utility.

RMS-11 Utilities

an indication that input records could not be written into an output
Indexed file because duplicate keys (for one or more key fields) were not
permitted:
If the summary appears on your terminal, the indicator is the message:
SOME DUPLICATE RECORDS NOT WRITTEN

If the summary is written into a file, RMSCNV supplies the indicator
Du P R c D =, followed by the f~rst 72 characters of the record that could not
be written. Each record left out of the output file is shown in the summary listing file.

Default If you do not include the SL switch, RMSCNV prints messages on
your terminal only when it encounters a fatal or nonfatal error.
9.2.4.3 Outflle Switches

• AP directs RMSCNV to append records to an existing Sequential file.
RMSCNV adds records read from the input file to the end of the output file.
AP and SU cannot be specified in the same command string.

NOTE
If the output file is not sequentially organized, RMSCNV
ignores the AP switch.

Default If you do not include the AP switch, RMSCNV's action depends
on the presence of the SU switch:
•

If you specified SU, the utility obeys the switch.

•

If you did not specify SU, RMSCNV:

- creates the next higher version of the file
- terminates with the following error message:
?CNl.l - - FI LE ALREADY EH I STS

• BL [:nnnn] directs RMSCNV to control the physical block size of the output
file when it is being created on magnetic tape. Any size specified by nnnn
must be from 18 through 8192 characters and should be a multiple of four
(otherwise, RMS-11 rounds it to the next higher multiple of four).

Default If you do not include the BL switch, RMSCNV uses a tape block
size of 512 characters.
• FO:val specifies the organization of the output file as one of the following:
SEQ -- Sequential file
REL -- Relative file
IDX - Indexed file

Default If you do not include the FO switch, RMSCNV assumes Sequential organization. The utility responds as described in "Effect,"
Section 9 .2.2.

RMS-11 Utilities

9-23

• LO directs RMSCNV to write records into the buckets of an Indexed file
according to the fill numbers established when the file was created. See
"Data Allocation" in RMSDEF, Section 9.3.4.6 and "Fill Number" in
Section 6.7.2.1.
Default If you do not include the LO switch, RMSCNV inserts as many
records into each bucket as possible.
• MA directs RMSCNV to activate the RMS-11 Mass Insert I/0 technique
(see Section 6.7 .2.2) and then use sequential put operations to insert records
into the outfile.
If you use the MA switch, the infile:
• must be sorted into ascending order according to the Primary Key field
of the output file. Otherwise, RMSCNV prints the following error
message and terminates processing.
?CNt.1 - - INPUT RECD RDS NOT IN ASCEND I NG ORDER

• should contain records with Primary Key values greater than the Primary Key value of any record already in the file. The Mass Insert
technique requires that records be inserted at logical end-of-file. If the
file is empty, there are no Primary Key values to exceed.
If the input records do not meet this requirement, RMSCNV continues
to process them. However, the utility does not achieve the improved
performance that is otherwise expected from Mass Insert.
Example You have two Indexed files, !DX.IN and !DX.OUT. The Primary Key of !DX.IN
is defined as bytes 0 through 15 of each record. The Primary Key of !DX.OUT is
defined as bytes 0 through 8 of each record. Since RMS-11 orders records in
Indexed files according to the Primary Key value, you can convert !DX.IN into
!DX.OUT using the MA switch with the default key of reference.
Note that if the Primary Key of !DX.OUT coincides with an Alternate key of
IDX.IN file, you can still use the MA switch as long as you specify the proper
Alternate Key in the KR switch (see Section 9.2.4.4).

Default If you do not include the MA switch, RMSCNV uses random put
operations to insert each record into the output file.
• PD [:[#Jx] directs RMSCNV to pad records read from the input file to the
output file's record length before writing them to the output file. Padding
character is specified as follows:
Switch

Character

PD
PD:x
PD:x

x is ASCII A-Z, 0-9, or special character except #, ? , and @

NULL (byte value 000 8 )
x is octal number 000-377 (43 for#, 77 for?, and 100 for@)

You use the PD switch only when the output file specifies fixed-length
records.

Default If you do not include the PD switch, and the input records are
shorter than the output file's records may be, RMSCNV terminates with the following message:
?CNl.I -- INPUT AND OUTPUT RECORD SIZES DO NOT CORRESPOND

9-24

RMS-11 Utilities

~."'r'.._

• SU directs RMSCNV to supersede an existing Sequential file. RMSCNV
utility supersedes a file in the output account with the same file-name,
extension, and version number. AP and SU cannot be specified in the same
command line.

Default If you do not include the SU switch, RMSCNV's action depends
on the presence of the AP switch:
•

If you specified AP, the utility obeys the switch.

•

If you did not specify AP, RMSCNV:

- creates the next higher version of the file
- terminates with the following error message:
?CNl.J - - FI LE ALREADY E}·{ I STS

• TR directs RMSCNV to truncate records read from the input file to
the output file's record length before writing them to the output file. The
trailing bytes of the records are truncated.

Default If you do not include the TR switch, and the input records are
longer than the input file's record may be, RMSCNV terminates
with the following error message:
?CNl.J -- INPUT AND OUTPUT RECORD SIZES DO NOT CORRESPOND

• WF directs RMSCNV to handle variable-with-fixed-control (VFC) format
records in either the input or output file. The utility processes the possible
combinations as follows:
ln1mt

VFC

Output

VFC

Processing
If the fixed control areas are the same size, RMSCNV performs a straightforward copy process; if the areas are not identical in length, the utility

terminates without converting the input file, printing the message:
?CNl.1 - - IN PUT AND OUT PUT F I>{ED CONTROL HEADER SIZES DO NOT CORRESPOND

VFC

FIX

VAR
STM

FIX

VAR
STM

VFC

The fixed control area of each input record is written as the first n bytes of
each output record, where n equals the size of the fixed control area. The
variable portion of the input record completes the output record.
The first n bytes of each input record are written into the fixed control
area of each output record, where n is the length of the fixed control area.
The remaining data in the input record is written into the variable portion
of the output record.

If you include the WF switch and neither file specifies VFC records,

RMSCNV terminates with the following error message:
?CNl.1 -- ILLEGAL USE OF /WF WITH RECORD FORMAT

RMS-11 Utilities

9-25

Default If you do not include the WF switch and one of the files contains

VFC records, the fixed control area of each record is ignored:
Input

Output

Processing

VFC

any

Only the variable portion of each record is read from the
input file and written to the output file.

any

VFC

Data is read from the input file, but is written only into
the variable portion of each output record.

9.2.4.4 lnflle Switches

• KR: n indicates the key that establishes the order in which records are
sequentially read from the Indexed input file and written to the output file,
where n=O for the Primary Key, n=l for the First Alternate Key, and so on
until n==9 for the Ninth Alternate Key.
RMSCNV will not process any input Indexed file with a key of reference
greater than nine. It terminates with the error message:

?CNlJ -- INVALID /KR VALUE

Default If you do not include the KR switch, RMSCNV uses the Primary

Key as the key of reference.
9.2.4.5 Command String Examples

• PAYROL.DAT/FO:IDX=NAME.FIL
RMSCNV reads each record of the input file NAME.FIL and examines the
contents in the Primary Key field (defined as an attribute of the output file
PA YROL.DAT). Then the utility inserts the record into PAYROL.DAT.
• NAME.DAT/FO:REL=MASTER.DAT/KR:l
RMSCNV reads the Indexed input file MASTER.DAT, using the First Alternate Key of the file to establish the sequence of access. Then the utility
writes the records sequentially into an empty Relative output file
NAME.DAT, starting with record cell one.
• ALPHA.BAR/FO:SEQ/AP=BETA.BAR/KR:2
RMSCNV reads the Indexed input file BETA.BAR, using the Second Alternate Key of the file to establish the sequence of access. Then the utility
appends the records onto the end of the existing Sequential output file
ALPHA.BAR.
If the AP switch had not been specified, RMSCNV would have created the

next higher version of ALPHA.BAR and written the records from
BETA.BAR into it.
• DELTA.DAT=GAMMA.DAT
RMSCNV creates the Sequential file DELTA.DAT and copies the records
from GAMMA.DAT into it.

9-26

RMS-11 Utilities

• NEWPAY.DAT/WF/FO:REL=OLDPAY.DAT
RMSCNV reads fixed-length records from the Indexed input file
OLDPAY.DAT, using the Primary Key of the file to establish the sequence
of access. Then the utility writes the records sequentially in a VFC format
into the existing Relative output file NEWPAY.DAT. Record format is an
attribute of each file. As each record is written, the first bytes become the
fixed control area.
• MT3:INVENT.BCK/rR/BL:l024/AP=SY:INVENT.DAT/KR:O
RMSCNV reads the Indexed input file INVENT.DAT, using the file's
Primary Key to establish the sequence of access. Then the utility writes the
records to magnetic tape, truncating them to the output file's record length
before adding them to the end of the existing file, and formatting the tape
blocks to 1024 bytes each.

9.2.5 Cautions
• If an existing Sequential file is used for output and neither the supersede

(SU) or the append (AP) switch is specified, the utility creates the next
higher version of the file and uses it.
• A Relative or Indexed output file must exist before RMSCNV is invoked.
The file may be created through an application program or the RMSDEF
utility.
• If you did not allow duplicate keys for one or more key fields of an Indexed

output file, RMSCNV bypasses any input record that would cause such
duplication to occur. The utility notifies you of its action as described under
"SL[:filespec]" in Section 9.2.4.2.
• If any input record is shorter than the Primary Key of an Indexed output

file, RMSCNV terminates with the error message:
'?CNt.1 - ·- ILLEGAL RMS RECORD SIZE

Any records processed before the too-short record may or may not have been
written to the output file, depending on a variety of circumstances,including
buffer sizes and I/0 techniques. You must examine the input file to determine its contents; a way to do this is:
CNV TI:=filespec/KR:O
Note that you .cannot fix the too-short record and then run RMSCNV with
the same command string: the output Indexed file may contain some or all
of the input records read before the exception record. You have to delete and
recreate the output file.
• When both the input and output files have variable-length records,
RMSCNV requires that the Maximum Record Size defined for the input file

RMS-11 Utilities

9-27

be less than or equal to the Maximum Record Size defined for the output
file. The utility terminates with the following error message if the MRS of
the input file is greater than that of the output file:

?CNl.J - - ILLEGAL RMS RECORD SIZE

Example An RMS-11 file with variable-length records, Maximum Record Size of 30 bytes,
is named VAR30.DAT. Another similar file, named VAR50.DAT, has a Maximum
Record Size of 50 bytes.

The following command string is valid:
VAR50.DAT=VAR30.DAT

The following command string is not valid:
VAR30.DAT=VAR50.DAT

• When both the input and output files have fixed-length records, RMSCNV
requires either the TR or PD switch if the fixed record lengths are not equal.
If neither switch is specified, the utility terminates with the error message:

~o""

?CNl.J -- INPUT AND OUTPUT RECORD SIZES DO NOT CORRESPOND

• When the input file contains variable-length records and the output file
contains fixed-length records, RMSCNV requires both the TR and the PD
switches. If both are not specified, the utility terminates with the error
message:
?CNt.1 -- SWITCH /TR OR /PD OR BOTH ARE NEEDED FOR THIS CONl.JERT

)~\

• You can use unit record devices for input and output files, restricted of
course by their capabilities:
Input
from terminal or card reader
Output to terminal or printer
You use a terminal for either output or input file with the file spec "TI:".
You end terminal input with the (CTFl./ZJ.
• Do not use either the infile or outfile specification as an argument in the SL
switch.

9.2.6 1/0 Techniques
RMSCNV uses the following I/0 techniques to maximize its performance:
• If either the output or input file is a Sequential file, RM SC NV sets the
Multi-Block Count to five blocks (see "I/0 Techniques," Section 3.3.3.5).

• When the output file is an Indexed file:

9-28

RMSCNV attempts to allocate extra buffers to cache the Root buckets of
all keys specified for the file. If the utility doesn't have enough task space
to cache all Roots, it continues processing with the two buffers required
for an Indexed file.

RMSCNV uses Deferred Write (see Section 7 .3 .2).

RMS-11 Utilities

9.3 RMSDEF Interactive Utility

9.3.1

Purpose

The interactive RMSDEF utility creates RMS-11 files, allowing you to
control all attributes of the files being created.

9.3.2 Effect
You specify file attributes by respondiiig to requests for data and questions
from RMSDEF. The method of questioning is outlined in Figure 9-1. The
figure shows the general flow of processing while Section 9.3.6 shows actual
messages and legal responses as well as the meaning of each attribute. You
can also get help from the utility by typing a question mark(?) in response to
any question or request for data.
RMSDEF can also build an indirect command file while you operate it. This
command file can be used thereafter to (re)create file(s) and can be modified
to create other similar files. See Sections 9.3.3.1 and 9.3.4.1.

NOTE
• Command files generated by RMSDEF Vl.5 are not compatible with RMSDEF Vl.8.
• Command files generated by RMSDEF are not compatible
across operating systems.
RMSDEF, however, does not write records into the file. The actual data
contents of the file must be loaded after the RMSDEF utility creates the file.
You can use either an application program, the RMSCNV utility, or the
RMSIFL utility to write records into the file.

NOTE
RMSDEF interprets all quantities you enter as decimal
numbers.
The following list indicates the information that RMSDEF always requests as
well as the requests that depend on specifications typed.
1. Command File?
a. If yes, file specification
b. If yes, create only command file or create both RMS-11 and command
file?

RMS-11 Utilities

9-29

2. File Specification
3. Data Structure
a. Minimum

1. File organization
2. Record format
3. Maximum record size
4. CARRIAGE RETURN control?
b. Possible

1. Size of fixed control area for VFC records
2. Maximum number of records in a Relative file
3. Block-spanning records in a Sequential file?

4. FORTRAN character control if no CARRIAGE RETURN control?
4.

Key Definition (Indexed files only)
a. Minimum

1. Position of key

2. Size of key
3. Data type

4. Name of key
5. Duplicate keys?

b. Possible
1. Change keys if duplicatable? if Alternate Key
2. Null key value? if Alternate Key
3. Null key character if null key value
5. File Structure
a. Minimum
1. Areas? (Indexed files only)

2. Placement control?
3. Initial allocation quantity

4. Default extension quantity
5. Contiguous?

9-30

RMS-11 Utilities

I I I

:u;

ii I

I I .... 14141 #¥¥¥ ,.,,,

Fl.gure 9-1:

KEY
IDENTIFICATION

RMSDEF Processing Flowchart

FILE
STRUCTURE

I _J

FILE
STRUCTUR:__

DATA-]-ALLOCATION
----------""I

FILE
CREATION

RMS-11 Utilities

b. Possible
1. Location if placement control

2. Exactly if placement control?
3. Type of alignment if placement control and areas
6. Data Allocation (Indexed files only)
a. Minimum
1. Number of bytes in data buckets filled
2. Number of bytes in index buckets filled
b. Possible
1. Area containing index Level 0 for each key if areas
2. Area containing index Levels 2+ for each key if areas
3. Area containing index Level 1 if areas
7. Protection

9.3.3 Call and Termination
9.3.3.1

Permanently Installed Utlllty

DEF [command string]
where command string is one of the following:
@filespec

if you want RMSDEF to read a command file named filespec
that was previously generated by RMSDEF (see Section
9.3.4.1). The utility processes the file, printing messages only
when it comes to the file creation phase (see Section 9.3.4.8).

HELP or ?

if you want an introductory HELP message for the RMSDEF

utility.

/ID

if you want the version number of the RMSDEF utility installed on your system. RMSDEF prints the following message:
THIS IS THE RMS DEFINE UTILITY, VERSION 1.snn

where nn is the patch level of the utility (see "Patch Level,"
Section 9.0 .2).
If you do not include a command string, the utility prints:
DD YOU WANT TD GENERATE A COMMAND FILE FOR FUTURE USECND>?

See Section 9.3.4 for the complete· dialog sequence.

9-32

RMS-11 Utilities

9.3.3.2 Unlnstalled Utlllty

RUN $RMSDEF
The utility prints:
DO

YOU

WANT

GENERATE

A COMMAND FILE

FOR FUTURE

USE(NO>?

See Section 9.3.4 for the complete dialog.
You may terminate RMSDEF at any time
by typing a (CTRL/ZJ. The current keyboard monitor or command interpreter
resumes control of your interface.
9.3.3.3 Terminating The Utlllty -

9.3.4 Process
9.3.4.1

Command Fiie

1. The terminal prints:
DO YOU WANT TO GENERATE A COMMAND FILE FOR FUTURE USE (NO>?

Type one of the following:

if you want to enter a filespec for an indirect command file
and RMSDEF to write entries into the file as you move
through the utility. Go to step 2.

N or (flgj)

if you do not want to build a command file. Go to Section
9.3.4.2.

@filespee

if you have already built a command file with RMSDEF
and want RMSDEF to read it now and create the specified
file. Go to Section 9.3.4.8.

/ID

if you want the utility to identify itself with a version number. RMSDEF prints the following message:
THIS IS THE RMS DEFINE UTILITY, VERSION 1.Bnn

where:

nn is the patch level of the utility (see "Patch Level,"
Section 9.0.2).
2. The terminal prints:
ENTER FILE SPECIFICATION FOR COMMAND FILE:

Type a filespec. See Appendix A for RMS-11 restrictions on file specifications defined by your operating system documentation.
3. The terminal prints:
DD YOU WANT TO CREATE THE FILE YOU WILL BE DESCRIBING<NO>?

RMS-11 Utilities

9-33

Type one of the following:

if you want RMSDEF to create the RMS-11 file being specified as well as the command file.

Nor IBm

if you do not want to create the RMS-11 file, only the
command file that can be used to create the file later.

NOTE
The command file created by RMSDEF takes the following
form shown. The utility follows each comment with the appropriate sequence of your entries, each on a separate line. Where
you enter only (8m to accept the default value, RMSDEF places
the CR/LF character sequence on a separate line in the command file.

;THE FIRST QUESTION ASKS FOR THE FILE SPECIFICATION.
;THE NEXT QUESTIONS DEAL WITH FILE ORGANIZATION & RECORD ATTRIBUTES
H HE F0 LL 0 WI NG QUEST IONS DE AL WITH KEYS (for Indexed files only)
;THE NEXT QUESTIONS DEAL WITH ALLOCATION AND PLACEMENT ATTRIBUTES
HHE NE><T QUEST IONS ASK ABOUT FI LL SIZES FOR KEYS (for Indexed files only)
;THE FOLLOWING QUESTIONS DEAL WITH FILE PROTECTION
9.3.4.2 Fiie Specification -

1. The terminal prints:
ENTER FILE SPECIFICATION:

Type one of the following:
files pee

if you want to create (or simulate the creation of; see Section 9.3.4.1, step 3) an RMS-11 file. Type a filespec as
defined by your operating system documentation; see also
Appendix A for RMS-11 restrictions on filespecs. Go to
Section 9.3.4.3.

@filespee

if you have built a command file (see "Command File,"
Section 9.3.4.1) and want RMSDEF to use it and create the
file specified. Go to Section 9.3.4.8.

2. The terminal prints:
IF THE FILE ALREADY EXISTS t DO YOU WANT TO SUPERSEDE IT(NO)?

It is possible that a file already exists with the file specification you entered in step 1. You must tell RMSDEF what to do if such a conflict

9-34

RMS-11 Utilities

occurs when it tries to create the file you describe during this interactive
session. On IAS/RSX-llM, a conflict results only if you specified a version number in step 1.
Type one of the following:
y

if you want RMSDEF to supersede an existing file.

N or (BDJ

if you do not want RMSDEF to create the file you will
describe by superseding a file that exists.

CAUTION
1. If you say "N" here, and

2. If you are generating a command file, and
3. If you describe other files after this one in
that command file, and
4. If, when you use the command file, a file
with the filespec entered in step 1 exists:
RMSDEF prints the following error message
and terminates processing of the command
file.
A FILE WITH THE FILE SPECIFICATION YOU ENTERED ALREADY EXISTS

9.3.4.3 Data Structure

1. The terminal prints:
ENTER FILE ORGANIZATIONCSEQ):

Type one of the following:
SEQ or (BDJ

for Sequential organization

REL

for Relative organization

IDX

for Indexed organization

NOTE
If you indicated a magnetic tape device in the filespec,

RMSDEF does not request a file organization. Since a magnetic tape file requires Sequential organization, the utility
prints:
SINCE YOU SPECIFIED A NON-DISK DEIJICE,
YOUR FI LE ORGANIZATION MUST BE SEQUENT I AL!

RMS-11 Utilities

9-35

2. The terminal prints:
ENTER RECORD FORMAT< t.JAR):

Type one of the following:
VAR or ffiTI)

if the records in the file have differing, or variable, lengths.

FIX

if the records in the file have the same, or a fixed, length.

VFC

if each record in the file has a control area with a fixed
length and a data area of no standard length, that is, Variable with Fixed Control.

STM

if the records in the file have no specific format, but are
delimited only by record terminator characters. The stream
format is permitted for Sequential disk files only.

UDF

if there are no records (or you do not want RMS-11 to
recognize records) in the file; this format is used only for
Block 1/0 files, such a,s RMS-11 back-up files. The undefined format is permitted for Sequential disk or tape files
only.

RMSDEF rejects a format that conflicts with the file organization specified, for instance, STM for Indexed files.
3. If you specified VFC in step 2, the terminal prints the following message;
otherwise, go to step 4.
ENTER SIZE OF FI }-{ED CONTROL AREA< 2) :

Type the decimal number of bytes in the fixed control area of each record
in the file. The minimum size is one byte; the maximum size is 255 bytes;
the default is two bytes.
4. The terminal prints:
ENT ER MA>{ I MUM RECORD SIZE:

or
ENT ER MA>{ I MUM RECORD SIZE< 0) :

Type a decimal number indicating the maximum number of bytes in any
record in the file. RMS-11 checks this value whenever a record access
opera ti on is requested for this file: if the record specified exceeds the
maximum size, RMS-11 returns an error. A size of zero disables the
RMS-11 check, but a nonzero value is required for all Relative files and all
files with fixed-length records.
5. If you specified REL in step 1, the terminal prints the following message;
otherwise, go to step 6.
ENTER MAH I MUM NUMBER OF RECORDS< 0) :

9-36

RMS-11 Utilities

Type a decimal number indicating the maximum number of records that
this Relative file will contain. RMS-11 checks this value whenever a record access operation is requested for this file: if the relative record number
specified exceeds the maximum record number, RMS-11 returns an error.
A~ sets the number to zero, which disables the RMS-11 check. The zero
allows the file to contain as many records as is physically possible (the
technical maximum is 2.14748 x 10 9 ).
6. If you specified SEQ in step 1, the terminal prints the following message;
otherwise, go to step 7.
WI L.L YOU ALLOW RECORDS TO CROSS BLOCK BOLIN DAR I ES (YES)?

Type one of the following:
Yor ~

if you want records to cross block boundaries.

if you do not want records to span blocks. If you specified
FIX in step 2 and a maximum record size greater than 512
in step 4, the terminal prints:

SINCE YOU SPECIFIED FI>(ED SIZE RECORDS t YOU MUST HAl.JE A MAHI MUM RECORD
SIZE LESS THAN 512 <THE SIZE OF 1 BLOCK) OR YOU MUST ALLOW RECORDS TO
CROSS BLOCK BOUNDARIES. PLEASE CHANGE ONE OF THESE.

RMSDEF repeats steps 4 and 6. Change either your
Maxim um Record Size or the answer to crossing block
boundaries.

~·

7. The terminal prints:
DO YOU WANT CARR I AGE RETURN CONTROL< YES)?

Type one of the following:
Yor ~

if you want each record to be preceded by a line feed character and followed by a carriage return character when it is
written to a unit record device (printer, terminal, and so
on). See the following note and go to appropriate section.

if you do not want CARRIAGE RETURN control and/or
you do want FORTRAN character control. Go to step 8.

8. The terminal prints:
DO YOU WANT FORTRAN CHARACTER CONTROL (NO)?

Type one of the following:

if you want the first byte of each record to be interpreted as
a FORTRAN forms control character when the record is
written to a unit record device.

Nor~

if you do not want FORTRAN character control.

RMS-11 Utilities

9-37

NOTE
If you indicated a magnetic tape device in the file specification,
at this point RMSDEF requests:
ENTER MAG TAPE BLOCK SIZE

C512):

Type a decimal number between 18 and 8192 representing the
number of bytes in each tape block. The number should be a
multiple of four; if it is not, RMS-11 rounds the number up to
the next multiple of four before writing it as an attribute. A~
sets the size to the default of 512 bytes.
The utility then bypasses other processing and immediately
requests protection information (see Section 9.3.4.7).

9.3.4.4 Key Definition - AB indicated by Figure 9-1, this section applies only
to Indexed files. RMSDEF begins this portion of dialog with the message:
IT'S TIME TO DEFINE THE PRIMARY KEY

1. The terminal prints:
ENTER DATA TYPE C STR):

Type one of the following:
STR or ~

if your key value is a string of alphanumeric characters.

INT

if your key value is a two- or four-byte signed integer.

BIN

if your key value is a two- or four-byte unsigned binary

number.
PAC

if your key value is a packed decimal number.

See "Key Data Type," Section 6.2.3, for a discussion of key data types.
2. The terminal prints:
ENTER POSITION OF KEY:

Type a decimal number indicating the position of the first byte of the key
within each record. For instance, if the key starts with the first byte of the
record, its position is 0. The second byte has position 1 and so on.
A position number must be specified for each segment of a segmented
string key; the numbers are separated by commas and enclosed in
parentheses. Integer, binary, and packed decimal keys cannot be
segmented.
Example A key has three segments that start at the first, fourth, and sixteenth positions.
They are described as follows:
COt3t15)

9-38

RMS-11 Utilities

3. The terminal prints:
ENTER SIZE OF KEY:

Type the decimal number of bytes in the key, that is, its length. Valid
lengths depend on the data type entered in step 1:
Data Type

Length Restrictions

STR

Minimum = 1 byte
Maximum= 255 bytes

INT

2 or 4 bytes

BIN

2 or 4 bytes

PAC

Minim um = 1 bytE~
Maximum= 16 bytes

A length must be specified for each segment of a segmented string key; the
numbers are separated by commas and enclosed in parentheses. A length
must be typed for each position number specified in step 1, but the sum of
all lengths cannot exceed 255. Integer, binary, and packed decimal keys
cannot be segmented.
Example The lengths of the keys in the step 2 example are entered as follows:
<2t2t1G)

The terminal prints:
ENTER NAME OF KEY< NONE) :

Type one of the following:
if you do not want to specify a key name.

name

if you want to name the key being defined; up to 32 ASCII
characters are allowed.

5. The terminal prints:
WILL YOU ALLOW DUPLICATE KEYS<dflt>?

Type one of the following:

if the file may contain more than one record with the same
value for this key. Alternate Keys must be specified as duplicatable before they can be specified as changeable (in
step 6).

if each record in the file must have a unique value for this
key. RMS-11 returns an error if duplication is attempted;
that is, put or update operations fail when the record has a
value in this key field exactly like a record already in the
file.

RMS-11 Utilities

9-39

Defaults and the values of dflt are:
Primary Key

Alternate Keys

Yes

When defining the Primary Key, go to step 10. Otherwise, go to step 6.
6. If you specified Yes in step 5, the terminal prints the following message;
otherwise, go to step 7.
WILL YOU ALLOW KEYS TO CHANGE< YES)?

Type one of the following:
Yor ~

if this Alternate Key can change values during an update
operation; that is, the record may be read with one value for
the key and rewritten with another value for the same key.

if this Alternate Key must not change after the record is
originally created.

7. The terminal prints:
DO YOU WISH TO DEFINE A NULL KEY 1.JALUE< NO)?

Type one of the following:

if you want the file to contain some records that cannot be
accessed via this key. See ''Key Characteristics,'' Section
6.2.5. Go to step 8 only if you specified a string key (step 1).
Otherwise, go to step 9; the other key data types have a null
key value of 0.

Nor~

if you do not want to specify a null value for this key. Go to
step 9.

8. The terminal prints:
ENTER NULL KEY 1.JALUE CHARACTER:

Type one of the following:

9-40

the single character itself, if it is not #, ?, @, or SPACE.

#40

for SPACE.

#43

for the reserved character #.

#77

for the reserved character ? .

#100

for the reserved character @.

#nnn

any octal byte value (000-377 8 ), specified by nnn.

RMS-11 Utilities

9. The terminal prints:
JU S T F I N I SHED ALTERNATE K E Y nnn

where:

nnn

starts with 1 and is incremented when you answer Yes to
the next question (step 10).

10. The terminal prints:
DO YOU WANT TO DEFINE MORE KEYSCNO>?

Type one of the following:

if you want to define more keys for the file. You may define
up to 254 alternate keys; however:
• The RMSCNV and RMSIFL utilities do not read higher
than the ninth Alternate Key, that is, the KR switch
must be less than or equal to nine (see Sections 9.2.4.4
and 9.5.4.4).
• Your application language may not support that many
keys. See your language manual.
RMSDEF then requests the information in (steps 1-9) for
each Alternate Key indicated; the Alternate Keys are defined in order, beginning with the First Al tern ate after the
Primary Key is defined.

Nor (80

if all keys for this file are defined.

9.3.4.5 FUe Structure

1. If you specified IDX for file organization, the terminal prints the following
message; otherwise, go to step 2.
DO YOU WANT TO DEF I NE AREAS< NO)?

Type one of the following:

if you want parts of this file to be logically different, with
different attributes. See "Areas," Section 6.3. The following questions (steps 2-10) are asked for each area.

Nor~

if you want this indexed file located in one area. RMSDEF
requests the following information (steps 2-10) for the file
once.

RMS-11 Utilities

9-41

2. The terminal prints:
DO YOU WANT PLACEMENT CONTROL CNO)?

Type one of the following:

if you want to specify an exact location on disk for this file
or area. Go to step 3.

N or (80

if you do not want to specifically locate this file or area. Go
to step 7.

3. On VAX systems only, the terminal prints:
IF YOU WANT TO SPECIFY t,lOLUME, ENTER ITS NUMBER WITHIN THE t,JOLUME
SETCO):

Type one of the following:
0 or (80

If you want the file or area placed:

• on the volume indicated by the filespec you entered (see
Section 9.3.4.2, step 1) - if that disk is not the root
volume of a defined volume set.
• on any volume in the set if the disk indicated by the
filespec is the root volume of a defined volume set.

A zero here essentially nullifies a YES answer in step 6.

if you want the file or area placed on another disk in the
volume set indicated by the filespec you entered, where n is
1 through the maximum number set for the volume.

On VAX systems, the terminal always prints the following message. On
IAS and RSX-llM, the terminal prints the following message only for
Areas 1+ in Indexed files.
ENTER TYPE OF AL I GNMENT ( LBN):

Type one of the following:

9-42

LBN or (80

if the location you will specify in step 5 is a Logical Block
Number on the disk volume.

VBN

valid only for Areas 1+ in Indexed files, if the location you
will specify in step 5 is a Virtual Block Number already
established within the file; that is, you are trying to closely
align this area with a defined area.

RMS-11 Utilities

valid only on VAX systems, if the location you will specify in
step 5 is the number of a cylinder on the disk volume. By
specifying CYL, you are directing the file processor to place the
first block of this file or area at or near the first logical block of
the cylinder you will specify in step 5.

CYL

'-''

5. The terminal prints:
ENTER LOCATION:

Type the decimal number location of the first block for this file area.
There is no default. For the first area defined, this number is:
• a Logical Block Number
• Device Cluster Number
For VAX users, if you specified CYL in step 4, you can enter here either:
• a valid number for a cylinder

'-''

• -1, if you want the area or file to start with the first block of any cylinder
on the disk
6. The terminal prints:
E>{ACTL Y (NO)?

Type one of the following:

if this file or area must start in the exact LBN location
specified in step 5. If this location is not available,
RMSDEF prints an error message after it tries to create the
file. Exact VBN locations are already taken, by definition.

Nor (80

if you will accept the closest approximation of the LBN or
VBN location specified in step 4, when the exact location is
not available.

7. The terminal prints:
ENTER INITIAL ALLOCATION IN BLOCKS (0):

Type a decimal number indicating the initial size of the file or area in
blocks. A (80 sets the value to zero: the area will be created, but it must be
extended before any records can be written into it. Since automatic file
extension is time-consuming, the file should be fully allocated when it is
created (see Section 6.6).

NOTE
If you have specified placement control (step 2), you should

type a nonzero allocation quantity. An inital allocation of
zero blocks essentially nullifies placement control: the file
processor does not allocate any blocks to the file, so it does
not use the placement control information.

RMS-11 Utilities

9-43

Example You can place· multiple files at the same LBN or DCN location, even answering YES to D<ACTL Y? (step 6), as long as
you use zero allocation quantities for all but one.

8. If you specified REL or IDX for file organization, the terminal prints the
following message; otherwise, go to step 9.
BUCKETSIZEC1):

Type a decimal number indicating the number of blocks in a bucket for
this file or area. The minimum is the number of blocks that can contain
one record (according to the size specified; see Section 9 .3.4.3, step 4); the
maximum is 32 or 15 blocks; the default is one. This number determines
the num her of blocks read into memory during each file access operation
and therefore affects processing speed and the amount of memory a program accessing this file requires. See "Bucket Size" Section 6.5.
9. The terminal prints:
ENTER DEFAULT E>CTENSION QUANTITY IN BLOCKS(<))

Type a decimal number indicating the number of blocks that should be
added to the file or area each time RMS-11 extends it. The Default Extension Quantity (DEQ) should be a multiple of the bucket size. RMS-11
requests this number of blocks from the operating system; RSTS/E, however, actually extends the file to the next full cluster larger than the
request.

A (8TIJ sets the value to zero: RMS-11 adds only the minimum space
required each time it expands the file or area. A definite, but reasonable
extension quantity speeds processing.
10. The terminal prints one of the following messages, except on RSTS/E, for
Areas 1+:
DO YOU WANT A CONT I GUOUS AREA (NO)?

or
DO YOU WANT A CONTIGUOUS FILE< NO)?

Type one of the following:
Y

if you want the disk space for this file or area allocated in
contiguous blocks. If the file processor cannot find that many
contiguous blocks, it will not create the file, even if enough
non-contiguous blocks are available.
A contiguous file or area may be extended although the
disk space added may not be contiguous with the original
allocation.
RSTS/E cannot extend a contiguous file. Be sure your
initial allocation quantity (step 7) is sufficient.

9-44

RMS-11 Utilities

11. If you answered Yes in step 1 (you have an Indexed file), the terminal
prints the following messages; otherwise, go to the next appropriate
section.
JUST FINISHED WITH AREA NUMBER nnn
DO YOU WANT TO DEFINE MORE AREAS< NO)?

Type one of the following:
y

N or IBill

if you want to specify attributes for another area of your
file. Areas are numbered sequentially, starting with zero.
The areas are associated with the index and data portions
of the file in the next section of the utility. Go to the step in
this section indicated below:
Step

Operating System

IAS or RSX-llM

RSTS/E

if you have defined enough areas for this file. Go to step 12.

12. If you defined one or more areas with a DEQ of zero, the terminal prints
the following message; otherwise, go to the next section.
ENTER FI LE DEFAULT E}<TENS I ON QUANT I TY ( 0) :

Type a decimal number indicating the number of blocks that should be
added to the file each time RMS-11 extends it. The DEQ should be a
multiple of the bucket size. A ~ sets the value to zero: RMS-11 will add
only the minimum space required each time it expands the file. A definite,
but reasonable extension quantity speeds processing.
9.3.4.6 Data Allocation -As indicated by Figure 9-1, this section applies only

to Indexed files. RMSDEF begins this portion of dialog with the message:
IT IS TIME FOR AREA NUMBERS AND FILL FACTORS FOR KEYS.

The questions are asked for each key defined (see Section 9.3.4.4). RMSDEF
begins the first session with the message:
THESE QUESTIONS ARE FOR THE PRIMARY KEY

Any further sessions begin with the message:
THESE QUEST! ONS ARE FOR ALTERNATE KEY NUMBER nnn

where:

nnn starts with 1.
l.

The terminal prints:
ENTER AREA NUMBER FOR DATA BUCKETS FOR TH IS KEY< 0):

Typ,e an integer (0-n) indicating the area already defined (see Section
9.3.4.4) that should contain the data portion (Level 0) of this key's index.

RMS-11 Utilities

9-45

2. The terminal prints:
THE BUCKET SIZE IS nn
ENTER FILL NUMBER FOR DATA BUCKETS CO>:

Type a decimal number of bytes in each of this key's data buckets that
should be used during the original population of the file. This number is
honored by the RMSCNV and RMSIFL utilities and may be honored by
MACR0-11 programs. See "Fill Number," Section 6.7.2.1.
A m:UJ sets the number to zero, indicating that buckets will be filled and
that no free space will be available in existing buckets for records added
after the file is populated.

3. The terminal prints:
ENTER AREA NUMBER FOR INDEH BUCKETS FOR THIS KEY CO):

Type an integer (0-n) indicating the area already defined (see Section
9.3.4.4) that should contain the upper portions (Levels 2+) of the index for
this key.
4. The terminal prints:
T HE BLICK ET S I Z E I S nn
ENTER FILL NUMBER FOR INDE>< BUCKETS CO)

Type a decimal number of bytes in each of this key's index buckets that
should be used during the original population of the file. This number is
honored by the RMSCNV and RMSIFL utilities and may be honored by
MACR0-11 programs.
A ffi:UJ sets the number to zero, indicating that buckets will be filled and
that no free space will be available in existing buckets for records added
after the file is populated.

5. The terminal prints:
ENTER AREA NUMBER FOR THE LOWEST I NOD< LEl.JEL FOR TH IS KEY ( 0):

Type an integer (0-n) indicating the area already defined (see Section
9.3.4.4) that should contain the lowest index lev~l (Level 1) portion of this
key's index.
If the area you specified for the upper portions of the index (Levels 2+)
and the area you specified for Level 1 have different bucket sizes,
RMSDEF prints the following message and returns to step 3:
THE AREA ASSOCIATED WITH THE LOWEST LEl.lEL INDE}{ BUCKET
HAS A DIFFERENT BUCKET SIZE THAN THE AREA ASSOCIATED WITH THE
HIGHER I NOE>< BUCK ET. TRY BO TH AGAIN.

9-46

RMS-11 Utilities

9.3.4. 7 Protection

1. The terminal prints:
SPEC I FY PROTECT I ON BY CLASS:
OWNER ( RWED ALLOWED) :

Type one of the following:
if you .want this file available to access (Read,
Write, Extend, Delete) by the account current when
the file is created.
NONE

if you do not want the file owner to have access to
this file after it is created.

R, W, E, and/or D

if you want to specify a level of protection between
none and all. You may specify one or more of the
letters representing Read, Write, Extend, and
Delete without separation.

2. The terminal prints:
GROUP ( RWED ALLOWED):

Type one of the following:
if you want this file available to access (Read,
Write, Extend, Delete) by accounts with the same
group or project number as the owner's account.

:3.

NONE

if you do not want the group members to have access to this file after it is created.

R, W, E, and/or D

if you want to specify a level of protection between
none and all. You may specify one or more of the
letters representing Read, Write, Extend, and
Delete without separation.

The terminal prints:
SYSTEM ( RWED ALLOWED) :

Type one of the following:
~

if you want this file available to access (Read,
Write, Extend, Delete) by system privileged
accounts.

NONE

if you do not want privileged accounts to have access
to this file after it is created.

R, W, E, and/or D

if you want to specify a level of protection between
none and all. You may specify one or more of the
letters representing Read, Write, Extend, and
Delete without separation.

RMS-11 Utilities

9-47

4. The terminal prints:
WORLD ( R ALLOWED) :

Type one of the following:
~_!)

if you want this file available for Read access only
by all accounts outside the owner, group, and privileged accounts.

NONE

if you do not want other accounts to have access to

this file after it is created.
R, W, E, and/or D

if you want to specify a level of protection other
than none and Read only. You may specify one or
more of the letters representing Read, Write, Edit,
and Delete without separation.
~

The terminal prints:
CLUSTERSIZE<O>:

Type a decimal number of blocks in a cluster for this file. The number is a
system-oriented extension quantity and must be equal to or a power of two
greater than the device clustersize (maximum of 256). A IBDJ sets the
clustersize to zero, which means RMS-11 defaults to the volume clustersize, and -1 sets the clustersize to 256.
2. The terminal prints:
PROTECTION<GO>:

Type a decimal protection number for the file. See Part I of the RSTS/E
System User's Guide for an explanation of and a guide to calculating
protection numbers.
9.3.4.8 Fiie Creation -The RMSDEF utility attempts to create the file.

Success -

If RMS-11 does not return an error, the utility prints:

Y0 UR F I LE HAS BE EN C RE ATE D ! !

files pee

If you chose to create a command file (see Section 9.3.4.1), RMSDEF also
prints the following message. Otherwise, the utility requests you to enter
another file specification (Section 9.3.4.2).
YOUR FI LE HAS BEEN PROCESSED AND A COMMAND FI LE GENERATED! !
DO YOU WANT TO CLOSE THE INDIRECT FILE<ND>?

- -

filespec

Type one of the following:

9-48

RMS-11 Utilities

if you are finished specifying files for the command file. The
utility returns to the question about command file generation
(see Section 9.3.4.1).

Nor~

if you want to specify another RMS-11 file and you want the
command file to include your input. RMSDEF continues to use
the command file originally specified and to obey your answer
to the "DO YOU WANT TO CREATE THE FILE YOU WILL
BE DESCRIBING?" question (see Section 9.3.4.1, step 3). The
utility returns to the request for a file specification (see Section
9.3.4.2).

You may start the process to create another file or type (CTR.. /~ to terminate the
utility.

Error - RMS DEF allows you to recover from one type of creation error. The
terminal prints an error description, followed by:
DO YOU WISH TO REENTER THE ENTIRE FILE SPECIFICATION

<YES>?

Type one of the following:
Y or ffi)

if you know how to correct the error and/or want to enter another filespec. The utility requests the filespec and attempts to
create the file using it.

if you do not know how to correct the error and/or want to start
again. RMSDEF returns to the file specification request (Section 9.3.4.2).

Other errors result in one of the following messages:
•

THIS FILE CANNOT BE CREATED
RMS STATUS CODE = nnnnnn

SYSTEM CODE = nnnnn11

• TH IS FI LE CAN NOT BE CREATED SINCE THE FI LE ALREADY E)<I STS AND YOUD ID
NOT SPECIFY SUPERSEDE.

"Error Code Mapping," Appendix B.2, defines the codes represented by
nnnnnn.
The utility returns to the file specification request to let you restart the
definition process (see Section 9.3.4.2).

9.4 RMSDSP Command Utlllty
9.4.1 Purpose
The RMSDSP utility provides a concise description of the RMS-11 file attributes of any file as well as the contents of back-up container files on magnetic
tape.

RMS-11 Utilities

9-49

9.4.2 Effect
The RM SD SP utility lists the following information for each file specified in a
proper command string. The information is either printed on the calling terminal or written into a disk file, depending on the format of the command
string.
LEGEND

filespec

a complete file specification

dd-mon-year hh:mm

the day, month, year, and time the file was created or
revised

NOTE
RMSDSP does not print null dates.
run

the operating system's revision number of the file

code

one of the following:
• one or more of the protection codes R (read), W
(write), E (extend), and D (delete)
• an integer specifying the protection status of the
file, in the form <nnn>

format

F!}{ED, l.JARIABLE, lTIFC, STREAM, or??? {unde-

fined)
length

an integer number of bytes set as the maximum record length

character control

CAR R I AG E RE T LI RN or F 0 RT RAN

num her of blocks

filsiz

number of blocks currently allocated to the file

unusd

number of blocks currently allocated to the file that
are not in use

type

the key data type as STR, INT, BIN, or PAC

changes

ND CHANG ES or CHANGEABLE

9.4.2.1 For Disk Sequential Flies filespec

FILE ORGANIZATION:
SEQUENTIAL
CREATED: dd-mon-year hh:mm
[REV I SEO: dd-mon-year hh:mm (run> J
FILE PROTECTION:
code
RECORD FORMAT:
format =length
RECORD ATTRIBUTES:
[character control J
CNO SPANNINGJ
FILE ATTRIBUTES:
ALLOCATION =filsiz EXTEND QUANTITY=nnnn
CCONTIGUOUSJ

9-50

RMS-11 Utilities

~
.
\

9.4.2.2 For Magnetic Tape Flies SEQUENTIAL

FILE ORGANIZATION:

filespec

FILE PROTECTION:
RECORD FORMAT:
RECORD ATTRIBUTES:
CNO SPANNING]
FILE ATTRIBUTES:
BL 0 CI\ S I ZE=nnnn

code
format= length
[character control J

9.4.2.3 For Relative Flies RELAT II.IE
FILE ORGANIZATION:
PROLOGUE VERSION NUMBER: n
dd-mon-year hh:mm <run) J
[REI.I I SEO:
CREATED: dd-mon-year hh:mm
FI LE PROTECT I ON:
code
RECORD FORMAT:
format= length
MAXIMUM RECORD NUMBER=nnnn
[character control J
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BUCKET SIZE=bk
ALLOCATION =filsiz EXTEND QUANTITY=nnnn
[CONTIGUOUS]

filespec

9.4.2.4 For Indexed Flies I NDE>{ED
FILE ORGANIZATION:
PROLOGUE VERSION NUMBER: n
CREATED: dd-mon-year hh:mm [REI.I I SEO: dcl-mon-year hh:mm <run> J
FILE PROTECTION:
code
RECORD FORMAT:
format= length
RECORD ATTRIBUTES:
[character control]
FILE ATTRIBUTES:
BUCKET SIZE=bk
ALLOCAT I ON=filsiz EXTEND QUANTITY=nnnn
[CONTIGUOUSJ
nnnn
NUMBER OF KEYS:

filespec

9.4.2.5 For Indexed Flies with FU Switch Specified - See the RMSDEF interactive utility in section 9.3 for an explanation of key attributes.
FI LE ORGANIZATION:
I NDE}<ED
PROLOGUE VERSION NUMBER: n
CREATED: dd-mon-year hh:mm rnE1.1 I SEO: dd-mon-year hh:mm <run) J
FI LIE PROTECT ION:
code
RECORD FORMAT:
format= length
RECIJRD ATTR IfWTES:
[character control]
FI LI:: ATTRIBUTES:
ALLOCATION=fi~k
EXTEND QUANTITY=nnnn
BUCKET SIZE=bk
[CONTIGUOUS]

filespec

AREA NUMBER: nnn
ALLOCATION REMAIN I NG = unused
BUCK ET S I ZE = nn EXTEND QUANT IT Y

nnnn

RMS-11 Utilities

9-51

PR I MARY KEY:
type
[NO J DU PL! CATES changes
CNULL =ccc J
ROOT VIRTUAL BLOCK NUMBER=nnnn ROOT LEVEL=nnn
1ST DATA BUCKET VIRTUAL BLOCK NUMBER=nnn
KEY NAME:
ccccccccccccccccccccccccccccccc
POSIT ION:
nnn
SIZE:
nnn
TOTAL KEY SIZE=nnn
MINIMUM RECORD LENGTH=nnn
DATA LEVEL FILL SIZE=nnnn
INDEX LEVEL FILL SIZE=nnnn
DATA AREA=nn
INDEX AREA=nn LOWEST LEVEL AREA=nn
ALTERNATE KEY: n
type
[NO J DU PL! CATES changes
[NULL =ccc J
ROOT VIRTUAL BLOCK NUMBER=nnnn ROOT LEVEL=nnn
1ST DATA BUCKET VIRTUAL BLOCK NUMBER=nnn
KEY NAME:
ccccccccccccccccccccccccccccccc
POSIT ION:
nnn
SIZE:
nnn
TOTAL KEY SIZE=nnn
MINIMUM RECORD LENGTH=nnn
DATA LEVEL FILL SIZE=nnnn
INDEX LEVEL FILL SIZE=nnnn
DATA AREA=nn
INDEX AREA=nn
LOWEST LEVEL AREA=nn

9.4.3 Call and Termination
9.4.3.1

Permanently Installed Utlllty

DSP [command string]
If you include a command string, the utility attempts to execute it and then

returns control to the current keyboard monitor or command interpreter.
If you do not specify a command string, RMSDSP assumes control of the user

interface and prints the prompt:
DSP>

You may type a command string or ~ to terminate the utility. When
RMSDSP has executed a command string, it reprints the prompt.

9.4.3.2 Unlnstalled Utlllty

RUN $RMSDSP
RMSDSP assumes control of the user interface and prints the prompt:
DSP>

You may type a command string or ~ to terminate the utility. When
RMSDSP has executed a command string, it reprints the prompt.

9-52

RM S-11 Utilities

9.4.4 Command String
Like all RMS-11 utilities, RMSDSP requires ANSI-standard labels on magnetic tapes. You must mount a tape volume to inform your operating system's
file processor that you are accessing an ANSI-standard tape. See Appendix F
for more details on RMS-11 magnetic tape handling.
If you do not mount your tape volume, RMSDSP does not accept wild card
specifications and responds to an explicit file-name with the tape device's
attributes.
9.4.4.1 General Form

[outfile=]infile[/switch] ..][,infile[/switch] ..]... ]
where:

outfile

is the filespec of a file to be created by the RMSDSP utility.
RMSDSP then writes the attributes or contents of the specified
input file(s) into the outfile. The default version of the outfile is the
highest existing version number plus one.
If an outfile specification is not included in the command string,
RMSDSP prints its information on the calling terminal.

No wild cards are allowed in the outfile specification.

infile

is the filespec of either:
• Any file whose RMS-11 attributes you want displayed.
• A back-up container file on magnetic tape whose contents you
want displayed. You must include the BP switch (see "Switches,"
Section 9.5.4.2) to indicate that the files are container files.
Wild cards are permitted in any field of the infile specification.

switch

may be a code shown in Table 9-3 and described under "Switches"
Section 9.5.4.2.
NOTE
A command string may also consist of the word "HELP" or a
question mark (?). RMSDSP responds with a HELP message.

Table 9-3: RMSDSP Utility Switches
Switch

Description

Default

/ID

Identify current version of RMSDSP.

No id.

IBP

List contents of ha.ck-up container file(s).

Display attributes of magtape files.

/FU

Display all indexed file attributes.

Basic display only.

RMS-11 Utilities

9-53

9.4.4.2 Switches

• ID causes RMSDSP to print its current version numbers in the form:
DSP -- IJERSION 1.enn

where nn is the patch level of the utility (see "Patch Level," Section 9.0.2).
This switch may appear alone as a command string.
• BP causes RMSDSP to read the specified input files as magnetic tape
container files (see "Command String" in RMSBCK, Section 9.1.4) and
display the names of the RMS-11 back-up files contained therein. All infiles
specified must be container files.
Default If you do not include the BP switch,RMSDSP prints the attributes

of the magtape files (see Section 9.5.2.2)
• FU causes RMSDSP to list the full attributes of an Indexed file, including
definitions for individual keys and attributes of and number of unused
blocks in areas, in addition to the basic data file attributes shown when the
FU switch is not specified (see section 9.5.2.5). Area information is printed
only when more than one area is defined.
Default If you do not include the FU switch, RMSDSP lists only the num-

ber of keys defined for Indexed file(s) and no area information (see
Section 9.5.2.4).
9.4.4.3 Command String Examples

•FILE.DAT
RMSDSP prints on the calling terminal the attributes of the file named
F1LE.DAT found on SY:. Attributes are displayed only for the highest
available version of the file.
• ATT.LST=*.*/FU
RMSDSP creates the file ATr.LST and writes into it a full listing of attributes for all files on SY: in the default account.
• MTO:ALPHA.BKP ,BETA .BKP/BP
RMSDSP lists on the calling terminal the names of all files within the two
container files ALPHA.BKP and BETA.BKP on the tape volume mounted
on MTO:.
• A scratch tape is mounted on a TE-16 9-channel tape drive that is part of a
PDP-11 computer system running under the RSX-llM operating system.
The operator allocates the tape drive, then initializes the tape using ANSIstandard labels with the commands:
ALL MM1:
I NI TtJOLUME MM 1: DSPTST

9-54

RMS-11 Utilities

The operator tries to examine the empty, unmounted tape with RMSDSP:
DSP MM1:*•*
?DSP -- ERROR WITH WILDCARDS
DSP MM1:
?DSP -- SYNTAX ERROR -

'rhen the operator mounts the tape, notifying the FllACP that the tape is
ANSI-standard:
MOUNT MM1:DSPTST

Again, the operator examines the empty tape with RMSDSP:
DSP MM1:
MM1: [303 ,351J.:1

FILE ORGANIZATION:

FI LE PROTECT ION:
RECORD FORMAT:
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512

CRWEDtRWEDtRWEDtRWEDJ
Fl){ED=512

SEQUENTIAL

DSP MM1:A.B.BAD
?DSP -- SYNTAX ERROR - BAD
DSP MM1:A.BAD
DSP -- FILE NOT FOUND - MM1:[303t351JA.BAD

The operator copies four RMS-11 files onto the tape using RMSCNV, then
examines the tape:
DSP MM1:*•*
MM1 :MD{SIM.B2S:O

FILE ORGANIZATION:

SEQUENTIAL

CRWEDtRWEDtRWEDtRWEDJ
FILE PROTECTION:
t.IAR I ABLE
RECORD FORMAT:
CARRIAGE RETURN NO SPANNING
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512
MM1 :MD{ROO.B2S:O

FILE ORGANIZATION:

SEQUENTIAL

FI LE PRClTECT ION:
CRWEDtRWEDtRWEDtRWEDJ
RECORD FORMAT:
VARIABLE
RECORD ATTRIBUTES:
CARRIAGE RETURN NO SPANNING
FILE ATTRIBUTES:
BLOCK SIZE=512
MM1 :Mnrns1 .B2S:O

FILE ORGANIZATION:

SEQUENTIAL

FI LE PROTECT ION:
CRWEDtRWEDtRWEDtRWEDJ
lJAR !ABLE
RECORD FORMAT:
RECORD ATTRIBUTES:
CARRIAGE RETURN NO SPANNING
FILE ATTRIBUTES:
BLOCK SIZE=512
MM1:MD{AS2.B2S:O

FILE ORGANIZATION:

SEQUENTIAL

FILE PROTECTION:
CRWEDtRWEDtRWEDtRWEDJ
t.IAR I ABLE
RECORD FORMAT:
CARRIAGE RETURN NO SPANNING
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512

RMS-11 Utilities

9-55

The file MIXSIM.B2S is listed first although it is the last file copied to the
tape. The FllACP provides directory information from the tape's current
position to the end and then rewinds the tape and reads to that position again.

The operator logs off the RSX-UM system, thereby dismounting and deallocating MMl :. The operator removes the tape from the drive and carries it to
another PDP-11 running under the RSTS/E operating system. After physically mounting the tape on another TE-16 9-channel drive, the operator logs
onto the RSTS/E system and tries to examine the tape with RMSDSP:
DSP MM1:*•*
Read>·
DSP MM1:
?DSP -- SYNTAX ERROR -

The operator mounts the tape, informing the File Processor that the tape is
ANSI-standard:

MOUNT MM1:DSPTST
MastaPe is in ANSI forMat
DENSITY IS 800 • PARITY IS ODD
Read>'

Now the operator examines the mounted tape:
DSP MM1:*•*

9-56

MM1:MIHROO.B2S

FILE ORGANIZATION:

FILE PROTECTION:
RECORD FORMAT:
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512

1.JARI ABLE
CARRIAGE RETURN NO SPANNING

MM 1: MI }<AS1. B2S

FILE ORGANIZATION:

FILE PROTECTION:
RECORD FORMAT:
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512

VARIABLE
CARRIAGE RETURN NO SPANNING

MM1 :MD<AS2.B2S

FILE ORGANIZATION:

FILE PROTECTION:
RECORD FORMAT:
RECORD ATTRIBUTES:
FI LE ATTRIBUTES:
BLOCK SIZE=512

<GO>
VARIABLE
CARRIAGE RETURN NO SPANNING

MM1:MI>{SIM.B2S

FILE ORGANIZATION:

FILE PROTECTION:
RECORD FORMAT:
RECORD ATTRIBUTES:
FILE ATTRIBUTES:
BLOCK SIZE=512

<BO>

RMS-11 Utilities

SEQUENTIAL

<GO>

SEQUENTIAL

<BO>

SEQUENTIAL

VARIABLE
CARRIAGE RETURN NO SPANNING

9.4.5 Cautions
• RMSDSP does not print allocation information for magnetic files. If you
want this data, use PIP.

9.5 RMSIFL Command Utlllty
9.5.1

Purpose

The RMSIFL command utility provides the fastest, most direct, and most
efficient method of populating an empty RMS-11 Indexed file. The utility can
read records from any type of RMS-11 file and load them into an Indexed file
that you have created. Unlike RMSCNV, RMSIFL does not use the standard
RMS-11 access methods to build the output file. Rather, RMSIFL exploits
the basic structure of Indexed files (discussed in Chapter 5) to insert data
records and construct indexes fast. As it does this, the utility creates optimally dense buckets at all levels of each index. RMSCNV, on the other
hand, can only optimize the Primary index when the input file is sorted by the
output file's Primary Key. However, you must use RMSCNV if the output file
contains records or you do not have room for the sort work files.

9.5.2 Effect
RMSIFL reads records from an input file and loads them into an output
Indexed file in a series of phases.
1. The utility examines the command string (described in Section 9.5.4),

rejecting it for syntax and/or other errors. When it has a valid command
string, RMSIFL determines if the input file must be sorted into ascending
order by the output file's Primary Key.
RMSIFL bypasses the sort phase (phase 2) and skips to phase 3 if:
• You included the Infile Switch NOSO (described in Section 9.5.4.4).
• If the key of reference specified for an Indexed input file is equal to or

contains the output Primary Key, based on position only. Nonstring key
types must match exactly, in starting position and length. String keys
must start in the same position, but the output Primary Key can be the
same length or shorter than the input key of reference.
You specify key of reference with the KR Infile Switch (described in
Section 9.5.4.4).
Example The ten-byte string input Alternate Key 2 starts at byte 0 and is specified as
the key of reference. The eight-byte string output Primary Key starts at byte
0. RMSIFL does not sort.
Example The ten-byte string input Primary Key starts at byte 0. The eight-byte string
output Primary Key starts at byte 3. RMSIFL sorts the input file.

RMS-11 Utilities

9-57

Example The three-byte packed decimal input Alternate Key 1 starts at byte 43 and is
specified as the key of reference. The three-byte packed decimal output Primary Key starts at byte 42. RMSIFL sorts the input file.

RMSIFL notifies you that it has started processing by printing the· following message:
PRIMARY KEY:

2. RMSIFL reads the input file in Sequential Access Mode, depending on the
organization of the input file:
Input File
Organization

Processing

SEQUENTIAL

RMSIFL reads the records in the Sequential Access Mode, starting
with the first record in the file.

RELATIVE

RMSIFL reads the records in the Sequential Access Mode, starting
with record cell 1.

INDEXED

RMSIFL reads the records in the Sequential Access Mode, following
the key of reference.

RMSIFL passes each record to modules extracted from the SORT-11
utility. These modules order the input records by ascending value in the
Primary Key field of the output file.
At the beginning of this phase, RMSIFL allocates disk space for three sort
work files, each 1.5 times the input file size.

NOTE
Performance in this phase depends entirely on the
SORT-11 modules. See the PDP-11 SORT Reference
Manual for more information on the space required for temporary sort files and guidance on the factors affecting sort
performance.
RMSIFL notes this phase by printing the following messages:
SORT HAS STARTED
SORT MERGE PHASE HAS FINISHED

If an error occurs while the SORT-11 modules are operating, RMSIFL

terminates with the following message:
SORTS ERROR CODE IN OCTAL:nnnnn

Refer to the PDP-11 SORT Reference Manual for a description of the
SORT-11 error codes.
At the end of this phase, RMSIFL deletes the sort work files.

9-58

RMS-11 Utilities

3. RMSIFL loads the ordered input records into the output file, constructing
Level 0 of the Primary index (discussed in "Data," Section 5.2.1) a bucket
at a time. The manner in which RMSIFL reads input records depends on
whether the input file was sorted in phase 2:
• If RMSIFL sorted the input file, the utility accepts the input records
one at a time from the sort modules. This technique avoids the need for
a temporary file containing the input records sorted in output Primary
Key order.
• If RMSIFL did not sort the input file, the utility reads the records from
the input file.

RMSIFL examines each input record to ensure that it is compatible with
the output file:
• If the record is not compatible, RMSIFL processes it as an exception
record according to the switch you specified in the command string. See
the discussion of the ER:[filespec] and NOER Infile Switches in Section
9.5.4.3 for the causes of exception records and their handling.
• If the record is compatible, RMSIFL inserts it into a Level 0 bucket
being built in memory.

As the utility places each record, the utility extracts the values for each of
the Alternate Keys defined for the file, if any:
a. RMSIFL combines the Alternate Key value with the Record's File
Address (RFA) of the user data record that supplied the key value.
RFAs are six bytes long. The utility then appends a byte with the key
number.
b. These fixed-length Alternate Key records are written into a separate
temporary file marked for delete. RMSIFL creates this file with an
initial allocation of 100 blocks and a Default Extension Quantity of 100
blocks. You can calculate the final size of this temporary file (ALQ) as
follows:
ALQ = (NDRF * NAK * (LKS + 7 + PAD))/512
where:
NDRF is the number of user data records in the input file,
NAK

is the number of Alternate Keys defined for the output file,

LKS

is the size of the longest Alternate Key defined for the output
file (and shown by RMSDSP), in bytes, {plus 1 for packed
decimal keys) and

PAD

is the number of null bytes added to word align the record:
PAD = 0 if LKS is an even number
PAD = 1 if LKS is an odd number

RMS-11 Utilities

9-59

Remember to round ALQ to the nearest multiple of 100.
During this process, the utility assembles the index levels of the Primary
index as well. RMSIFL puts an index record into a Level 1 bucket for each
Level 0 bucket it writes to the file. When it writes a Level 1 bucket to the
file, the utility inserts an index record in a Level 2 bucket, and so on, up to
and including the Root.
4. RMSIFL sorts the Alternate Key temporary file by key number and
ascending key value. Nonstring key data types are converted into a stringlike representation for sorting only.
At the beginning of this phase, RMSIFL allocates the disk space for sort
work files, each 1.5 times the size of the Alternate Key temporary file.
The utility notes this phase by printing the following messages:
ALTERNATE KEY<S>
SORT HAS STARTED
SORT MERGE PHASE HAS FINISHED

At the end of this phase, RMSIFL deletes the sort work files.
5. RMSIFL builds the Alternate indexes one at a time, from Level 0 up to
the Root, using the sorted temporary file for input.
The utility notes the construction of each Alternate index by printing the
following message:

ALTERNATE KEY<S> nnn

where:

nnn starts with 1.
~.

6. RMSIFL prints the following messages when the output file is completely
built, then terminates the processing of the current command string.
NUMBER OF INPUT RECORDS: nnnnnnnnn
NUMBER OF OUTPUT RECORDS: nnnnnnnnn
NUMBER OF EXCEPTION RECORDS: nnnnnnnnn

NOTE

RMSIFL extends the output file if it cannot contain the input
records and the index(es). The utility minimizes this overhead
by extending the file a set quantity each time it requires space.
This quantity is the integral multiple of bucket size greater
than 50. On RSTS/E, RMSIFL forces a contiguous output file
to be noncontiguous, then extends it.

9-60

RMS-11 Utilities

9.5.3 Call and Termination
9.5.3.1 Permanently Installed Utlllty

IFL [command string]
If you include a command string, the utility attempts to execute it and then

returns control to the current keyboard monitor or command interpreter.
If you do not specify a command string, RMSIFL assumes control of the user

interface and prints the prompt:·
IFL>

You may type a command string or ~ to terminate the utility. When
RMSIFL has executed a command string, it reprints the prompt.
9.5.3.2 Unlnstalled Utlllty

RUN $RMSIFL
RMSIFL assumes control of the user interface and prints the prompt:
IFL>

You may type a command string or ~ to terminate the utility. When
RMSIFL has executed a command string, it reprints the prompt.

9.5.4 Command String
9.5.4.1 General Form

outfile[/switch] .. ]=infile[/switch] ..]
where:

outfile

is the filespec of an existing Indexed file that is to receive records
from the input file. Wild card characters are not permitted in any
field of the specification. The default version is the highest version.
RMSIFL checks for the following conditions in the output file. If
they are not met, the utility prints the indicated error message and
terminates execution of the command string.
• The output file must exist before you run the utility; otherwise,
RMSIFL prints the message:
? I FL - - FI LE NOT FOUND - filespec

• The output file must have no data in it besides the Prologue
written when the file was created. Otherwise, RMSIFL prints the
message:
?IFL --

INDEHED OUTPUT FILE NOT EMPTY

RMS-11 Utilities

9-61

• The output file must be an RMS-11 Indexed file. Otherwise,
RMSIFL prints the message:
? I FL - -

OUTPUT FI LE MUST BE AN INDEXED FI LE

• The output file must not have a bucket size greater than five
blocks. Otherwise, RMSIFL prints the message:
? I FL

- -

THERE IS NOT ENOUGH I FL MEMORY FOR THE CURRENT

COMMAND LI NE

• The output file cannot have more than ten keys defined. Otherwise, RMSIFL prints the message:
? I FL - -

TOO MANY KEYS

filespec

is the filespec of the input file that is the source of the records
written to the output file. Wild card characters cannot appear in any
field of the filespec.

infile

You can specify only one input file, but it can have any RMS-11 file
organization. RMSIFL does not affect the input file in any way.
The utility opens the input file access read, allow no-write. Therefore, no writing task can access the input file while RMSIFL is
operating.
The input file cannot have more than ten keys defined. Otherwise,
RMSIFL prints the message:
? I FL - -

TOO MANY KEYS

filespec

You cannot use a back-up file produced by RMSBCK as input.
RMSIFL terminates with the following error message:
? I FL - -

0 PEN ERR 0 R 0 N F I LE

?IFL --

FATAL RMS ERROR

files pee

STS = -01424 tSTV

= 00000

may be a code shown in Table 9-4 and described in Section 9.5.4.2
through 9.5.4.4.

switch

NOTE
A command string may also consist of the word "HELP" or a
question mark (?). RMSIFL responds with a HELP message.
9.5.4.2 Global Switches -

• ID causes RMSIFL to print its current version number, in the form:
I FL

- -

1.JERS ION 1. Bnn

where:
nn is the patch level of the utility (see "Patch Level," Section 9.0.2).

This switch may appear alone as a command string.

9-62

RMS-11 Utilities

'fable 9-4: RMSIFL Utility Switches

Type

Default
Process

Description

Switch

String

? or HELP

Print HELP message.

No help.

Global

/ID

Identify current version

No id.

Outfile

/ER

Write exception records on terniinal.

Same.

/ER:filespec

Write exception records into
specified file.

Write exception records on terminal.

/LO

Honor fill numbers.

Fill buckets.

/NOER[:SJ

Stop processing immediately if
input record is incompatible.

Write exception records on terminal.

/PD[:[#Jxl

Pad input records to output
record length.

Handle input as exception record if different lengths.

/TR

Truncate input records to output record length.

Handle input as exception record if different lengths.

/CL:nnn

Set sort work files' clustersize.

Use algorithm.

!DE:dvn[:dvn[:dvn[:]]]

Reassign devices for sort work
files.

Create and use sort
work files on SY:

/KR:n

Key of reference number.

Primary Key (n=O).

/NOSO

Do not sort before loading.

Sort input file before
loading.

Infile

9.5.4.3 Outflle Switches

• ER: [filespec] directs RMSIFL to continue processing if it encounters a record in the input file that cannot be written into the output file. Exception
records exist for one of the following reasons:
•

You specified the NOSO Infile Switch and the utility found a record that
was not in ascending order by output Primary Key value. The error
message is:
I FL - - E}-{CE PT I ON RECORD:

•

You did not allow duplicates in at least one key of the output file, but an
input record contains a duplicate value in that key field. The error message is:
!FL -- EHCEPTION RECORD:

•

RECORD OUT OF SEQUENCE

DUPLICATE KEY WHERE NOT ALLOWED

An input record is not long enough to contain the Primary Key of the
output file. The error message is:
I FL - - EXCEPT I ON RECORD:

RECORD TOO SHORT FOR PR I MARY KEY

RMS-11 Utilities

9-63

•

An input record is too long to fit in a bucket of the output file. The error
message is:
I FL - -

•

D(CE PT I ON RECORD:

An input record is not compatible with the output file's fixed-length
record format and you have not specified the PD and/or TR switches.
The error message is:
I FL - - EHCE PT I ON RECORD:

•

RECORD DOESN'T FIT FI >(ED LENGTH

An input record is longer than the output file's variable-length Maximum Record Size and you have not specified the TR switch. The error
message is:
IFL -- EHCEPTION RECORD:

•

RECORD TOO LONG FOR BUCK ET SIZE

RECORD TOO LONG FOR MM( I MUM RECORD SIZE

An input record contains an illegal packed decimal key: a digit half-byte
is greater than 9 or the sign half- byte is less than 10. The error message
is:
IFL -- D(CEPTION RECORD:

RECORD CONTAINS AN ILLEGAL PACKED KEY

If RMSIFL encounters an exception while processing the Alternate
index(es), it flags as deleted the data record in the Primary Level 0. The
utility does not remove the record from the file. The record may be compressed during later use of the Indexed file, as described in "Key Selection,"
Section 6.2.
You can add corrected exception records to the output file after RMSIFL
terminates with either an application program or RMSCNV (possibly using
your terminal as the input device; see Section 9.2.5).
If you do not specify an argument (filespec), RMSIFL prints the appropriate
error message, then the exception record on your terminal.
If you specify a nondisk file or a terminal, the utility creates it as an
RMS-11 Sequential file with variable-length records when the first exception record is encountered. Then, RMSIFL writes both the error message
and the exception record into the file. RMSIFL uses this procedure for every
exception record until it finishes processing.
If you specify a disk file, the utility creates it as an RMS-11 Sequential file
with VFC records when the first exception record is encountered. ·Then,
RMSIFL writes the exception record into the variable portion of the record
and a four-byte exception code into the fixed control area of the record. The
codes are:

Code Meaning

001:
002:
003:
004:
005:
006:
007:

9-64

Record is out of Primary Key sequence
Record contains an illegal duplicate value
Record is too short to contain output Primary Key
Record is too long to fit in output bucket
Record is not correct size for output fixed-length record
Record is too long for output Maximum Record Size
Record contains an illegal packed decimal key.

RMS-11 Utilities

See also the NOER switch.

Default If you do not include the ER switch, RMSIFL prints the appropriate error message, then the exception record on your terminal.
• LO directs RMSIFL to write records into buckets according to the fill numbers established when the file was created. See "Data Allocation" in
RMSDEF, Section 9.3.4.6, and "Fill Number," Section 6.7.2.1.

Default If you do not include the LO switch, RMSIFL inserts as many
records as possible into each bucket.
• NOER[:S] directs RMSIFL to print an error message, delete the output file,
and terminate processing immediately if it encounters an input record that
cannot be written into the output file; see the ER: [filespec] switch for reasons and error messages.
RMSIFL deletes the output file unless you specify the Save argument (:S)
in the NOER switch. Termination at this point leaves the output file incomplete: it is not a valid file of any organization and cannot be reused for any
RMS-·11 purpose.
Example Your output file is large and contiguous and you don't want to lose any of that
contiguous space to other users.

Default If you do not include the NOER switch, RMSIFL responds according to the version of the ER switch you used.
• PD[: [#]x] directs RMSIFL to pad records read from the input file to the
output file's record length before writing them to the output file. Padding
character is specified as follows:
Switch

Character

PD
PD:x
PD:#x

x is ASCII A-Z, 0-9, or special character except #, ? , and @

NULL
xis octal number 000 through 377 (40 for SPACE, 43 for#, 77 for?, and 100 for@)

You use the PD switch only when the output file specifies fixed-length
records.

Default If you do not include the PD switch, and the input records are
shorter than the output file's records may be, RMSIFL treats
them as exception records.
See EH and NOER switch in this section.
• TR directs RMSIFL to truncate records read from the input file to the
output file's Maximum Record Size before writing them to the output file.
The trailing bytes of the records are truncated.

Default If you do not include the TR switch, and the input records are
longer than the output file's record may be, RMSIFL treats them
as exception records.
See ER and NOER switch in this section.

RMS-11 Utilities

9-65

9.5.4.4 lnflle Switches • CL:nnn specifies a clustersize for RMSIFL to use when it creates work files

to sort input records and its special Alternate Key records. Large files require large clustersizes so that the User File Directory can contain the
retrieval pointers.
Example With a pack clustersize of 16, a file 80,000 blocks long cannot be created.

The clustersize you specify must be a 0 or a power of two from 1 through
256; otherwise, RMSIFL terminates with the following message:
? I FL - -

THE CLUSTERS I ZE IS GREATER THAN 256. OR NOT A POWER OF 2

If you specify a value less than the pack clustersize for the disk on which the

sort work files are created, RMSIFL uses the pack clustersize.
Default If you do not include the CL switch, RMSIFL uses the following

algorithm to calculate a clustersize:
1. Evaluate the expression:
(FILESIZE * 1.5)/512
where:
FILESIZE is the size of the input file or the Alternate Key
temporary file.
2. Find the nearest power of two less than the value of the expression and not greater than 256.
• DE:dun[:dun[:dun[:]]] directs RMSIFL to reassign the devices where it allo-

cates the three sort work files. You can specify from one to three disk devices
with the switch, where dun is a physical device name and number or a
logical name up to six alphanumeric characters. You separate device names
by a colon (:); the final colon is optional.
If you include the DE switch in the command string, RMSIFL reassigns all

three sort work files:
•

If you specify one device name, the utility reassigns the three work files

to that device.
•

If you specify two device names, the utility reassigns one work file to the

first device and two work files to the second.
•

If you specify three device names, the utility reassigns one work file to

each device.
For optimum performance, you should assign each sort work file to a different device. However, if you have only two devices available, assign the first
and third files to one device and the second file to the other device.
Example /DE:DMO:DM1:DM2
Example /DE:DMO:DMl :DMO

9-66

RMS-11 Utilities

Default RMSIFL allocates the sort work files on SY:.
• KR:n directs RMSIFL to read an input Indexed file according to the key
specified by n, where n is 0 for the Primary Key, 1 for the First Alternate
Key, and so on, up to 9.
The KR switch can eliminate the need for sorting the input file; see
"Effect," Section 9.5.2.
If you use the KR switch with a non-Indexed file, RMSIFL terminates with

the following message:
?IFL -- /KR NOT ALLOWED FDR SEQUENTIAL OR RELATil.JE FILE

Default If you do not include the KR switch, and the input file is Indexed,
RMSIFL reads the file via the Primary index.
• NOSO directs RMSIFL to bypass its sort phase of processing because you
have already sorted the input file. If the input file is not in order by ascending value of the output file's Primary Key, RMSIFL's response depends on
whether you have included the NOER switch or a version of the ER switch
(see Section 9.5.4.3).

Default If you do not include the NOSO switch, RMSIFL sorts all records
from the input file into the proper order. See "Effect," Section
9.5.2.

9.5.5 Cautions
• RMSIFL ignores the fixed control area of VFC records in either input or
output file.
• When both the input and output files have fixed-length records, RMSIFL
requires either the TR or PD switch if the fixed record lengths are not equal.
If neither switch is specified, the utility terminates with the error message:
? I FL - -

IN PUT AND OUT PUT RECORD SIZES DO NOT CORRESPOND

• When the input file contains variable-length records and the output file
contains fixed-length records, RMSIFL requires both the TR and the PD
switches. If both are not specified, the utility terminates with the error
message:
? I FL - - SW ITCH /TR OR I PD OR BOTH ARE NEEDED FOR TH IS RECORD

9.6 RMSRST Command Utility
9.6.1

Purpose

The RMSRST utility reverses the action of the RMSBCK utility. Where
RMSBCK creates back-up copies of files, RMSRST takes those back-up files
as input and produces standard RMS-11 files as output. The structure, content, and attributes of these restored files are those of the original files when
they were backed up.
RMS-11 Utilities

9-67

9.6.2 Effect
RMSRST provides you with the following capabilities:
Explicit and Implicit File Selection
You restore a single file or a collection of files with each call of the
RMSRST utility. Multiple files are specified explicitly with a string of file
specifications or implicitly with wild cards. You may select from a wild
card group of files by name or on the basis of back-up date.
Choice of Restoration Account and/or Volume
You may restore file(s) into any account on any disk device to which you
have proper access privileges.
Extended Diagnostic Messages
When the utility operates in the Query mode, RMSRST gives you the
option of continuing or terminating processing when special errors occur
(described in Section 9.6.4.2). You thereby ensure that the utility does not
terminate the processing of a collection of files due to errors in the processing of one file. If the Query mode is explicitly disabled, RMSRST terminates when it encounters the special errors. See the QU switch in Section
9.6.4.2 for examples of diagnostic messages.
Data Integrity Checks
RMSRST can perform extensive data integrity checking as each back-up
file is restored. You can direct the utility either to read file contents after
they are restored or to both read and check file contents on a byte-by-byte
basis after they are written. RMSRST automatically retries read errors if
processing continues to a normal termination.
These integrity checks give you a choice about how reliable the restored
files are:
• You can rely on software and hardware accuracy and restore your files in
minimum time.
• You can verify that you can read the restored files, adding the read time
to the minimum time.
• You can guarantee that the restored files can be read and that they
match the back-up files, adding both read and compare times to the
minimum time.
Summary Listing
You can specify that RMSRST list a summary of the files selected and
restored and of error messages. This summary can be generated at your
terminal or written into a file specified in the utility's command string.
See the SL switch in Section 9.6.4.2 for a sample of the summary.

9-68

RMS-11 Utilities

9.6.3 Call and Termination
9.6.3.1

Permanently Installed Utlllty -

RST [command string]
If you include a command string, the utility attempts to execute it and then

returns control to the current keyboard monitor or command interpreter.
If you do not specify a command string, RMSRST assumes control of the user

interface and prints the prompt:
RST>
l

You may type a command string or (CTRL/zl to terminate the utility. When
RMSRST has executed a command string, it reprints the prompt.
9.6.3.2 Unlnstalled Utlllty -

RUN $RMSRST
RMSRST assumes control of the user interface and prints the prompt:
RST>

You may type a command string or (CTRL/zl to terminate the utility. When
RMSRST has executed a command string, it reprints the prompt.

9.6.4 Command String
9.6.4.1

General Form -

outfile[/switch] .. ]=infile[/switch] ..][,infile[/switch] ..]... ]
where:

outfile

is the filespec of a file(s) to be restored by the utility. The file-name
and extension must be wild cards, that is, *.*;therefore, you cannot
rename files as they are restored by RMSRST. The version can be
omitted or can be a wild card. In both instances, the version used is
that of the original file.
RMSRST interprets the account field as follows:
Account Number
None

Files Restored To
Default account.
Original accounts from which the files were backed up;
the numbers are stored in the back-up files.

Explicit

infile

Specified account.

is the file specification of a back-up container file or a disk file in
back-up format. You can use wild cards to specify multiple files. If

RMS-11 Utilities

9-69

version is omitted and the input file resides on magnetic tape (a
container file), the following defaults are applied:
• If wild cards appear in either or both file-name and extension, all

versions of all files satisfying the wild cards are selected.
• If both file-name and extension are explicitly specified, the first

physically occurring instance of the file with version 0 is selected.

NOTE
To restore a specific data file from magtape, you must
use the container file-name as the infile specification
and specify the data file-name in the SE infile switch.
switch

may be a code shown in Table 9-5 and described in Sections 9.6.4.2
through 9.6.4.4.

NOTE
A command string may also consist of the word
"HELP" or a question mark (?). RMSRST responds
with a HELP message.

Table 9-5: RMSRST Utility Switches

Switch

Description

Default
Process

String

? or HELP

Print HELP message.

No help.

Global

/ID

Identify current
RMSRST utility.

/QU

Enable Query mode.

/SL[:filespec]

Provide summary of activity.

No summary.

/FR

Change protection codes.

Original
protection.

IRA*

Read after writing.

NORA

/RC*

Check after writing.

NORC

/SU

Supersede existing files.

NOSU

!BD:dd-mon-yy

Restore disk files based on date
of back up.

No date
checking.

/OA: [act nbr]

Select files based on original
account.

OA: [*•*)

/SE:filespec or
/SE: (files pee, ... )

Select specified file(s) for restoration from back-up container
file.

SE:*·*;*

Type

Outfile

Infile

version

No id.

9-70

RMS-11 Utilities

9.6.4.2 Global Switches -

• ID causes RMSRST to print its current version number, in the form:
RST -- VERSION 1.Bnn

where nn is the patch level of the utility (see "Patch Level," Section 9.0.2).
This switch may appear alone as a command string.
• QU enables the Query mode (default). When the Query mode is enabled,
the RMSRST utility allows you to continue or terminate processing when
one of the following occurs:
•

A read error on an input file.

•

A read-after-write error on an output file when the RA switch is specified.

•

A check-after-write error when the RC switch is specified.

•

The table allocated internally for data integrity checks or automatic
retry of read errors is full.

When one of these errors occurs, the utility prints a diagnostic message
specifying the type of error and the name of the file being processed. If you
answer Y (yes), RMSRST continues processing the current file and command string. If you answer N (no), the utility terminates processing immediately, bypassing the rest of the command string.
Example

BCK - - CHECK AFTER WR I TE ERROR ON OUT PUT FI LE - filespec
t,lBN vbnl TO vbn2. CONT I NUE < Y t N)?

Example BC K - - I NT E GR IT Y CHECK TABLE FULL • C 0 NT I NU E < Y t N ) ?

When you disable the Query mode through the NOQU switch, RMSRST
prints the appropriate message and terminates processing after it encounters one of the errors.

Default If you specify no version of the QU switch, Query mode is enabled.
• SL[:filespec] directs RMSRST to provide a summary of processing, including:
-

the command string

the names of files successfully restored

the names of files not restored, with associated error messages

diagnostic messages produced in Query mode

a summary of input errors and of errors remaining in restored output
files following automatic retry of data integrity checks (when you specify
RA or RC).

If a file specification is not included in the switch, RMSRST prints the
summary listing on your terminal. If a file is specified, the utility creates the

file and writes the summary listing into it.

RMS-11 Utilities

9-71

Example RMSRST - VERB I ON 1. 800 22-JAN-1978 1 Lt: 19: 00
*•*=DB1:*+*/SL:RST.LST
RST--FILEPROCESSINGCOMPLETE-SY:[303t357JSPEWDOC.CMD;L13
RST - - FI LE PROCESS I NG COMPLETE - SY: [ 303 , 357 J TECO. TEC; 1
RST -- FILE ALREADY E}<ISTS - SY:[3031357JUSERGUIDE.Rl.JW;3
RST -- FILE ALREADY EHISTS - SY:C3031357JREFMNLUPO,Rt.iw;s

Default If you specify no version of the SL switch, RMSRST only prints

messages on your terminal if it encounters a fatal or non-fatal
error condition.
9.6.4.3 Outflle Switches -

• FR directs RMSRST to change the restored file's system protection codes.
RMSRST modifies the protection code in the back-up file when it restores
the data to disk. If the files are restored with the operating system they were
backed up with, the utility changes the protection code so the default
account, the one operating the utility, becomes the owner of the files. If the
restoring operating system is different, RMSRST sets the protection code to
the system default.

~·~

NOTE
If files are restored to an account other than the one used
when they were backed up and FR is not specified, the utility
assigns the files to the indicated account (see ·"General
Form," Section 9.6.4.1), but the protection code is not
changed, that is, the original owner is still the owner, and so
on.
Default If you do not include the FR switch, RMSRST does not change the

protection code from the value copied when the file was backed
up.
• RA directs RMSRST to read the restored file after writing it. The utility
reads back each block of a file after it has been completely restored: if the
device hardware detects a read error, RMSRST's reaction depends on
whether the Query mode is enabled or not. See QU under "Global Switches,"
Section 9.6.4.1.

"""'

NOTE
You cannot use the RA and RC switches in the same command string. RMSRST prints the following error message
and terminates execution of the command string.
?BCK - - CONFLICT I NG 0 PT ION - I sw

where sw is RA or RC depending on which is second in the
command string.
Default If you do not include the RA switch, RMSRST does no read-after-

writer data integrity checking.

9-72

RMS-11 Utilities

_ _, __ _ _ _ _ _ _ _ _ _ _ _. ._ _ _ _ _ _ _ _ _..._ _ _ _ _ _ _" "_ _ _ _ _ _ _ _ _ _lllllllll_!lmllll

• RC directs RMSRST to check the restored file after writing it. After a file
has been completely restored,. the utility reads each block of the file back
into memory and compares it with the corresponding block from the input
back-up file. If an error occurs during this process (a hardware read error or
a mismatch between the contents of the two blocks), RMSRST's reaction to
the error depends on whether the Query mode is enabled or not. See QU
under "Global Switches," Section 9.6.4.1.
See note under RA in this section.

Default If you do not include the RC switch, RMSBCK does not check the
restored file after writing it.
• SU causes RMSRST to supersede each file in the output account with the
same file-name and extension as a file being restored.

Default If you do not include the SU switch, :f{MSBCK does not supersede
files in the output account. If the utility encounters a file with the
same file-name, extension, and version as an input file, it prints
the following nonfatal error message and continues processing.
RST - - FI LE ALREADY EH I STS - filespec

9.6.4.4 lnflle Switches -

• BD:dd-mon-yy directs RMSRST to restore disk files based on back-up
date. You can combine this switch with wild cards to identify specific files
within a group of disk files. RMSRST selects for processing only those files
that satisfy the wild card specification and that were backed up on the
specified date.
The BD switch cannot be used if the back-up files are stored on magnetic
tape. Since RMSRST cannot access the creation date of a magnetic tape
file, it prints the following error message if BD is included in a magnetic
tape infile specification:
?RST - ILLOGICAL USE OF SWITCH - /BO

Default If you do not include the BD switch, RMSRST applies no date
criterion to the selection of files for restoration.
• OA: [act nbr] directs RMSRST to select files based on the account from
which the original files were backed up. You can combine this switch with
wild cards to identify specific files within a group of disk or tape files.
RMSRST selects for processing only those files that satisfy the wild card
specification and that were backed up from the specified account.

Default If you do not include the OA switch, RMSRST applies no account
number criterion to the selection of files for restoration.

RMS-11 Utilities

9-73

• SE:filespec or SE:(filespec, ... ) directs RMSRST to select specified file(s) for
restoration from magnetic tape container files only. This switch allows you
to select explicitly one or more files from container files. The switch argument, filespec, can contain only file-name, extension, and version; any can
be a wild card. Up to ten filespecs may be included, enclosed in parentheses
and separated by commas if more than one.

If version is omitted from filespec, RMSRST applies the following defaults:

•

All versions of the file represented by filespec are restored if any of the
following are true:
- The file-name or extension in filespec is a wild card.
- The OA switch is not specified for the container file.
- The OA switch is specified for the container file, but contains wild
cards.

•

The first physically occurring version ofthe file represented by filespec is
restored if:
- Neither the file-name or extension in filespec is a wild card.
- The OA switch is specified and contains no wild cards.

Default If you do not include the SE switch, RMSRST uses all contained
files as input to the restoration process.
9.6.4.5 Command String .Examples -

• *·*=MTO:ALPHA.DAT
RMSRST searches the volume mounted on MTO: for the first physically
occurring version of container file ALPHA.DAT. The utility restores the
contained files to their original form into the current account on SY:. Since
the SU switch was not specified, the current account on SY: must not
contain a file with the same file-name, extension, and version as a file being
restored: RMSRST does not restore duplicate files; instead, it prints the
following error message and continues processing.
RST

- -

FI LE ALREADY E}O STS

filespec

• *.4SU=MT1:*.*
RMSRST searches the volume mounted on MTl: for all container files. For
each such file, it restores the contained files into the current account on
SY:. If the current account contains a file with the same file-name, extension, and version as a file being restored, the disk file is superseded.
• *.4SU/SL=DK0:*.4BD:30-0CT-78,MTO:*.*
RMSRST reads all files under the current account on DKO: that were
backed up on October 30, 1978, and restores them to the current account on
SY:. Then the utility restores files within all container files on the tape

9-7 4

RMS-11 Utilities

mounted on MTO:. Throughout the process, the utility supersedes any existing file with the same file-name, extension, and version as the file being
restored. RMSRST also prints a summary of activity at your terminal.
• [100,101*·*/RC/SU=MTO:*.*/SL:[lOO,lOJREST.LST
RMSRST restores all files within all container files on the tape on MTO:
into account [100, 10] on SY:, superseding disk files with the same filenames, extensions, and versions. The utility checks after writing each restored file. Since Query mode is enabled by default, diagnostic messages
may appear at your terminal as a result of these checks. The utility also
creates a file named REST.LST in the output account and writes a summary of its activity into it.
• *·*=MT2:MASTER.BKP/SE:(PARTS.DAT,CUST.FIL)
RMSRST searches the volume mounted on MT2: for the first physically
occurring instance of the container file MASTER.BKP. When this file is
found, RMSRST searches MASTER.BKP for all versions of files
PARTS.DAT and CUST.FIL. These files are restored to their original form
in the default account on SY:.
• *.4SU/FR=MTO:ACCT.BKP/SE:*.DAT */OA:[l00,111
RMSRST restores all versions of all files with an extension of DAT and an
original account of [100, 11] into the default account on SY: from container
file ACCT.BKP on MTO:, changing the protection code as it does.· The
utility supersedes files in the default account if necessary.
• DK2:[*,*1*·* *=MTl:*.*
RMSRST replaces all files within all container files on MTl: into their
original accounts on DK2:.
• [200, 111*·4FR=MT3:A200.BAC/OA: [200, 111
RMSRST rewrites all files with an original account of [200,11] in container
file A200.BAC into account [200,11] on SY:. The tape on MT3: was created
with the RMSBCK utility on the RSTS/E operating system, but the files
are being restored with RSX-llM. RMSRST therefore changes the protection code on each file to the system default as the file is written onto the
system device.

9.6.5 Cautions
• Files can be restored only on disk devices.
• A system disk cannot be restored as a bootable volume.
• A file contiguous when backed up will not be contiguous when restored.
However, you can use RMSDEF and RMSCNV to make a contiguous file.
• Do not use either the infile or outfile specifications as an argument in the SL
switch.

RMS-11 Utilities

9-75

·~:

App~endix A

RMS-11 and the Operating Systems

RMS-11 is available as an interface between user application programs and
data storage devices on the following DIGITAL operating systems.

A.1

IAS
A.1.1

RMS-11 Restrictions on IAS

None.

A.1.2 IAS Restrictions on RMS-11
None.

A.1.3 Compatlbillty with Other Fiie Managers
In addition to RMS-11, File Control Services (FCS-11) is available on IAS
systems. FCS-11 and RMS-11 can reside on the same computer system and
one program can use both record access systems with caution. The structure
of RMS-11 Sequential files is identical to that of files created by FCS-11.
Therefore, you can use FCS-11 to access RMS-11 Sequential files and viceversa.
However, if you use FCS-11 to open an RMS-11 Relative or Indexed file,
FCS-11 will logically destroy the file when it closes the file.
Within the Sequential organization, the following RMS-11 record formats are
compatible with FCS-11:
RMS-11

FCS-11

Fixed
Variable
VFC

FLR
VLR
SVLR (fixed control area must be equal to two bytes)

A-1

A.1.4 Asynchronous Operations
IAS allows tasks to perform operations asynchronously. You use this technique so that your task can perform functions during 1/0 operations, rather
than wait until the I/O activity is over.
A.1.4.1 RMS-11 Synchronous Environment - By default, the Task Builder
incorporates the synchronous RMS-11 routines into tasks. These routines do
not enable or disable Asynchronous System Traps (ASTs). You can program
routines that run at AST level; however, those routines must not:

• initiate an RMS-11 operation while another is still in progress. RMS-11
synchronous routines are vulnerable to this type of procedure: the results of
the interrupted operation are unpredictable. Therefore, you should avoid
using RMS-11 routines at program level and AST level at the same time.
To enable simultaneous operations at the different levels, you must take
special precautions to avoid interference between the operations.
• cause the task segment containing the RMS-11 routines to be overlaid while
an RMS-11 operation is in progress. The results in such a situation are
unpredictable.
A.1.4.2 RMS-11 Asynchronous Environment - You can task build with the
RMS-11 Asynchronous Executive modules (see Nonoverlaid RMS-11 in
"Task Building with RMS-11 Routines," Section 8.1 for details), thereby
creating the RMS-11 Asynchronous Environment. File operations must still
be performed synchronously. In order to perform asynchronous record operations, you must also set up an asynchronous Record Access Block (RAB) and
turn on the RB$ASY in the RAB ROP field. See the RMS-11 MA CR0-11
Refere nee Manual for details.

When your task initiates an RMS-11 asynchronous record operation, RMS-11
disables Asynchronous System Traps (ASTs) while it is running. RMS-11
enables ASTs when it must wait and when it returns control to your program.
While your program is in control during RMS-11 I/0 operations, the program
can disable ASTs for critical operations. However, the outstanding RMS-11
operation typically will not be able to complete until ASTs are enabled again.
The RMS-11 Asynchronous Environment restricts operations as follows:
• You can initiate asynchronous RMS-11 operations at AST level provided
the associated Record Access Stream is not in use. However, operations
begun at AST level cannot complete until the task reverts to program level.
Therefore, any associated wait operation ($WAIT macro) must not be performed at AST level.
• Any completion routine associated with an asynchronous record operation is
typically executed at AST level and is considered part of the record operation:
-

A-2

The completion routine cannot initiate a synchronous RMS-11 operation. RMS-11 returns the error code ER$AST.

RMS-11 and the Operating Systems

·~

The completion routine cannot use the $WAIT macro. RMS-11 aborts
the task.

Asynchronous RMS-11 operations initiated by the completion routine
cannot complete until control is returned to RMS-11. You do this by
issuing the $RETURN macro as the end of the completion routine.
RMS-11 resumes control and performs an exit from AST level. Control
returns to your program, although an AST is still outstanding because of
the asynchronous operation initiated by the completion routine.

• Be careful initiating synchronous RMS-11 operations from the AST level
while outside a completion routine. You must ensure that no resource required by the operation is exclusively controlled by an operation initiated at
program level. Since the operation performs at AST level, it cannot be
interrupted: the operation can loop continuously, waiting for that resource
to be freed.
Example Your program initiates an asynchronous record operation on an Indexed file.
When control returns to the program, it initiates a synchronous operation on that
same Indexed file. If the asynchronous operation locks the file Prologue or a
bucket required by the synchronous operation, the synchronous operation loops,
repeating its request for the locked blocks. Since the asynchronous operation
cannot resume to unlock the blocks, the synchronous operation will never break
out of its loop.

"-"

A.2 RSTS/E
A.2.1

RMS-11 Restrictions on RSTS/E

RMS-11. does not allow switches on file specifications.

A.2.2 RSTS/E Restrictions on RMS-11
The following RMS-11 capabilities are not available on RSTS/E:
• Asynchronous 1/0 operations
• Magnetic tape volume handling after RMS-11 opens a file on the tape
Example Data integrity switches on the RMSBCK and RMSRST utilities.
Example The macros $REWIND (for magtape files only), $NXTVOL, and $SPACE.

• Multi-volume magnetic tape files
• Windowsize adjustment
• Extending contiguous files
• Placement control for file areas other than Area 0

RMS-11 and the Operating Systems

A-3

A.2.3 Compatibility with Other Fiie Managers
In addition to RMS-11, BASIC-PLUS performs file handling. A stream
ASCII (terminal-format) file created by BASIC-PLUS can be read as an
RMS-11 Sequential file with Stream record format as long as the file is nullfilled from the end of the last record to the physical end-of-file.

A.3 RSX-11M
A.3.1

RMS-11 Restrictions on RSX-11 M

None.

A.3.2 RSX-11 M Restrictions on RMS-11
None.

A.3.3 Compatibility with Other File Managers

~I
•f?

"t"""

•

In addition to RMS-11, File Control Services (FCS-11) is also available on
RSX-llM systems. FCS-11 and RMS-11 can reside on the same computer
system and one program can use both record access systems with caution. The
structure of RMS-11 Sequential files is identical to that of files created by
FCS-11. Therefore, you can use FCS-11 to access RMS-11 Sequential files
and vice-versa.
However, if you use FCS-11 to open an RMS-11 Relative or Indexed file,
fCS-11 will logically destroy the file when it closes the file.
Within the Sequential organization, the following RMS-11 record formats are
compatible with FCS-11:
RMS-11

FCS-11

Fixed
Variable
VFC

FLR
VLR
SVLR (fixed control area must be equal to two bytes)

A.3.4 Asynchronous Operations
RSX-llM allows tasks to perform operations asynchronously. You use this
technique so that your task can perform functions during 1/0 operations,
rather than wait until the 1/0 activity is over.
A.3.4.1 RMS-11 Synchronous Environment - By default, the Task Builder
incorporates the synchronous RMS-11 routines into tasks. These routines do
not enable or disable Asynchronous System Traps (ASTs). You can program
routines that run at AST level; however, those routines must not:

• initiate an RMS-11 operation while another is still in progress. RMS-11
synchronous routines are vulnerable to this type of procedure: the results of

"""'

A-4

RMS-11 and the Operating Systems

the interrupted operation are unpredictable. Therefore, you should avoid
using RMS-11 routines at program level and AST level at the same time.
To enable simultaneous operations at the different levels, you must take
special precautions to avoid interference between the operations.
• cause the task segment containing the RMS-11 routines to be overlaid while
an RMS-11 operation is in progress. The results in such a situation are
unpredictable.

A.3.4.2 RMS-11 Asynchronous Environment - You can task build with the
RMS-11 Asynchronous Executive modules (see Nonoverlaid RMS-11 in
"Task Building with RMS-11 Routines," Section 8.1 for details), thereby
creating the RMS-11 Asynchronous Environment. File operations must still
be performed synchronously. In order to perform asynchronous record opera tions, you must also set up an asynchronous Record Access Block (RAB) and
turn on the RB$ASY in the RAB ROP field. See the RMS-11 MACR0-11
Reference Manual for details.

When your task initiates an RMS-11 asynchronous record operation, RMS-11
disables Asynchronous System Traps (ASTs) while it is running. RMS-11
enables ASTs when it must wait and when it returns control to your program.
While your program is in control during RMS-11 I/O operations, the program
can disable ASTs for critical operations. However, the outstanding RMS-11
operation typically will not be able to complete until ASTs are enabled again.
The RMS-11 Asynchronous Environment restricts operations as follows:
• You can initiate asynchronous RMS-11 operations at AST level provided
the associated Record Access Stream is not in use. However, operations
begun at AST level cannot complete until the task reverts to program level.
Therefore, any associated wait operation ($WAIT macro) must not be performed at AST level.
• Any completion routine associated with an asynchronous record operation is
typically executed at AST level and is considered part of the record operation:
-

The completion routine cannot initiate a synchronous RMS-11 operation. RMS-11 returns the error code ER$AST.

The completion routine cannot use the $WAIT macro. RMS-11 aborts
the task.

Asynchronous RMS-11 operations initiated by the completion .routine
cannot complete until control is returned to RMS-11. You do this by
issuing the $RETURN macro as the end of the completion routine.
RMS-11 resumes control and performs an exit from AST level. Control
returns to your program, although an AST is still outstanding because of
the asynchronous operation initiated by the completion routine.

'-''
RMS-11 and the Operating Systems

A-5

• Be careful initiating synchronous RMS-11 operations from the AST level
while outside a completion routine. You must ensure that no resource required by the operation is exclusively controlled by an operation initiated at
program level. Since the operation performs at AST level, it cannot be
interrupted: the operation can loop continuously, waiting for that resource
to be freed.
Example Your program initiates an asynchronous record operation on an Indexed file.
When control returns to the program, it initiates a synchronous operation on that
same Indexed file. H the asynchronous operation locks the file Prologue or a
bucket required by the synchronous operation, the synchronous operation loops,
repeating its request for the locked blocks. Since the asynchronous operation
cannot resume to unlock the blocks, the synchronous operation will never break
out of its loop.

A.4 VAX/AME
A.4.1

RMS-11 Restrictions on VAX/AME

RMS-11 utilities do not support global wild card characters in file specifications; they do not follow subdirectories in all directories possibly covered by
the wild card indication.
However, RMS-11 utilities do support wild card characters within named
directories as well as within Master File Directories to the User File Directory
level, but not to the subdirectory level.

A.4.2 VAX/AME Restrictions on RMS-11
The following RMS-11 capabilities are not available on VAX/AME:
• Write sharing. As long as programs open RMS-11 files with access no-write
and allow no-write declarations, they can share the files. However, if the
first program declares write access, other programs are denied access to the
file; and if a reading program opens a file, no writing program is allowed to
open it.
• RMS-11 Resident Library

A.4.3 Asynchronous Operations
VAX/AME allows tasks to perform operations asynchronously. You use this
technique so that your task can perform functions during 1/0 operations,
rather than wait until the 1/0 activity is over.
By default, the Task Builder
incorporates the synchronous RMS-11 routines into tasks. These routines do
not enable or disable Asynchronous System Traps (ASTs). You can program
routines that run at AST level; however, those routines must not:
A.4.3.1

RMS-11 Synchronous Environment -

• initiate an RMS-11 operation while another is still in progress~ RMS-11
synchronous routines are vulnerable to this type of procedure: the results of

A-6

RMS-11 and the Operating Systems

A.4.3.2 RMS-11 Asynchronous Environment - You can task build with the
RMS-11 Asynchronous Executive modules (see Nonoverlaid RMS-11 in
"Task Building with RMS-11 Routines," Section 8.1 for details), thereby
creating the RMS-11 Asynchronous Environment. File operations must still
be performed synchronously. In order to perform asynchronous record opera tions, you must also set up an asynchronous Record Access Block (RAB) and
turn on the RB$ASY in the RAB ROP field. See the RMS-11 MA CR0-11
Reference Manual for details.

You can initiate asynchronous RMS-11 operations at AST level provided
the associated Record Access Stream is not in use. However, operations
begun at AST level cannot complete until the task reverts to program level.
Therefore, any associated wait operation ($WAIT macro) must not be performed at AST level.

·• Any completion routine associated with an asynchronous record operation is
typically executed at AST level and is considered part of the record operation:
-

The completion routine cannot initiate a synchronous RMS-11 operation. RMS-11 returns the error code ER$AST.

The completion routine cannot use the $WAIT macro. RMS-11 aborts
the task.

RMS-11 and the Operating Systems

A-7

• Be careful initiating synchronous RMS-11 operations from the AST level
while outside a completion routine. You must ensure that no resource required by the operation is exclusively controlled by an operation initiated at
program level. Since the operation performs at AST level, it cannot be
interrupted: the operation can loop continuously, waiting for that resource
to be freed.
Example Your program initiates an asynchronous record operation on an Indexed file.
When control returns to the program, it initiates a synchronous operation on that
same Indexed file. If the .asynchronous operation locks the file Prologue or a
bucket required by the synchronous operation, the synchronous operation loops,
repeating its request for the locked blocks. Since the asynchronous operation
cannot resume to unlock the blocks, the synchronous operation will never break
out of its loop.

A-8

RMS-11 and the Operating Systems

Appendix B
Rl\nS-11 and· the Programming Languages

B. ii

Implementation In Languages
.RMS-11 is the record management software for the following PDP-11 programming languages:
BASIC-PLUS-2
DIBOL
MACR0-11
PDP-11 COBOL
RPG II
All features of RMS-11 supported by an operating system (generally described in Table B-1) are available to the MACR0-11 programmer.
The other, higher level languages support a subset of RMS-11 features. The
specific implementation of these features and the syntax used to invoke them
are described in the language documentation.

B.:2 Function Chart
Table B-1 identifies PDP-11 programming languages and the RMS-11 features they support. It is not a definitive source for language capabilities. See
your latest language documentation for its current RMS-11 implementation.
The explicit information shown indicates features available through the language without the help of RMS-11 utilities or other software. However, the
flag # indicates that a user can specify the feature via an RMS-11 utility
(primarily RMSDEF) and then use it in the language.

B-1

Table B-1: RMS-11 Features Supported by Programming Languages
The notation SIR/I indicates that the feature is supported by the language for all file organizations, Sequential (S), Relative (R), and Indexed (I). Blanks are substituted for a letter when the
language does not support the feature for an organization.

Features

File Organizations:
SEQUENTIAL
RELATIVE
INDEXED
File Operations:
Create a file
Single area
Multiple areas
Initial allocation quantity
Default extension quantity
Contiguous allocation
Fill number
Open a file
By file specification
By file ID
Close a file
Delete a file
Examine an Indexed file's structure
Implicit file extension
Explicit file extension
Record Formats:
FIXED
VARIABLE
VFC
STREAM
UNDEFINED (BLOCK 1/0)
Record Access Modes:
SEQUENTIAL
RANDOM
RECORD FILE ADDRESS
DYNAMIC SHIFT BETWEEN
MODES
Record Transfer Modes:
MOVE
LOCATE

MACR0-11

BASIC
PLUS-2

COBOL

RPG II

DIBOL

YES
YES
YES

$CREATE
Yes
Yes
Yes
Yes
Yes
Yes
$OPEN
Yes
Yes
$CLOSE
$ERASE

OPEN
Yes
No#
Yes

Yes
Yes
No#

OPEN 1
Yes
No#

OPEN
Yes
No
CLOSE/END
KILL

OPEN
Yes
No#
Yes
Yes
Yes
Yes
OPEN
Yes
No
CLOSE

Yes 2
Yes
No
Yes 3
No

OPEN
Yes
No
CLOSE
XCALL
DELET

$DISPLAY
Yes
$EXTEND

Yes
No

SIR/I
SIR/I
SIR

SIR/I
SIR/I
No

SIR/I
SIR/I
SIR
S4
No

Y~s

SIR/I

s
s

SIR/I
SIR/I
No
No
No

SIR/I

Yes

SIR/I

SIR/I
R/I
No

Yes

Yes
Yes

Yes
No

A DIBOL program can create Sequential files only.
(continued on next page)
Automatically before RPG II cycle execution.
3
Automatically after RPG II cycle execution.
4 Sequential files with stream format records are accessible for input only by a DIBOL task.
1

B-2

RMS-11 and the Programming Languages

"""'

Table B-· 1: RMS-11 Features Supported by Programming Languages
(Cont.)
Features
Record Operations:
GET
PUT
Honor fill number
UPDATE
Honor fill number
FIND
DELETE
Release locked bucket
Set up Record Access Stream
Terminate Record Access Stream
Set pointer back to file beginning
Truncate Sequential disk file
Number of Record Access Streams
per file:
SINGLE
MULTIPLE
I/0 Techniques:
Synchronous 1/0
Asynchronous I/O
Multiple Block Count (SEQ only)
Root Caching (IDX only)
Mass Insert
Retrieval Window Specification
Clustersize
Index Keys:
MAXIMUM NUMBER
SEGMENTED KEYS
DATA TYPES
String
15-bit Signed Integer
31-bit Signed Integer
16-bit Unsigned Binary
32-bit Unsigned Binary
Packed Decimal
DUPLICATES:
Primary
Alternate
CHANGE VALUE DURING
UPDATE:
Primary
Alternate
NULL KEY VALUE
MATCH DURING RANDOM
ACCESS:
Exact
Approximate
Generic
Sharing:
SEQUENTIAL
RELATIVE
INDEXED
I/0 UNIT LOCKING
1/0 Buffer Handling:
CENTRAL TO TASK
PRIVATE FOR EACH FILE
DYNAMIC ALLOCATION FROM
ADDRESS SPACE
Magnetic Tape Operations:

MACR0-11

BASIC
PLUS-2

COBOL

$GET
$PUT

GET
PUT

READ
WRITE

Implicit
Implicit

Yes
$UPDATE
Yes
$FIND
$DELETE
$FREE
$CONNECT
$DISCONNECT
$REWIND
$TRUNCATE

Yes
UPDATE
Yes
FIND
DELETE
No
CONNECT
Yes
RESTORE
SCRATCH

No
REWRITE
No
START
DELETE
No
No
No
No

No
Implicit
No
CHAIN
No
No
No
No
No
No

READ/READS
WRITES
STORE
No
WRITE
No
No
DELETE
No
No
No
No
No

SIR/I

R/I

SIR/I
No

Yes
Yes

Yes
No
Yes
Yes
No
Yes
Yes

Yes
No
No
No
No
No
No

Yes

Yes
Yes
Yes
Yes
Yes

Yes
No
Yes
Yes
Yes
Yes
Yes

Yes
No
No
No
No

255
Yes

255
No#

255
No

225";
No

255
No

Yes
Yes
Yes
Yes
Yes
Yes

Yes
No
No
No
No
No

Yes
Yes

.Yes
Yes

No
Yes

Yes
Yes

No
Yes
Yes

No
Yes
No#

Yes
Yes
Yes

Yes
Yes
No

Yes
Yes
Yes

Multiple
readers
Read/Write
Read/Write
Automatic

Yes
Yes

No
No

Yes
Yes

No
No

Yes

DIBOL

RPG II

s RPG II task default is buffers for four keys; however, each time a program is executed, i.t can use only one key.

RMS-11 and the Programming Languages

B-3

8.3 ERROR CODE MAPPING
Each higher level language checks for RMS-11 completion codes after each
file and record operation. If the code indicates anything other than unconditional success, the language Object- or Run-Time System maps the RMS-11
error code to a value compatible with the language's error reporting structure.
Table B-2 shows:
• the RMS-11 error codes available to the MACR0-11 programmer and the
higher language OTS or RTS
• the equivalent higher level language error codes available to programmers
using those languages
• a brief explanation of the cause of the error. The descriptions often include
references to RMS-11 control structures familiar only to the MACR0-11
programmer.
When Table B-2 indicates that the Status Value (STV) field contains a file
processor code, you should refer to the description of such codes in one of the
following manuals:
• Error code appendix of the IAS/RSX-111/0 Operations Reference Manual
• User recoverable error messages in an appendix of the RSTS/E Programming Manual. Note that the value returned in STV is the negative of the
decimal value shown in the Programming Manual. That is, if STV contains
"-20 10 ,'' look up "20 10 ."
When Table B-2 indicates that the Status Value field contains an FSS directive error code, you should refer to the error code appendix of the RSTS/E
System Directives Manual for a description of the codes.

B-4

RMS-11 and the Programming Languages

·~

(

Table B-2: Language Error Code Mapping
Symbolic
Value

MACRO-II
Status Code

I ER$ABO I 177760s/-16
ER$ACC

177740s/-32

ER$ACT

177720s/-48

MACRO-II
Status Value

BASICPLUS-2

I ER$STK/ER$MAP I

255 1

File processor
error code

PDP-11
COBOL

RPG II

DIBOL

Description

I -16

Operation aborted: Stack save area exhausted or
memory-resident control structures corrupted.

162

-32

File processor error: File processor could not access
the file.

255 1

-48

File activity precludes operation.

Example You attempted to close a file before an
asynchronous record operation finished.

s=
UJ.
I
.........

ER$AID

177700s/-64

XAB address

255 1

-64

The area indicated by the XAB does not exist in the
file.

ER$ALN

177660s/-80

XAB address

255 1

-80

Illegal alignment value for Placement Control.

ER$ALQ

177640s/-96

XAB address

-96

ER$ANI

177620s/-112

128

Illegal allocation quantity: The quantity exceeds
65K blocks during file creation on a non-Large Files
RSTS/E system or equals zero during an explicit file
extension operation.

ER$AOP

177600s/-128

-112

Records in a file on ANSI-labeled magnetic tape are
variable-length, but not in ANSI-D format.

ER$AST

177560s/-144

255 1

-128

Invalid type of allocation.

-144

Invalid operation at AST level: You attempted to
issue a synchronous operation from an asynchronous
record operation completion routine.

-160

File processor error: Read failure on file attributes.

255

.........
$Xl

::s

0..
("'t-

::r"
(t)

XAB address

255

'"O

(Jq

.....
$Xl

8
8

5·

ER$ATR

177540s/-160

(Jq

t"4

$Xl

File processor
error code

252

File processor
error code

252

177530s/-168

::s

(Jq

i::

$Xl

ER$ATW

177520s/-176

(Jq
(t)

Invalid File ID. See ER$FID.

30
-176

File processor error: Write failure on file attributes.

-192

"File bucket size exceeds maximum for operating
system.

ER$BKS

177500s/-192

255 1

Cl:'
~
1

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

(continued on next page)

Table B-2: Language Error Code Mapping (Cont.)

Symbolic
Value

MACRO-II
Status Code

MACR0-11
Status Value

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

Description

ER$BKZ

177460s/-208

XAB address

255 1

-208

Area bucket size exceeds maximum for operating
system.

ER$BLN

177440s/-224

255 1

-224

Control block (FAB, RAB, or XAB) length is
invalid.

ER$BOF

177430s/-232

129

-232

Beginning of file detected during magnetic tape
spacing operation.

ER$BPA

177420s/-240

255 1

-240

3
3

Invalid 1/0 buffer: Private buffer pool not on word
boundary.

ER$BPS

177400s/-256

255 1

-256

Invalid 1/0 buffer: Private buffer pool size not a
multiple of two bytes.

ER$BUG

177360s/-272

255 1

-272

RMS-11 aborts your task because it detected an
internal error. Contact a DIGITAL Software
Specialist.

ER$CCR

177340s/-288

255 1

-288

You attempted to connect more than one record -access stream to a Sequential file.

ER$CHG

177320s/-304

130

-304

100

During an update operation, you attempted to
change a key field that does not allow changes.

ER$CHK

177300s/-320

-320

Indexed file bucket corrupted. Do as many of the
following steps as are necessary:

00
I

P-4
P-4

$:D

::s

0...
M--

::r
('D
""d

a
$:D

Er
aq
$:D

::s
aq
c::
$:D

('D

r:tJ

1. Move the disk pack containing the file to another
device and try the process again. If it works, the
error was caused by a hardware read' failure.
2. Recreate the file using the RMSIFL or RMSCNV
utility. If this works, the corrupted bucket was in
an index bucket not used during sequential access by Primary Key.
3. Restore the file from your last backup copy.

ER$CLS

177260J-336

File processor
error code

-336

File processor error: During RMS-11 file close
operation.

(continued on next page)

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

)

-~)

(

Table B-2: Language Error Code Mapping (Cont.)

I Q~"h.i.H~ -1 MAPIUL11

..LT.&.C-..'-".&.•'-.1- .I. .I.

Value

Status CodE;

Status Value

·!:u..~u-·_
PLUS-2

ER$COD

177240s/-352

XAB address

255 1

ER$CRE

177220s/-368

File processor
error code

162/16

ER$CUR

177200s/-384

131

ER$DAC

177160s/-400

File processor
error code

ER$DAN

177140s/-416

XAB address

ER$DEL
ER$DEV

::0
~

- UJ.i.i..ii.IV.i..&...,.

M,H..,»lLn

J.Y.Lr.1.'-J.I."''-'- .I. .I.

~r11J.A.'-.!-

...um~_u
....,... - ......
COBOL

I
RPG II

I
Description

XAB type is invalid for the organization or
operation.

-368

File processor error: File processor could not create
file.

-384

113

No Current Record: Delete, truncate, or update operation was not immediately preceded by a successful get or find.

252

-400

File processor error: File processor deaccess failure
during RMS-11 file close operation.

255 1

-416

Invalid area number specified in Key XAB DAN
field.

177120s/-432

132

-432

Record accessed by RF A access mode has been
deleted.

1771 OOs/-448

133

-448

• Syntax error in device name
• No such device
• Inappropriate device for operation

-352

.......
.......
$»

DIBOL

::s
p,..
("to

t:l"'
tD

3°

(Jq
l""S

Example You attempted to create an Indexed file
on magnetic tape.
ER$DFW

177070s/-456

t'-1

ER$DIR

177060J-464

::s
(Jq

ER$DME

177040s/-480

255

ER$DNF

177020J-496

ER$DNR

177000J-512

$»

e
e5·

File processor
error code

255 1

-456

File processor error: File processor could not write
bucket; RMS-11 deferred the I/0 operation until it
needed the I/0 buffer for another bucket because
the user program specified Deferred Write.

-464

Syntax error in filespec directory name.

-480

Dynamic memory exhausted: An RMS-11 buffer
pool had insufficient free space.

-496

Directory not found.

-512

374

Device not ready.

(Jq

$»

c
$»

(Jq

tD
Cll

I
'I

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

(continued on next page)

~
I

Table B-2: Language Error Code Mapping (Cont.)

CX>

Symbolic
Value

MACRO-II
Status Code

MACR0-11
Status Value

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

ER$DPE

176770s/-520

File processor
error code

255 1

-520

Device positioning error.

::l
p.

ER$DTP

176760s/-528

XAB address

255 1

-528

Invalid key data type.

M'"

ER$DUP

176740s/-544

134

2/22 2

-544

102

Invalid record operation: You attempted to insert a
record that would cause duplicate values in a key
field where duplicates are not allowed.

ER$ENT

176720s/-560

162

-560

ER$ENV

176700s/-576

135

-576

425

File processor error: File processor could not enter
filespec in directory.
You attempted an operation when the RMS-11
routines were not in the task: In MACR0-11, the
operation or file organization was not specified in
ORG$ macro; in BASIC-PLUS-2, you didn't specify correct switches with BUILD command.

ER$EOF

176660s/-592

-592

• For record processing: End of file.
• For Block 1/0: Invalid VBN.

ER$ESS

176640s/-608

255 1

-608

Expanded file-name string area in NAM block too
short.

ER$EXP

176630s/-616

255 1

-616

File expiration date not reached.

ER$EXT

176620s/-624

255 1

-624

File processor error: During RMS-11 file extension
operation.

ER$FAB

176600s/-640

255 1

-640

F AB BID field does not contain FB$BID.

ER$FAC

176560s/-656

136

-656

Invalid record operation: Operation does not match
access declaration made when file was created or
opened.

ER$FEX

176540s/-672

-672

You tried to create a file that exists.

ER$FID

177530s/-168

255 1

-168

Invalid file ID.

~00
I

~
~

::r"
a>

Description

!JQ

1-j

8
8

File processor
error code

!JQ

t'-4
~

::l

!JQ

a>
00

File processor
error code

(continued on next page)

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.
"2" means operation succeeded.
"22" means operation failed.

,:)

)

(

c·

Table B-2: Language Error Code Mapping (Cont.)
I Symbolic ! MACR0-11

Status Code

MACR0-11
Status Value

J BASIC-

Value

PLUS-2

PDP-11
COBOL

ER$FLG

176520s/-688

XAB address

137

RPG

DIBOL

-688

I
Description

_ Invalid combination of key .characteristics.

I Example No duplicate key values, but key values
can change during update operations.

138

-704

File locked by another user: You cannot access the
file because your sharing specifications cannot be
met.

162

-720

File processor error: File processor could not find
filespec in specified directory.

176440s/-736

-736

File not found during file open operation.

ER$FNM

176420s/-752

-752

Syntax error in file-name.

ER$FOP

176400s/-768

139

-768

Invalid file access option specified in FAB FOP
field.

ER$FSS

176370s/-776

255 1

-776

RSTS/E monitor error: the File-name String Scan
routine is unable to parse the file-name string supplied by RMS-11.

ER$FUL

176360s/-784

24/34

-784

_ Device full: RMS-11 cannot create or extend file.

ER$FLK

176500s/-704

ER$FND

176460s/-720

ER$FNF

File processor
error code

r:nI
t--1.
t--1.
~

::s

FSS directive
error code

0..
("T-

::r
('O

~
~
0

(Jq
~

/95 3

ER$IAN

176340s/-800

ER$IDX

ER$IFI

255 1

-800

Invalid area number specified in Key XAB IAN
field.

176320s/-816

140

-816

Specified index was not created. This code can only
occur in the STV field when STS contains
ER$RNF.

176300s/-832

255 1

-832

FAB IFI field contains invalid value.

ER$IMX

176260s/-848

-848

More than 254 keys and/or areas defined or multiple
Summary, Protection, or Date/Time XABs present
during operation.

ER$INI

176240s/-864

255 1

-864

$INIT or $INITIF macro call never issued.

XAB address

s
s
s·

(Jq

t"-4

::s

(Jq

i::
!l:>

(Jq

XAB address

255

('O

en
~
I

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.
"24" means boundary violation during WRITE operation.
"34" means file couldn't be extended during a seauential WRITE ooeration.

(continued on next page)

o:2
I

"""'"
=

Table B-2: Language Error Code Mapping (Cont.)
Symbolic
Value

MACR0-11
Status Code

ER$IOP

176220s/-880

MACR0-11
Status Value

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

141

-880

Description

r:nI

Illegal operation.

.......

Example You attempted to truncate a nonSequential file.

.......

$:\:I

::s

0..

Example You attempted to delete or extend a magnetic tape file.

M-

::r"
Ct)

""d
...,

Example You issued a Block 1/0 operation to a
stream not connected for block operations.

(Jq

s
3

Example You issued a record operation to a stream
connected for Block 1/0 operations.

s·

(Jq

t""

$:\:I

ER$IRC

176200s/-896

142

-896

Illegal record encountered in Sequential file: Record-length field is invalid.

ER$ISI

176160s/-912

255 1

-912

RAB ISi field is invalid. You may have altered it or
failed to connect the stream.

ER$KBF

176140s/-928

143

-928

No key specified during random record operation:
RAB KBF field equals 0.

ER$KEY

176120s/-944

143

23/244

-944

Negative Relative Record Number during random
operation or bad format in packed decimal key
value.

ER$KRF

176100J-960

144

-960

Invalid key of reference:

::s
(Jq
s::
~
r:JJ

• During random get or find operation.
• During connect or rewind operation: Error code is
returned for the first sequential get or find operation following the connect or rewind.
ER$KSZ

176060J-976

ER$LAN

176040J-992

1
4

XAB address

145

-976

Invalid key size

255 1

-992

Invalid area number specified in Key XAB LAN
field.
(continued on next page)

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.
"23" means record was not found during random READ of a Relative file.
"24" means boundary violation during WRITE operation.

)

)
I

)

(

Table B-2: Language Error Code Mapping (Cont.)
I_ _ .. ! __ . ___ __
! __ . ___ -_-_----! _. ____
I---=~=-~--Symbolic
Value

MAlJHU-11
Status Code

ER$LBL

176020s/-1008

MAlJHU-11
Status Value

HAS ICPLUS-2

COBOL

RPG II

DIBOL

Description

146

-1008

I Invalid medium: Magnetic tape is not labeled in

PUP-11

accordance with ANSI standards.

=:a

ER$LBY

176000s/-1024

-1024

Logical channel busy: You attempted to create or
open a file using a logical channel in use; that is,
you already opened a file on that channel.

ER$LCH

1757608/-1040

-1040

Invalid logical channel or unit number.

ER$LEX

175750s/-1048

XAB address

255 1

-1048

You attempted to extend an area containing an
unused extent.

ER$LOC

175740s/-1056

XAB address

255 1

-1056

Invalid location during Placement Control.

ER$MAP

175720s/-1072

255 1

-1072

Memory-resident data structures, such as I/O buff:
ers, corrupted. This code can occur in the STV field
when Status Code contains ER$ABO.

ER$MKD

175700s/-1088

252

-1088

File processor error: File processor could not mark
file for deletion.

ER$MRN

175660s/-1104

147

-1104

• Maximum Record Number field contains a negative value during creation of Relative file.
• Relative Record Number for random operation to
Relative file exceeds Maximum Record Number
specified when file was created.

ER$MRS

175640J-1120

148

-1120

Maximum Record Size is zero during file creation
and one of the following is true:

00
I

~
~

i;:g

::;$

0..
~

('D

File processor
error code

~
0

..,

(Jq

i;:g

s
s5·

(Jq

t"-4

• Record format is fixed
• File organization is Relative.

i;:g

(Jq

i;:g
(Jq

ER$NAM

175620J-1136

255 1

-1136

Odd address in FAB NAM field on file open, creation, or erase operation.

ER$NEF

175600s/-1152

149

-1152

You attempted a put operation to a Sequential file
when stream is not positioned to end-of-file.

('D

tp
"""""
"""""

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

(continued on next page)

.....

Table B-2: Language Error Code Mapping (Cont.)

Symbolic
Value

MACRO-II
Status Code

ER$NID

I
.........
.........

::s

MACR0-11
Status Value

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

Description

175560s/-1168

255 1

-1168

Dynamic memory exhausted: Not enough buffer
area to open an Indexed file .

ER$NPK

175540s/-1184

150

-1184

You attempted to create an Indexed file without
defining a Primary Key.

ER$0PN

175520s/-1200

FIB
error code

-1200

File processor error: During RMS-11 file open
operation.

ER$0RD

175500s/-1216

XAB address

255 1

-1216

XABs not ordered properly.

ER$0RG

175460s/-1232

255 1

-1232

Invalid file organization.

ER$PLG

175440s/-1248

-1248

A Prologue block may be corrupted. Do as many of
the following steps as are necessary:

0...
M'-

::J"'

(t)

1-1

$l:)

3
3

s·

t"4

$l:)

::s

1. Move the disk pack containing the file to another
device and try the process again. If it works, the
error was caused by a hardware read failure.
2. Recreate the file using the RMSIFL or RMSCNV
utility.
3. Restore the file from your last backup copy.

$l:)
~
(t)

)

ER$POS

l 75420J-1264

XAB address

151

-1264

ER$PRM

175400s/-1280

XAB address

255 1

-1280

ER$PRV

175360s/-1296

-1296

114

Privilege violation: access to the file denied by the
operating system.

ER$RAB

175340s/-1312

255 1

-1312

RAB BID field does not contain RB$BID.

ER$RAC

175320s/-1328

152

-1328

Invalid or illogical record access option specified in
RAB RAC field.

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

You specified a key position beyond the end of the
record.
File directory entry contains date and time information not semantically correct. The file may be corrupted. Recreate field using RMSIFL or
RMSCNV utility.

(continued on next page)

)

(

Table B-2: Language Error Code Mapping (Cont.)
BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

1753008/-1344

255 1

-1344

You specified both Carriage Return control and FORTRAN forms control.

ER$RBF

175260s/-1360

255 1

-1360

RAB RBF field contains an odd address (Block VO
access only).

ER$RER

175240s/-1376

252

-1376

File processor error:

O~UJ.UU.U'-i

C!-·-L.~1~~

MACRO~ll

MACRO-!!

Value

Status Code

Status Value

ER$RAT

I
File processor
error code

Description

• In record processing: Read failure on file block.
• In Block I/0, VBN = 0, an illegal value.

ER$REX

175220s/-1392

153

-1392

113

Record exists: During a put operation in random
mode to a Relative file, you tried to insert a record
into a cell containing a record.

ER$RFA

175200s/-1408

173

-1408

Invalid RF A during RF A access.

ER$RFM

175160s/-1424

167

-1424

Invalid record format.

ER$RLK

175140s/-1440

154

-1440

Target bucket locked by another task or another
stream in the same program.

ER$RMV

175120s/-1456

File processor
error code

252

-1456

File processor error: File processor could not remove
filespec from directory.

ER$RNF

175100s/-1472

ER$IDX

155

-1472

144

ER$RNL

17 5060s/-1488

255 1

-1488

NO
ERROR

Record specified during random get or find operation
does not exist in Relative or Indexed file. For Indexed
files only, STV may contain ER$IDX.
You initiated a free operation, but no bucket was
locked.

ER$ROP

175040s/-1504

255 1

-1504

Invalid record processing option or illogical combination of values specified in RAB ROP field.

ER$RPL

175020s/-1520

255 1

-1520

File processor error: Read failure on file Prologue.

ER$RRV

175000s/-1536

-1536

Invalid RRV record encountered in Indexed file. File
may be corrupted. Recreate field using RMSIFL or
RMSCNV utility.

00
i--"
......
~

0..
("'t--

::r"
('D

'"1

~
'"1
~

8
8

s·

crq

t""'
~

crq
~

crq
('D

trj

File processor
error code

I
~
"""""'

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

(continued on next page)

Table B-2: Language Error Code Mapping (Cont.)

I
,.....
~

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

Description

174760s/-1552

255 1

-1552

Record stream active: In asynchronous environment,
you attempted a record operation on a stream that is
performing an operation.

ER$RSZ

1747408/-1568

156

-1568

• Record size exceeds one of the following:
- MRS for variable-length or VFC records.
- magnetic tape block size.
- data bucket size of Indexed file.
- 51010 bytes and block spanning not allowed.
- end of I/0 buffer during Locate Mode put operation on sequential file.
• Record is too short to contain Primary Key of In~
dexed file.
• Record size is zero during Block I/0.
• Record size is not equal to size of Current Record
for update operation on a disk Sequential file or on
an Indexed file where Primary Key duplicates are
allowed.
• Record size does not equal MRS for fixed-length
records.

ER$RTB

174720s/-1584

157

-1584

Record too big for user buffer: RMS-11 could not
move entire record retrieved by get operation to user
buffer. This error does not destroy the context of the
stream. Rather, the stream's context is updated as if
the operation were completely successful and as
much of the record as possible is moved to the u:ser
buffer.·

ER$RVU

17 4710s/-1592

255 1

-1592

During a put or update operation, RMS-11 moved
the specified record to the file successfully, but could
not update one or more Record Reference Vector
(RRV). The file is corrupted, but you can retrieve
every· record via the Primary index. Therefore, you
should create a new Indexed file and populate it from
the bad file using either the RMSIFL or the
RM SC NV utility.

Symbolic
Value

MACR0-11
Status Code

ER$RSA

MACR0-11
Status Value

CJ).

"""""'
"""""'

0..
~

::r
('D
~

(Jq
~

(Jq

t'-4

(Jq

s::

~
(Jq
('D

r:J:J

Actual record
size

(continued on next page)

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

)

)
4

)

(

Table B-2: Language Error Code Mapping (Cont.)
..... ,...Rr.,. • •
I
....... A,...Rr.. • • I .......
I
I
I
I
I
I
I
I
rur-11 I
I
I
Value
Status Code
Status Value
PLUS-2
COBOL
RPG II DIBOL

• •. I

~ymoonc

lYll\.~

COT,...

v-11

'--'

v-11

Tto~..-..

I_ 174700s/-1600
-

158

-1600

Invalid record operation: During a sequential put
operation to an Indexed file, Primary Key of record
to be written is not equal to or greater than key of
previous record.

ER$SHR

17 4660s/-1616

168

-1616

You specified allow write declaration for Sequential
file.

ER$SIZ

174640s/-1632

159

-1632

Invalid key size.

XAB address

17 46208/-1648

255

Example Key is .bigger than Maximum Record
Size.

30
-1648

During asynchronous record operation, RMS-:-11
found the stack is too big to save. This code can only
occur in the STV field when Status Code contains
ER$ABO.

-1664

The interface between RMS-11 and the operating
system has changed. Report this error on an SPR to
DIGITAL.

00
I

ER$SYS

17 4600s/-1664

!:I:>
~

Directive or
QIO status
error code

255 1

0..
~

::r"
ro

Description

ER$SEQ

ER$STK

;-;.
;-;.

• •

na"'I'--'-

ER$TRE

174560s/-1680

-1680

Index in Indexed file is corrupted. Recreate file using RMSIFL or RMSCNV utility.

ER$TYP

174540s/-1696

-1696

Syntax error in extension, such as more than three
characters specified.

ER$UBF

174520s/-1712

255 1

-1712

User buffer improperly specified: address is either
zero or buffer is not word-aligned (for Block I/0 access only).

ER$USZ

174500s/-1728

255 1

-1728

User buffer length equals zero.

ER$VER

17 4460s/-17 44

-1744

Syntax error in file version number.

ER$VOL

174440s/-1760

255 1

-1760

Nonzero relative volume number.

ER$WCD

174430s/-l 768

255 1

-1768

Explicit or default file specification field contains
wild card character.

ER$WER

1744208/-1776

252

-1776

File processor error: Write failure on file block.

...;

!:I:>

s
s5·

t""

!:I:>

r:/l

XAB address

,....I
01

File processor
error code

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

(continued on next page)

Table B-2: Language Error Code Mapping (Cont.)

"""""

BASICPLUS-2

PDP-11
COBOL

RPG II

DIBOL

-1784

File processor error: Device is write locked.

File processor
error code

252

-1792

File processor error: Error while writing Prologue.

XAB address

255 1

-1808

F AB XAB field or XAB NXT field contains an odd
address.

17 4340s/-1824

255 1

-1824

Explicit or default file specification contains extraneous field.

SU$DUP

2/2

172

NO ERROR

Conditional success: A record inserted into an Indexed file by a put or update operation contains at
least one key value present in another record.

SU$IDX

3/3

169

NO ERROR

A put or update operation on an Indexed file ended
successfully, but RMS-11 could not opti~ize the
index structure during the operation. Therefore,
RMS-11 will require extra I/0 operations to retrieve
the record. With this success code, RMS-11 may
include an RMS-11 error code to indicate why the
index structure was not updated in the STV field of
the control block.

SU$RRV

4/4

169

NO ERROR

No longer a valid completion code. See ER$RVU.

Symbolic
Value

MACRO-II
Status Code

ER$WLK

174410s/-1784

ER$WPL

174400s/-1792

ER$XAB

174360s/-1808

ER$XTR

8
8
i:r
()Cl
t""

MACRO-II
Status Value

Description

UJ
I

..........

..........
~

::s

0..

::J"'
ct)

"'d

()Cl

'"1
~

::s

()Cl

~
()Cl
ct)

r.r:i

On IAS/RSX-llM, value of ERR is same as MACR0-11 Status Code.

)

,._;

Appendix C
RMS-11 Disk-Resident Overlays

C.1

Overlay Structures
Figure C-1 is a graphic illustration of the overlay structures defined by the
standard RMS-11 ODL files, RMSllS.ODL and RMSllX.ODL. From this
figure, you can get a feeling for the extent and complexity of the RMS-11
overlay structures. You can also see what overlay segments are loaded into
memory before the segment containing the called routine is loaded.
Example Your program initiates an update operation and causes a call to an RMS-11 routine
in the overlay segment labeled with the factor name RMI03U. Before the operating
system can execute this routine, it must ensure that the following segments are in
memory:

RM SALL
RMS REC
RMOX36

The following conventions were used in creating Figure C-1:
• The parenthetical comment map name indicates a cosmetic term used on
the Memory Allocation Map produced by the Task Builder.

C-1

Example The standard RMS-11 ODL file RMSllS.ODL contains the factor:

RMSFIL: .FCTR

RMSFAB-LB:C1 ,1JRMSLIB/LB:ROIFLF:ROFSET:ROFSEI-RMSFL

RMSllS.ODL also contains the ODL statement:

+NAME

RMSFAB

Finding this combination of statements, the Task Builder describes the overlay
segment defined in the RMSFIL factor with the name RMSFAB. FAB is the
acronym for File Access Block, the internal control structure RMS-11 and the
user program use to describe a file. Since the RMSFIL factor starts the series of
overlay segments that contain file operation routines, the name RMSFAB is
appropriate.
If RMSFAB were left out of the factor, the Task Builder would use the name of
the first module or factor in the line; in this case, the utility would use ROIFLF, a

less meaningful name.

• Factor names are in bold print.
• Each indentation of one space represents a concatenated factor or module
within a factor. Each separate series of indented names represents an overlay segment.
• Factors or modules equally indented either overlay each other or the top one
is only a factor or map name.
Example The following portion of the RMSllS.ODL chart shows:

• The factor RMOXOX concatenated with the module RMSSYM.
• The fact that RMOXOX is only the name for a factor that starts with another
factor name (RMSIO). The two factor names are equally indented.
Factor names such as RMOXOX are used because the Task Builder limits the
length of input lines. Such names essentially extend factors.
• The factor RMSCBL is concatenated with the factor RMSIO. The intervening
module names comprise RMSIO.

RMSSYM

RMOXOX
RMSIO
ROCA CH
RORLCH
ROMAPC
RORWBF
ROUNLK
RMSCBL

C-2

RMS-11 Disk-Resident Overlays

JFigure C-1: RMSllS.ODL and RMSllX.ODL Overlay Structures

RMS11S.ODL

RMS11X.ODL

RMSROT (ref. by primary ODL file)
RMSSYM
RMOXOX
RMSIO
ROCACH
RORLCH
ROMAPC
RORWBF
ROUNLK
RMSCBL
RORTCB
RO AC BB
RORTDB
ROM DAT
RMSCOM
RORMSE
ROSAVR
ROIDPB
RMSCM1
ROAUTO
ROIMPA
RMSEXC
RO EX SY
RO RS ES
ROWTBS
RORMSA

RMSROT (ref. by primary ODL file)
RMSSYM
RMOX3X
RMSIO
ROCA CH
RORLCH
ROMAPC
RORWBF
ROUNLK
RMSCBL
RORTCB
ROACBB
RORTDB
ROM DAT
RMS COM
RORMSE
ROSAVR
ROIDPB
RMSCM1
ROAUTO
ROIMPA
RMSEXC
ROEXSY
RORSES
ROWTBS
RORMSA
RMSIXO
RORLSB
ROGETB

RMS-11 Disk-Resident Overlays

C-3

Figure C-1: RMSl lS.ODL and RMS 1lX. ODL Overlay Structures (Cont).
RMS11S.ODL

~00
I

.......
.......

0~·

i::s:;"
I

:;:o
CD
00

5.:
CD
::s

!!'+-

0
<
CD
""i

~
00

RMSALL (ref. by primary ODL file)
RMS11 (map name)
RMSFIL
RMSFAB (map name)
ROIFLF
ROFSET
ROFSB
RMSFL
AOC LCM
RM FILE
ROA EOP/RORXAC
RM SfLO
RMSFNM
AOPRFN
ROXPFN
ROMKWA
ROASLN
ROINIT
RMSFL1
ROALDB
ROALBD
ROALIO
RMSDPC
AODPYC
ROCKSM
RMSFL1
ROALDB
ROALBD
FIOALIO
RMSCLS
R1CLOS
R2CLOS
RMSCLD
ROCKSM
RllEXTD
ROEXTD
RMR1RL
R1RLBK
R1NXBK
RMSOPN
ROOP FL
RMSOPO
FIORXDI
ROMFNB
RORD50
RMOXYO
RMOPIN (map name)
ROCKSM
R10PFL
R20PFL
MSCRO
RMSCRE (map name)
FIOCRXI
RMSCR1
R1CRFL
R2CRFL
1CRCK
MCRC.K
R2CRCK
MCREA
ROCRFL
AO MF NB
RORD50
RMSCR2

)

0
RM SOOP
R1WTLS
R1CKEF

RORECR
RORXCR
RORXMD
RORXDI
MOX21
R2WPLG
R2BFMT
ROCKSM
RMSFL1
ROALDB
ROALBD
ROALIO
RMSMIC
RMSERA
ROERFL
RMSER1
RORXMD
RMSOPO
RORXDI
ROMFNB
RORD50
RMSEXT

RMIN10

R1GET
R1GSET
R1GBLD
RMOU1P
R1PUT
R1PSET
R1PUNR
R1PBLD
MOU1U
R1UPDA
H1UBLD
R1DELE
R1PSET

RMSBLK
RMSBMC (rrep name)
RORWBI
RMMISC
RMSMIS (map name)
RMMISO
ROFREE
R1TRUN
RMMIS1
RORWIN
R2CALC
RMWATR
ROWATR
RMR1WL
R1WTLS
RMR1RL
R1RLBK
R1NXBK
RMEXTD
ROEXTD
RMMAGT
ROMAGT
RMOX26
RMSREL (map name)
RMI02C
R210CK
R2CALC
ROCK SM
RMI02A
R2FIND
R2GSET
RMI02G
R2GUPD (map name)
R2GET
RMI02U
R2UPDA
R2PSET
RMI020
[
R2DELE
RM102P
R2PUT
R2PSET
RMI02H
R2BFMT
R2EXTD
RM EX TD
ROEXTD

ROEXTO
ROCKSM
R2BFMT
RMEXID
AOEXTD
RMSFL1
ROAL DB
ROALBD
ROALIO
RMSFEC
RMSRAB (map name)
RMSCDX
RMSCD (map name)
R1CONP
R2CONP
RMSCDO
R1DISC
R2DISC
RMSCD1
ROCCLN
RORSB
RO ALBS
RMSFL1
ROAL DB
ROALBD
FIOAUO
RMSCD2
RMR1RL
R1RLBK
R1NXBK
RMR1WL
R1WTLS
RMEXID
ROEXTD
MSSQO

R1NXBK
RMEXTD
ROEXTD

(continued on next page)

")
._11

)

_ _;•

)

(

Figure C-1: RMSllS.ODL and RMSUX.ODL Overlay Structure (Cont.)
RMS11X.ODL
RMSALL (ref. by primary ODL file)
RMS11 (map name)
RMSFIL
RMSFAB (map name)
ROIFLF
ROFSET
ROFSB
RMSR.
RICLCM
RMFILE
ROREOP/RORXAC

RMS FLO
RMSFNM
ROPRFN
ROXPFN
ROMKWA
ROASLN
ROINIT
RMSFL1
ROA LOB
ROALBD
ROALIO
RMSO PC
RODPYC
ROCKSM
RMSFL1
ROALDB
ROALBD
ROAUO
RMsa.s
R1CLOS
R2CLOS
RMSCLO
R3CLOS
ROCKSM
RMEXTD
ROEXTD
RMR1RL
R1RLBK
R1NXBK
RMSOPN
ROOPFL
RMSOPO
RORXDI

00
I

.......

t::t
;·
~

ROMFNB
RORD50
RMOXYO
RMOPIN (map name)
ROCKSM
R10PFL
R20PFL
RMOX30
R30PFL
R3RPLG
RM SCRO
RMSCRE (map name)

!;:l:j
~

s:
~

::s

("t-

0
<
~
i-;

ROCRXI

~
00

RMSCR1
R1CRFL
R2CRFL
R3CRFL
MCRCK
1CRCK
R2CRCK
R3CRCK

l
c

rL r
~I MCREA

ROCRFL
ROMFNB
RORDSO

:~~

RORXCR
RORXMD
RORXDI

CRXX
RMOX21
R2WPLG
R2BFMT
ROCKSM
RMSFL1
ROALDB
ROALBD
ROALIO
RMOX31
R3WPLG
ROCKSM
ROALBD
RM EXTD
ROEXTD

MSMIC
MSERA
ROERFL
RMSER1
RORXMD

RMSOPO
RORXDI

ROMFNB
RORDSO
MSEXT
ROEXTO
ROCKSM
R2BFMT
RMEXTO
ROEXTD
RMSFL1
ROAL DB
ROALBD
ROALIO

MSREC
RMSRAB (map name)
RM SC DX
RMSCD (map name)
R1CONP
R2CONP
RMSCDO
R3CONP
R1DISC
R2DISC
R3DISC
RMSC01
ROCCLN
RORSEI
RO ALBS
RMSFL1
ROAL DB
ROAL BO
ROALIO
RMSCD2

RMR1RL
R1RLBK
R1NXBK
RMR1WL
R1WTLS
RMEXTD
ROEXTD

RMSSQO
RMSSEO (map name)
RMR1RL
R1RLBK
R1NXBK
RMEXTD
ROEXTO
RMSQOP
R1WTLS
R1CKEF
RMIN10
R1GET
R1GSET
R1GBLD
RMOU1P
R1PUT
R1PSET
R1PUNR
R1PBLD
RMOU1U
R1UPDA
R1UBLD
R1DELE
R1PSET
RMSBLK
RMSBMC (map name)
RORWBJ
RMMISC
RMSMIS (map name)
RMMISO
ROFREE
R1TRUN
RMMIS1
RORW1N
R2CALC
RMWATR
ROW ATR
RllR1WL
R1WTLS
RMR1RL
R1RLBK
R1NXBK
RM EXTD
ROEXTO
RMMAGT
ROMAGT
RMOX26
RMSREL (map name)
RMI02C
R210CK
R2CALC
ROCKSM
RMI02A
R2FINO
R2GSET
RMI02G
R2GUPD (map name)
R2GET

iii~

RM102U
R2UPDA
RMI020
R2DELE
RMI02P
R2PUT
R2PSET
RM102H
R2BFMT
R2EXTD

RMEXTD
ROEXTD
OU35
R31SID

RMEXTO
ROEXTD
LMOX36
RMSIDX (map name)
RIM030
RMOU3C
R3FROO
R3GKEY
R3FPAT
R3MISC
R3WBKT
RMOU30
R3FRFA
R3FNDR
R3NBKT
R3KREF
RllOU3E
R3SDBK
R3SKRE
ROCMKY
ROCKSM
ROGPTR
RMl031
RMI036
R3GET
R3RND
RMIN3S
R3FRSE
R3POSE
R3POSR
R3FRRV
RMIN3K
R3GSET
R3GTRE
R3GRPT
R3FRRF
R3FRKE
RMI03P
R3PUT
R3PSET
RMOU3Q

-l '

RMEXTD
R3BFMT
RO EX TD
RMOU33
R3PIXC
R3FRRV
R3DLSI
MI03U
R3UPDA
R3USET
RMOU3Q
RMOU34
R31UDR

OU3F

OU3H

UDI
3BSRT
R31UDC
OU3G
3BSPL
3BRRV
R3URRV
MOU3A
R3ALOC
R3BFMT
RMEXTD
ROEXTD
MOU35
R31SID

ffMOU34
R31UDR

OU3H

SDI
3BSRT
3SSPL
OU3A
R3ALOC
R3BFMT
RMEXTD
ROEXTD
RMOU36
R31UDX
R31KEY
RMOU3M
R31KYI
R3KSPL
[ R3ROOT
RMOU3A
R3ALOC
R3BFMT
RMEXTD
ROEXTD
·RMOU3P
R3MKID
RMOU3A
R3ALOC

OU3F

UDI
3BSRT
R31UDC
OU3G
BSPL
3BRRV
R3URRV
RMOU3A
R3ALOC
R3BFMT

SDI
3BSRT
R3SSPL
OU3A
R3ALOC
R3BFMT

I fi

RMEXTD
ROEXTD
MOU36
R31UDX
R31KEY

IRMOU3M
R31KYI
R3KSPL
R3ROOT
RMOU3A
R3ALOC
R3BFMT
RMEXTD
ROEXTD
RMOU3P
R3MKID
RMOU3A
R3ALOC
R3BFMT
RMEXTD
ROEXTD
RMI030
R3DELE
R3DSET
R3RPLC
R3SKDL
RMOU33
R3PIXC
R3FRRV
R3DLSI

.
I

C.2 RMS-11 ODL FILES

C-6

C.2.1

RMS16X.ODL

RMS CM 1:
RMSROT:
RMSI ){0:

RMSFL:
RMSFNM:
RMSDPC:
RMS FL 1:
RMSCLS:
RMSCLO:
RMSDPN:
RMCRCK:
RMCREA:
RMSCRO:
RMSALL:
RMSFLO:
RMSMIC:
RMSERA:
RMSE>(T:
RMOXYO:
RMOH30:
RMSCR1:
RMCR>rn:
RM0){21:
RM0>(31:
RMSCDH:
RMSCD1:
RMSCDO:
RMSCD2:
RMSREC:
RMOJ<2G:
RMID2C:

• FCTR
• FCTR
• FCTR
.NAME
+NAME
+NAME
.NAME
+NAME
.NAME
+NAME
+NAME
+NAME
+NAME
+NAME
+NAME
+NAME
+NAME
+NAME
• FCTR
.FCTR
.FCTR
+FCTR
• FCTR
• FCTR
, FCTR
• FCTR
.FCTR
.FCTR
+FCTR
• FCTR
+FCTR
.FCTR
.FCTR
.FCTR
.FCTR
, FCTR
• FCTR
• FCTR
.FCTR
• FCTR
.FCTR
• FCTR
• FCTR
• FCTR
• FCTR
.FCTR

LB: [ 1 ,1 JRMSLIB/LB:ROAUTO:ROIMPA
LB: [ 1 11 J RMSL IB/LB: RMSSYM-RMl<XXX
LB: [ 1 11 J RMSL IB/LB: RORLSB: ROGETB
RMS11
RM SF AB
RM SR AB
RM SC RE
RMSCD
RMOPIN
RMSSEQ
RM SR EL
R2GU PD
RMS ID){
RMSBMC
RMSMIS
RMDEXT
RMDRLB
RMDWTL
LB:[111JRMSLIB/LB:ROCLCM-RMFILE-RMSFLO
LB:[l 1lJRMSLIB/LB:ROINIT:ROPRFN:ROXPFN:ROMKWA:ROASLN-RMSFLl
LB:[l i1JRMSLIB/LB:RODPYC
LB: [ 1 11 J RMSL I BIL B: ROAL DB: ROALBD: ROAL ID
LB: [ 1 , 1 J RMSL I B/LB: R1CLOS: R2CLOS-RMSCLO
LB:[l 11JRMSLIB/LB:R3CLOS:ROCKSM-RMEXTD-RMR1RL
LB: [ 1 11 J RMSL I B/LB: ROOPFL-RMSOPO·-RMOHYO
LB: [ 1 11 JRMSLI B/ LB: R1 CRCK: R2CRC K: R3CRCK
LB: [ 1 11 JRMSLI B/LB: ROCRFL: ROMFNB: RORD50-RMSCR2
RMSCRE-LB:Cl 11JRMSLIB/LB:ROCRHI-RMSCR1
RMS11-*<RMSFIL1RMSREC>
RMSFNM-<RMSDPN-RMSDPC-RMSCDX1RMSCR01RMSCLS-RMSMIC-RMSCDO>
RMSERA-RMSEl<T
LB: [ 1 1l] RMSL I B/LB: ROERFL-RMSER 1
LB: [ 1 , 1 J RMSL I B/LB: ROE XTO: R2BFMT
RMOPIN-LB:[l 11JRMSLIB/LB:ROCKSM:R10PFL:R20PFL-RMOX30
LB: [ 1 1 l JRMSLI B/LB: R30PFL: R3RPLG
LB:[l 11JRMSLIB/LB:R1CRFL:R2CRFL:R3CRFL-RMCRCK-RMCREA-RMCRXX
RM0)<21-RM0){31
LB:[l 11JRMSLIB/LB:R2WPLG:R2BFMT:ROCKSM
LB: [ 1, 1 JRMSLIB/LB: R3WPLG-RMEHTD
LB:[l 11JRMSLIB/LB:R1CONP:R2CONP:R3CONP:RORSEI-RMSCD1
LB: [ 1 d JRMSLIB/LB: ROCCLN :ROALBS
LB:[l 11JRMSLIB/LB:R1DISC:R2DISC:R3DISC-RMSCD2
LB:[l 11JRMSLIB/LB:ROCCLN-RMR1WL
RMSRAB-RMEXTD-<RMOH2G-RMSBLK-RMSSCl01RMOH3GI
RMSREL-RMID2C-RMID2A
LB : [ 1 , 1 JRMSLI BI LB: R2 IOCK : R2CALC: ROCK SM

RMID2A:
RM ID2G:
RM I02P:
RMI02H:
RM ID2U:
RMI02D:
RMOl-(36:
RMI030:
RMI031:
RMI03G:
RMIN3K:
RMIN3S:

.FCTR
.F.CTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
• FCTR
, FCTR
+FCTR
+FCTR
• FCTR

LB: (1 i1JRMSLIB/LB:R2FIND:R2GSET-RMID2G-RMID2P
LB: [1 dJRMSLIB/LB:R2GET-RMI02U-RMID2D
LB:[l 11JRMSLIB/LB:R2PUT:R2PSET-RMI02H
LB: [ 1 11 J RMSL I B/LB: R2BFMT: R2El<TD
LB: [1 i1JRMSLIB/LB:R2UPDA
LB:[l 11JRMSLIB/LB:R2DELE
RMSIDH-RMI030-RMI031
RMOU3C-RMOU3D-RMOU3E-RMI03P-RMI03U-RMDU3A
<RMID3G-RMID3D-RMOU33 1RMDU3Q)
LB: [ 1 11 JRMSLI BI LB: R3G ET: R3F I ND- RM I N3 S-RM IN 3K
LB:[l 11JRMSLIB/LB:R3GSET:R3GTRE:R3GRPT:R3FRRF:R3FRKE
LB:[l 11JRMSLIB/LB:R3FRSE:R3POSE:R3POSR:R3FRRV

RMS-11 Disk-Resident Overlays

~
_j

f~MI03P:

"-"'.

_,,

RMOU3Q:
f~M I03U:
RMI03D:
RMOU33:
RMOU34:
RMOU35:
RMOU3G:
RMOU3A:
RMOU3C:
RMOU3D:
RMOU3E:
RMOU3F:
RMOU3G:
RMOU3H:
RMOU3J:
RMOU3K:
RMOU3L:
RMOU3M:
RMOU3N:
RMOU30:
RMOU3P:
RMSBLK:
RMMISC:
RMMISO:
RMMIS1:
RMWATR:
RMMAGT:

, FCTR
, FCTR
,FCTR
,FCTR
, FCTR
, FCTR
,FCTR
,FCTR
,FCTR
,FCTR
, FCTR
,FCTR
, FCTR
, FCTR
, FCTR
,FCTR
, FCTR
, FCTR
,FCTR
, FCTR
,FCTR
,FCTR
, FCTR
, FCTR
,FCTR
,FCTR
,FCTR
,FCTR

LB: [1 d JRMSLI B/L.B: R3PUT: R3PSET
<RMOU341RMOU35-RMOU3G-RMOU3P)
LB: [ 1 d JRMSLI B/L.B: R3UPDA: R3USET
LB: [ 1 , 1 JRMSLI B/L.B: R3DELE: R3DSET: R3RPLC: R3SKDL
LB: Cl t1JRMSLIB/LB:R3PIHC:R3DLSI
LB:C111JRMSLIB/LB:R3IUDR-RMOU3J-RMOU3F-RMOU3K-RMOU3G
LB: [1 dJRMSLIB/L.B:R3ISID-RMOU3H
LB:[l 11JRMSLIB/L.B:R3UIDX:R3IKEY-RMOU3M-RMOU30
LB: [111 JRMSLIB/L.B:R3ALOC:R3BFMT
LB:[! 11JRMSLIB/LB:R3FROO:R3GKEY:R3FPAT:R3MISC:R3WBKT
LB: [l11JRMSLIB/L.B:R3FRFA:R3FNDR:R3NBKT:R3KREF
LB: [111JRMSLIB/LB:R3SDBK:R3SKRE:ROCMKY:ROCKSM:ROGPTR
LB: [1 dJRMSLIB/LB:R3IUDI:R3BSRT
LB: [1 ,1JRMSLIB/LB:R3BRRV:R3URRlJ
LB: [ 1 , 1 J RMS LIB I LB: R3 I SD I : R3BS RT - RMO U3L
LB: [1 t1JRMSLIB/LB:R3IUDC
LB: U d JR MS LIB/LB: R3BSPL
LB: U t1JRMSLIB/Ul:R3SSPL
LB: Cl t1JRMSLIB/LB:R3IKYI
LB: [ 1, 1 JRMSLI B/LB: R3KSPL
LB: [ 1 d JR MS LIB/LB: R3ROOT-RMOU3N
LB: [ 1, 1 JRMSL IB/LB: R3MK ID
RMSBMC-LB: [ 1 ti J RMSLI B/LB: RORWB I-RMM ISC
RMSMIS-RMMISO
LB:Cl11JRMSLIB/LB:ROFREE:R1TRUN-RMMIS1
LB: [ 1, 1 JRMSLIB/LB: RORWIN-RMWATR-RMMAGT
LB: [ l 11JRMSLIB/LB:ROWATR
LB:Cl 11JRMSLIB/LB:ROMAGT

C.2.2 RMS20X.ODL

RMSCM 1 :
RMSROT:
RMS I NO:

RMSFL:
RMSFNM:
RMSDPC:
RMSFL 1:
RMSCLS:
RMSCLO:
RMSOPN:
RMCRCK:
RMCREA:
RMSCRO:
RMSALL:
RMSFLO:

,FCTR
,FCTR
,FCTR
,NAME
.NAME
,NAME
.NAME
,NAME
,NAME
.NAME
,NAME
,NAME
,NAME
,NAME
,NAME
,NAME
.NAME
, FCTR
, FCTR
, FCTR
, FCTR
,FCTR
,FCTR
,FCTR
,FCTR
,FCTR
,FCTR
, FCTR
,FCTR

LB: [ 1 ti JRMSLIB/LB:ROAUTO:ROIMPA
LB: [ 1 ti JRMSLI B/LB: RMSSYM-RMNNXl<
LB: [ 1 , 1 J RMSL I B/LB: RORLSB: ROGETB
RMSll
RMSFAB
RMSRAB
RMSCRE
RMSCD
RMOPIN
RMS SEQ
RMSREL
RMSIDN
RMSBMC
RMSMIS
RMDD:T
RMDRLB
RMDWTL
LB:Cl 1lJRMSLIB/LB:ROCLCM-RMFILE-RMSFLO
LB:Cl 1lJRMSLIB/LB:ROINIT:ROPRFN:ROXPFN:ROMKWA:ROASLN-RMSFLl
LB:Cl 11JRMSLIB/LB:RODPYC
LB: [ 1 , 1 JRMSLI BI LB: ROAL DB: ROAL BO: ROA LI 0
LB: [ 1, 1 JRMSLIB/LB: R1CLOS: R2CLOS-RMSCLO
LB: [ 1 , 1 J RMS LIB I LB: R3CLOS: ROCK SM-RMENTD-RMR 1 RL
LB: [ 1 , 1 JRMSLI B/LB: ROOPFL-RMSOPO-RMONYO
LB: [ 1 , 1 JRMSLI B/ LB: R 1 CRCK: R2CRCK: R3CRCK
LB: [ 1 , 1 JRMSLI B/ LB: ROCRFL-RMSCR2
RMSCRE-LB:[i 11JRMSLIB/LB:ROCRXI-RMSCR1
RMSll-*<RMSFIL1RMSREC)
RMSFNM-<RMSOPN-RMSDPC-RMSCDN-RMSCR01RMSCLS-RMSMIC-RMSCDO>

'-"'
RMS-11 Disk-Resident Overlays

C-7

RMSMIC:
RMSERA:
RMSEHT:
RMOXYO:
RMOX30:
RMSCRl:
RMCRXX:
RMOH21:
RMOH31:
RMSCDH:
RMSCDl:
RMSCDO:
RMSCDZ:
RMSREC:
RMOX28:
RMID2C:
RMID2A:
RMIOZG:
RMID2P:
RMID2H:
RMID2U:
RMID2D:
RMOH38:
RM !030:
RMI03G:
RMIN3K:
RMIN3S:
RMID3P:
RMOU3Q:
RMID3U:
RMI03D:
RMOU33:
RMOU3a:
RMOU35:
RMOU38:
RMOU3A:
RMDU3C:
RMOU3D:
RMOU3E:
RMOU3F:
RMOU3G:
RMOU3H:
RMDU3J:
RMOU3K:
RMOU3L:
RMDU3M:
RMOU3N:
RMOU30:
RMDU3P:
RMSBLK:
RMMISC:
RMMISO:
RMMIS1:
RMWATR:
RMMAGT:

·c-8

.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
, FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
, FCTR
, FCTR
, FCTR
, FCTR
, FCTR
, FCTR
.FCTR
.FCTR
.FCTR
.FCTR
,FCTR
.FCTR
.FCTR
.FCTR
.FCTR
.FCTR
,FCTR
, FCTR
, FCTR

RMSERA-RMSEXT
LB: E 1 • 1 JRMSL IB I LB: ROERFL- RMSER 1
LB: E 1 • 1 JRMSL I B/LB: ROEHTO: R2BFMT
RMDPIN-LB: El 11JRMSLI6/LB:ROCKSM:R1DPFL:R20PFL-RM0){30
LB: El .1JRMSLIB/LB:R30PFL:R3RPLG
LB: [ 1 '1 JRMSLI B/ LB: R 1 CRFL: R2CRFL: R3CRFL-RMCRCK-RMCREA-RMCRHH
RMOX21-RMOX31
LB: E111JRMSLIB/LB:R2WPLG:R2BFMT
LB: [ 1 • 1JRMSL16/LB: R3WPLG
LB: El 11JRMSLIB/LB:R1CDNP:R2CONP:R3CONP:RORSEI-RMSCD1
LB: El 11JRMSLIB/LB:ROCCLN:ROALBS-RMEXTD
LB: El 11JRMSLIB/LB:R1DISC:R2DISC:R3DISC-RMSCD2
LB: [ 1 .1 JRMSL IB/LB: ROCCLN-RMR 1WL
RMSRAB-RMEXTD-<RMSBLK-RMOX28rRMSSQO-RMOX38>
RMSREL-RMI02C-RMI02A
LB: [ 1 t1 JRMSLI B/LB: R2 I OCK: R2CALC: ROCK SM
LB: El 11JRMSLIB/LB:R2FIND:R2GSET-RMID2G-RMID2P
LB: [1 t1JRMSLIB/LB:R2GET-RMI02U-RMID2D
LB: El t1JRMSLIB/LB:R2PUT:R2PSET-RMID2H
LB: El dJRMSLIB/LB:R2BFMT:R2EXTD
LB: [ 1 • 1 JRMSL I B/LB: R2UPDA
LB: El 11JRMSL16/LB:R2DELE
RMSIDH-RMID30-RMI03G-RMI03P-RMID3U-CRMOU3Q)
RMOU3C-RMOU3D-RMOU3E-RMOU3A
LB: [1 rlJRMSLlB/LB:R3GET:R3FIND-RMIN3S-RMIN3K
LB: [ 1 • 1 JRMSLI B/ LB: R3GSET: R3GTRE: R3GRPT: R3FRRF: R3FRK E
LB: [ 1•1 JRMSLIB/LB: R3FRSE: R3PDSE: R3POSR: R3FRRl,I
LB: El r1JRMSLIB/LB:R3PUT:R3PSET
<RMOU33-RMID3D-RMOU3PrRMOU3a,RMOU35-RMOU38)
LB: El .1JRMSLIB/LB:R3UPDA:R3'lJSET
LB: [ 1'1 JRMSLI B/LB: R3DELE: R3DSET: R3RPLC: R3SKDL
LB: El t1JRMSLIB/LB:R3PIXC:R3DLSI
LB: [ 1 • 1 JRMSLI B/LB: R3 I UDR-RMDU3J-RMOU3F-RMOU3K-RMOU3G
LB: El t1 JRMSLIB/LB:R3ISIO-RMOU3H
LB: El 11JRMSLIB/LB:R3UIDH:R3IKEY-RMOU3M-RMOU30
LB: El 11JRMSLIB/LB:R3ALOC:R3BFMT
LB: El 11JRMSL16/LB:R3FROO:R3GKEY:R3FPAT:R3MISC:R3WBKT
LB: El 11JRMSL16/LB:R3FRFA:R3FNDR:R3NBKT:R3KREF
LB: E 1•1 JRMSLlB/LB: R3SDBK: R3SKRE: ROCMKY: ROCK SM: ROGPTR
LB: [ 1 t1 JRMSL16/LB:R3IUDI :R3BSRT
LB: E1 r1JRMSLIB/LB:R3BRRV:R3URRV
LB: C1 t1JRMSL16/LB:R3ISDI:R3BSRT-RMOU3L
LB: El t1 JRMSLlB/LB:R3IUDC
LB: El t1JRMSLIB/LB:R3BSPL
LB: El t1JRMSLIB/LB:R3SSPL
LB: Cl t1JRMSLIB/LB:R3IKYI
LB: E1 t1JRMSL16/LB:R3KSPL
LB: [ 1 t1 JRMSLI B/LB: R3RODT-RMOU3N
LB: E1 t1JRMSLIB/LB:R3MKID
RMSBMC-LB: Cl t!JRMSLIB/LB:RORWBI-RMMISC
RMSMIS-RMMISO
LB: El tlJRMSLIB/LB:ROFREE:RlTRUN-RMMISl
LB: El t!JRMSLIB/LB:RORWIN-RMWATR-RMMAGT
LB: E1 rlJRMSLIB/LB:ROWATR-RMRlWL-RMRlRL
LB: C1 t1JRMSL16/LB: ROMAGT

RMS-11 Disk-Resident Overlays

""~".

Appendix D
RMSDFN Command Utility

You should use the RMSDEF interactive utility (described in Section 9.3),
rather than the RMSDFN command utility, to create RMS-11 files:
• RMSDEF enables you to define all attributes of the file.
• RMSDEF interacts with you as you define the file. The utility requests
information according to your entries and provides HELP messages to guide
your choices.
• RMSDEF is actively supported.

D.1
~

Purpose
The RMSDFN command utility creates RMS-11 files. You define or default
file attributes in the command string invoking the utility. RMSDFN then
stores the attributes within the newly created file.
NOTE
The conventions discussed in Sections 9.0.1 and 9.0.2 also
apply to RMSDFN.

D.2 Effect
To create a file with RMSDFN, you need specify only the name and attributes
of the file. Optionally, you can specify an initial allocation quantity. However,
RMSDFN does not write records into the file. You can employ either an
application program or the RMSIFL or RMSCNV utility to populate the file
after RMSDFN creates it.
The attributes you can specify when creating a file vary with the file organization selected. However, all file definitions include the following information:
• file organization
• record format
• allocation quantity

D-1

• a contiguous or noncontiguous allocation
• system protection code
Indexed files also require a Primary Key. Optionally, you can define up to
nine Alternate Key definitions per file, including whether the keys can be
changed during update or may contain the same value in multiple records.

D.3 Call and Termination
D.3.1

Permanently Installed Utility

DFN [command string]
If you do not specify a command string, RMSDFN assumes control of the user

interface and prints the prompt:
DFN>

You can type a command string or (QJ]Ql) to terminate the utility. When
RMSDFN has executed a command string, it reprints the prompt.
If you include a command string, the utility attempts to execute it and then

returns control to the operating system.

D.3.2 Uninstalled Utility
RUN $RMSDF
RMSDFN assumes control of the user interface and displays the prompt:
DFN >

You can type a command string or lQ'.IBQl) to terminate the utility. When
RMSDFN has executed a command string, it reprints the prompt.

D.4 Command String
D.4.1

General Form

outfile[/switch] .. ]
where:

D-2

outfile

is the filespec of the file to be created. Wild card characters cannot
appear in any field of the specification. The file specified should not
exist in the account indicated.

switch

can be a code shown in Table D~l and described under "Switches,"
Section D.4.2.

RMSDFN Command Utility

Table D-1: RMSDFN Utility Switches
Type

,,._,,

Switch

Description

Default

Global

/lD

Identify current
RMSDFN utility.

Out.file

/AL:nnn

Allocation quantity.

AL:O

/BK:nn

Number of blocks per bucket.

BK:l

/CO

Contiguous allocation.

NOCO

/F'O: val

File organization.

FO:SEQ

/KY: n:m[: (/OP:val[: val])]

Key definition.

No keys.

/PR: options

Protection specification.

/SY:RWED
/OW:RWED
/GR:RWED
/WO:RWED

/PR:nnn

Octal protection code.

See /PR:options

/PR:<nnn>

Protection specification.

PR: <60>

/RF: val[: nnnn]

Record format and record size.

RF:VAR:O

version

No id.

D.4.2 Switches
• ID causes RMSDFN to print its current version numbers, in the form:
DFN -- VERSION 1.5

This switch can appear alone as a command string.
• AL: nnn specifies the initial allocation quantity for the file, where nnn is a
number of blocks.
• BK:nn specifies the number of blocks per bucket for Relative and Indexed
files. The default is one block per bucket; the maximum value for nn is ;~2 or

......,.

15 .

• CO indicates that RMSDFN should attempt the initial file allocation as
physically contiguous blocks.
• FO:val indicates the file organization being created. You can specify one of

the following values:
SEQ - Sequential file (default)
REL -- Relative file
IDX - Indexed file
The value set for file organization determines what additional switches can
or must appear in the command string. For example, if FO:IDX is specified,
then at least one key definition (KY) must also appear.
• KY:n:m[:(/OP:val[:ual])] defines an Indexed file key. You can specify a

maximum of ten key definitions for a single Indexed file. RMSDFN interprets the first one in the command string (scanning from left to right) as the

RMSDFN Command Utility

D-3

Primary Key definition, the next one as the First Alternate Key, the next as
the Second Alternate Key, and so on.
You must specify at least one key definition, that is, the Primary Key
definitions for Indexed files.
The minimum general form of a key definition is:
/KY:n:m
where:
n is the location of the first byte of the key field within the record; 0
represents the first byte of the record, 1 the second, and so on.
m is the key field length in bytes; the maxim um length for a key field is 255

bytes.
Example /KY:9:16

This switch defines a key field that begins in the tenth byte of each record and is
sixteen characters long.

You can also specify key characteristics with the KY switch using the following general form:

/KY :n: m: (/OP: val[:val])
where:
val

is one of the following:
DUP

if duplicate key values are allowed to occur among records
in the file. DUP is the default for Alternate Keys.

CHG

if Alternate Key fields are allowed to change values during
an update operation. CHG is the default for Alternate Keys.

NODUP

if duplicate key values cannot occur among records in the
file. NODUP is the default for Primary Keys.

NOCHG if key fields cannot change during an update operation.
NOCHG is the default and only acceptable value for Primary Keys.
RMS-11 does not allow certain combinations of characteristics, depending
on the type of key (Primary or Alternate) being defined. Table D-2 summarizes these combinations.
Table D-2: Key Characteristic Combinations

D-4

CHG

NO CHG

Key Type

DUP

NODUP

DUP

NODUP

Primary

Error

Allowed

Default

Alternate

Default

Error

Allowed

RMSDFN Command Utility

NOC HG

When specifying the OP switch, you must use the parentheses shown in the
general form.
Example /KY:0:8:(/0P:DUP:CHG)

This switch defines an Alternate Key. In each record of the file, the key field
begins in the first byte and is eight bytes long. Duplicate key values can occur
among records of the file and key fields are allowed to change during update
operations.

• PR:options specifies the system protection code for the file being created.
Protection is available for the four IAS/RSX-llM categories:
System - access by the system's accounts
Owner - access by the owner
Group - access given other members in the owner's project
World - access given all other accounts
You can specify whether the accounts in each category can read, write,
extend, or delete the file. Therefore, the full format of the PR switch is:
/PR[/SY[:RWED][/OW[:RWED]l[/GR[:RWED]][/WO[:RWEDll
where:
/SY, /OW, /GR,
and /WO

are the subswitches that specify protection categories for
the file. If no subswitch is specified for a category, that
category is allowed complete access to the file.

R, W, E, and D

in any order, represent the Read, Write, Extend, and
Delete privileges granted to a category. If a subswitch is
specified, but no privileges are indicated, none are allowed for the category.

Protection can also be specified as an octal value in the PR switch form:
/PR:nnn
where:

nnn is the octal representation of the protection to be assigned to the file.
Refer to the RSX-llM Utilities Procedures Manual for a description
of the format of the octal protection word.
• PR:<nnn> specifies the protection status of the file. See the RSTS/E User's
Guide for a discussion of the values of nnn.
• RF:val[:n] specifies the record format and size for the file,
where:

val is the record format expressed as one of the following:
FIX - fixed-length
VAR - variable-length
VFC - variable-with-fixed-control; fixed control area defaults to two
bytes; restricted to Sequential and Relative files
STM - stream; restricted to disk Sequential files

RMSDFN Command Utility

D-5

specifies one of the following:
- when FIX is specified, the actual record size; a nonzero value is
required: there is no default length.
- when VAR, VFC, and STM are specified, the maximum record size
in bytes. If you enter a nonzero value, RMS-11 checks records written into the file against this maximum length, rejecting those records
that are too long. However, if 0 or no value is specified, RMS-11 does
not check record length. A nonzero value is required for Relative
files; that is, RMSDFN rejects the default value of 0 with the following message if file organization is REL.
?DFN --

ILLEGAL OPTION 1.JALUE -

/RF

If the format is VFC, the length does not include the fixed control area.

D.4.3 Examples
• DK1:ALPHA.DAT/FO:REL/CO/RF:FIX:128
RMSDFN attempts to create a Relative file, with a minimum initial allocation (one block) made contiguously. The records are 128 bytes long. The
bucket size defaults to one block.
• TEST.INP
RMSDFN attempts to create a Sequential file named TEST.INP. The records are variable in format with no specified maximum length. The initial
allocation is zero blocks.
• MASTER.FIL/FO:IDX/KY:0:20
RMSDFN attempts to create this minimally specified Indexed file. The
records are variable in format with no specified maximum length. Bucket
size defaults to one block. The Primary Key begins in the first byte of each
record and is 20 bytes long.
• TRANS .DAT/FO :IDX/BK:4/RF:FIX: 112/KY :0: 12/KY :19:6/KY :29 :4(/0P: NOCHG:DUP)
RMSDFN attempts to create the Index file TRANS.DAT. Each bucket
contains four blocks and the fixed-length records are 112 bytes long. Each
record contains three key fields:

D-6

Primary Key

begins in the first byte of each record and is 12
bytes long. By default, duplicate key values and
changes in key values during updates are not
allowed.

First Alternate Key

begins in the twentieth byte of each record and is
six bytes long. By default, duplicate key values
and changes in key values during updates are
allowed.

RMSDFN Command Utility

Second Alternate Key

begins in the thirtieth byte of each record and is
four bytes long. By specification, the key is duplicatable, but not changeable.

D.SI Cautions
0

RMSDFN creates Relative and Indexed files only on disk devices.

RMSDFN does not limit the record size specified in the RF switch. A file
therefore can be created with attributes indicating record sizes greater than
the maximums shown in Chapters 3, 4, and 6. However, processing such a
file causes unpredictable results.

RMSDFN Command Utility

D-7

Appendix E
Utillity Error Messages

E.1

RMS DEF Interactive Utility
BAD U!Ct TRY AGAIN

Description

The account number in the file specification is not valid for your computer
system for one of the following reasons:
• Either one or both parts of the account number exceeds the maximum value
allowed by your opera ting system:
Operating
System
IAS/RSX-11 M
RS'I'S/E

Range
1 through 377 (octal)
1 through 254 (decimal)

• The specified account has not been set up by the system manager.
Suggested Action

Type the file specification with a valid account number.

THE DEFAULT E}<TENS ION QUANT !TY MUST BE A MULTIPLE OF THE BUCKET SIZE•

Description

You typed a decimal number of blocks for Default Extension Quantity, but
that number is not an integral multiple of the bucket size you specified.
Suggested Action

Type the next higher multiple of bucket size for Default Extension Quantity.

E-1

** DEVICE IS FULL **
Description

RMSDEF attempted to create the file you defined, but could not allocate the
number of blocks you specified in the way you specified.
Suggested Action

If you specified a large contiguous initial allocation for the file, redefine the
file requesting either a smaller initial allocation or a noncontiguous allocation.
If you did not specify a large initial allocation, either contiguous or noncontig-

uous, then the specified disk is full. You should either:
• delete unnecessary files from the disk and reorganize empty blocks in a
single contiguous part of the disk using the DSC utility
• use a different disk pack on the device

** DEVICE NOT AVAILABLE **
Description

RMSDEF attempted to create the file you defined, but could not access the
specified device because it was not ready or on-line.
Suggested Action

Make the specified device ready or specify another device; and try again.
Description

RMSDEF attempted to create the file you defined, but the file processor
indicates that the specified account has not been set up.
Suggested Action

Specify another account number or request the system manager to create the
specified account; and try again.

ERROR IN LUN ASSIGNMENT
Description

RMSDEF assigned a logical channel erroneously.
Suggested Action

Recall the utility and try the dialog again. If this message occurs again,
submit a Software Performance Report as described in Section 9.0.2.6.

E-2

Utility Error Messages
:a a

JO -

ERROR WITH INDIRECT FILE
Description

You directed RMSDEF to read an indirect file, but the utility detected an
error in the file that is not covered by other error messages.
Suggested Actl on

Recall the utility and specify the indirect file. If this message occurs call your
local DIGITAL representative.
FILE SPECIFICATION IS TOO LONG
Description

You directed RMSDEF to read an indirect file, but the specified file contains
either a file specification longer than 120 bytes or a line (record) longer than
80 bytes.
Suggested Action

Edit the indirect file, shortening the file specification or line. Recall the utility
and specify the indirect file.
I-0 ERROR IN PROCESSING INDIRECT FILE
Description

You directed RMSDEF to read an indirect file, but the utility detected an 1/0
error during its processing.
Suggested Action

Recall the utility and specify the indirect file. If this error occurs again,
recreate the indirect file. If the new file fails, call your local DIGITAL
representative.
ILLEGAL DEVICEt TRY AGAIN
Description

RMSDEF attempted to create the file you defined, but encountered an error:
either the specified device is not on your computer system or the utility has
erroneously assigned its logical channel to another device.
Suggested Action

If the device is not on your system, specify another device and try again. If the

error occurs for a device recognized by your system, submit a Software Performance Report as described in Section 9.0.2.6.

Utility Error Messages

E-3

ILLEGAL FILE SPECIFICATIONt TRY AGAIN
Description

RMSDEF detected a syntax error in the file specification.
Suggested Action

Retype the file specification correctly.

ILLEGAL KEY SIZE. TYPE ? FOR HELP.
Description

You typed a key size that is not valid for the key data type you specified.
Suggested Action

Type a question mark (?) to print a HELP message showing the size restrictions for the different data types. Then, type a valid key size.

ILLEGAL VALUE. TYPE ? FOR HELP.
Description

You typed a value that is not a proper reply to the current question or request.
Suggested Action

Type a question mark (?) to print a HELP message that should help you
determine the proper values. Then type a proper value for the question or
request.

INVALID DATA IN COMMAND FILE
Description

You directed RMSDEF to read an indirect file, but the a line in that file did
not satisfy the utility's data requirements. If you specified the indirect file at
command level, RMSDEF returns to that level. However, if you specified the
indirect file after the utility printed its prompt, RMSDEF continues its dialog
with you rather than the indirect file.
Suggested Action

Correct or recreate the indirect file and try again, or continue the interactive
dialog.

E-4

Utility Error Messages

KEY EXCEEDS MAXIMUM RECORD SIZE.
Description

You typed a size for the key being defined which extends the key beyond the
end of the record you defined.
Suggested Action

Type a smaller size or terminate the current session with a CTRL/Z and start
again, defining a larger Maximum Record Size.

MAXIMUM NESTING DEPTH EXCEEDED IN INDIRECT FILE
Description

An indirect command file contains an indirect command. RMSDEF does not
permit nesting of indirect commands.
Suggested Action

Type the contents of one or both of the indirect command files as a sequence
of individual command strings.

OPEN ERROR IN INDIRECT FILE
Description

You directed RMSDEF to read an indirect file, but the utility could not open
the file. The file may not exist as specified; it may be locked; or there is some
other problem with the file.
Suggested Action

Correct the problem and try again.

** PRIVILEGE VIOLATION **
Description

RMSDEF attempted to create the file you defined, but the operating system
denied the utility access.
Suggested Action

Correct the problem and try again.

Utility Error Messages

E-5

SEGMENTS ARE NOT ALLOWED WHEN THE KEY rs NUMERIC
Description

You attempted to indicate segments for a key you defined as numeric (integer,
binary, or packed decimal).
Suggested Action

Enter one size or position.

SUM OF SEGMENTED KEY SIZES rs GREATER THAN 255+
Description

You typed a series of sizes for a segmented key. However, the sum of those
sizes is greater than 255 bytes, the maximum size for one key.
Suggested Action

Type the series of sizes again, making one or more of the segments smaller so
that their total is 255 or less. You could make two keys out of the one you are
attempting to define: reduce the segments in this key and when RMS DEF
asks if you have more keys to define, type "Y" and specify another key
containing the rest of the segments.

SYNTAX ERROR IN INDIRECT FILE SPECIFICATION
Description

You typed a file specification for an indirect command file, but there was a
syntax error in it.
Suggested Action

Type the file specification correctly.

THE BUCKET FILL SIZE rs GREATER THAN THE ACTUAL BUCKET SIZE.
Descrlptlo n

You typed a decimal number of bytes as a fill number. However, that number
is greater than the total bytes in the bucket you specified for the current area.
Suggested Action

Type a fill number smaller than the bucket size.

E-6

Utility Error Messages

THE NUMBER OF VIRTUAL BLOCKS PER BUCKET IS NOT BIG ENOUGH GIVEN
THE MAXIMUM RECORD SIZE YOU SPECIFIED.
Description

You typed a bucket size that would not contain the largest record you indicated in the Maximum Record Size question.
Suggested Action

Enter a larger bucket size; or terminate the utility, recall it, and redefine the
file with no or a larger Maximum Record Size.
TOO MANY AREAS! ! !
Description

You typed "Y" in response to the question:
DO YOU WANT TO DEFINE MORE AREASCNO>?

However, you have defined 255 areas, the maximum for one file.
Suggested Action

Continue with the RMSDEF interactive session and complete the file
definition.
TOO MANY KEYS!
Description

You typed "Y" in response to the question:
DO YOU WANT TO DEFINE MORE KEYSCNO>?

However, you have defined 255 keys, the maximum for one file.
Suggested Action

Continue with the RMSDEF interactive sess10n and complete the file
definition.
TOO MANY SEGMENTS
Description

You are defining a segmented key. However, the series of positions or the
series of sizes you typed contained more than eight numbers, the maximum
segments for one key.
Suggested Action

Type the number series again, using no more than eight values.

Utility Error Messages

E-7

WARNING - BE AWARE THAT THE FILENAME HAS BEEN TRUNCATED.
Description

In the file specification you just typed, you specified a file-name longer than
nine or six characters. RMSDEF continues the interactive session, but the
utility uses only the maximum number of characters allowed in the file-name
when it tries to create the file.
Example On a RSTS/E system, you typed the filespec:

PAYROLL.DAT
RMSDEF uses only "PAYROL" when it tries to create the file.

Suggested Action

Continue the session unless you cannot accept the file-name truncation.
~.

WARNING - BE AWARE THAT THE FILE TYPE HAS BEEN TRUNCATED.
Description

In the file specification you just typed, you specified a file extension longer
than three characters. RMSDEF continues the interactive session, but the
utility uses only the first three characters in the file extension when it tries to
create the file.
Example You typed the filespec:

PAYDAY.FILE
RMSDEF uses only "FIL" when it tries to create the file.

Suggested Action

Continue the session unless you cannot accept the shorter extension.

YOU ENTERED A DIFFERENT NUMBER OF SEGMENTS IN THE KEY POSITION
QUESTION.
Descrlptlo n

You are defining a segmented key. However, the series of key sizes you typed
does not contain the same number of segments as the series of key positions
you typed for the last question.
Suggested Action

Type the key size series again, specifying the same number of segments that
you used in the key position series.

E-8

Utility Error Messages

YOU HAVE SPECIFIED A VIRTUAL BLOCK NUMBER WHICH IS NOT WITHIN
THE BOUNDARIES OF THE FILE+
Description

You are using Virtual Block Number (VBN) alignment to place a file
area other than Area 0. However, the number you typed in response to the
LOCATION: request is not a valid VBN for the file you have defined to this
point.
Suggested Action

Recalculate the extent of the file you have defined to this point by adding the
initial allocation quantities for all the areas defined, starting with Area 0. Use
the sum to help you type a valid VBN.
YOU ORIGINALLY GAVE A BAD FILE SPECIFICATION WHICH WAS OUTPUT TD
YOUR COMMAND FILE+ YOU MUST EDIT THE CHANGE IN YOURSELF,
Description

When you typed a filespec for the file you were defining, you made an error.
RMSDEF allowed you to retype the filespec at the end of the interactive
session, after it failed to create the file. However, you also directed the utility
to create an indirect file during the session. That indirect file now has an
incorrect file specification at its beginning.
Suggested Action

Before you can use the indirect file, you must change that invalid file specification with a file editor.

,__,.

E.2 Command Utllltles
The format of an error message indicates the severity of the condition
encountered:
• A question mark (?) appearing as the first character of the message indicates a fatal error condition. Fatal error conditions cause the utility to
terminate processing at the point the condition was encountered.
• The ahsense of a question mark as the first character of the message indicates a nonfatal error condition. When a nonfatal error condition occurs, the
utility terminates the processing of only the file causing the error. The
utility continues processing any additional files specified in the command
string.
fu both types of error messages, the three character name of the utility en-

countering the error precedes the main text of the message.

Utility Error Messages

E-9

Italicized lowercase letters in the error messages listed in this section indicate
that the utility substitutes a value in the message printed at your terminal.
The following conventions indicate these substitutions:

CONVENTIONS
fipcode

A numeric value representing a file processor error code. When
an error message contains a code, refer to the description of
such codes in the:
• Error code appendix of the IAS/RSX-11 I/0 Operations
Reference Manual
• User recoverable error messages in an appendix of the
RSTS/E Programming Manual. Note that the code is the
negative of the decimal value shown in the Programming
Manual. That is, if the code is -20 10 , look up 20 10 •

rmscode

date
device
account

A numeric value representing an RMS-11 error code. When an
error message contains a code, refer to the description of such
codes in Appendix B.2.
A date you specified within the command line.
A device specification.
A User File Directory (UFD) or User Identification Code (UIC)
specification.

filespec

A file specification.

switch

A switch appearing within the command string.

utl

·~.

The three-character name of the utility encountering the error.

value

A switch value appearing within the command string.

v bn

A virtual block number.
~.

utl -- CHECK AFTER WRITE ERROR ON OUTPUT FILE - filespec,
l.JBN vbnl TO vbn2, CONT I NUE ( Y ,N)?

Description

This diagnostic error message prins on your terminal if Query Mode
is enabled and the RC switch (check after writing) is specified.
During data integrity checking, the utility found that the contents of
virtual blocks vbnl through vbn2 in the indicated file are not identical to the associated virtual blocks of an input file.
Suggested Action

Type "Y" to continue processing despite the error. Type "N" to
terminate processing the indicated output file and bypass the processing of any remaining files.

E-10

Utility Error Messages

?utl - -

CLOSE ERROR ON FI LE filespec t ERROR CODE= fipcode

Description

The specified input or output file could not be properly closed.
Suggested Action

Type the command string again. If the error recurs, run a validity
check of the file structure using the VFY/CLEAN utility on the
volume to determine if it is corrupted.

?utl --

COMMAND FILE NEST ING DEPTH E}·{CEEDED

Description

An indirect command file contains an indirect command. RMS-11
does not permit nesting of indirect commands.
Suggested Action

Type the contents of one or both of the indirect command files as a
sequence of command strings.

?utl - -

COMMAND LINE TOO LONG

Descrl ptlo n

The command string you typed is longer than the utility's command
string buffer (158 characters). Therefore, you have typed too many
continuation lines.
Suggested Action

Use several command strings to perform the functions of the original
command string. If the condition occurs repeatedly, submit an SPR
to DIGITAL.

?utl --

CONFLICTING OPTION - switch

Description

The indicated switch contradicts a previous switch within the same
command string (scanning from left to right).
Suggested Action

Type the command string with the correct switch.

Utility Error Messages

E-11

?utl - - CREATE ERROR ON filespec t ERROR CODE= fipcode

Descrl ptlo n

The specified file could not be created by the utility.
Suggested Action

Refer to the appropriate manual.

?utl - - DEi.JI CE NOT READY - device

Description

The message is self-explanatory.
Suggested Action

Correct the situation and type the command again.

?utl --

DEIJICE OFF LINE - device

Description

The indicated device is on the system, but access is prohibited for
one of the following reasons:
• The device is not ready.
• No volume is mounted on the device.
• The device is currently reserved by another job.
• The device requires privileges for use and you are not privileged.
• The device is disabled.
Suggested Action

Correct the problem and type the command string again.

?utl -- DEl.JICE WRITE PROTECTED - device

Description

The utility cannot access the device for write operations.
Suggested Action

Write enable the device.

E-12

Utility Error Messages

utl -- DEl.JICE/FILE IS FULL·- devicelfilespec

Description

The utility cannot create or extend the file on the device because of
insufficient space.
Suggested Action

Type the command using another device for output files or copy the
indicated file to another device and retry the command. You can
also delete unneeded files on the indicated device and retype the
command.
?utl --

DI RECTORY NOT FOUND ·- filespec

Description

The account does not exist on the specified device.
Suggested Action

Type the command with the correct account specification.
?utl - -

D !RECTORY SYNTAH ERROR - filespec

Description

The account portion of the file specification does not conform to the
syntax rules.
Suggested Action

Type the command string with the correct syntax.
utl -- DUP RCD= record

Description

The utility encountered an input record that would cause duplicate
values in an output Indexed file key field that does not allow duplicates. The utility did not insert the record into the output file, but
wrote the first 72 characters into the summary listing file.
Suggested Action

If the exception records are valid, redefine the output file to allow

duplicates in the necessary key fields, then repeat the command.
You could also insert the exception records into the current output
file, changing the key values causing the duplication. You can do
this by creating a Sequential file containing the records and using
RMSCNV, or you can use RMSCNV with your terminal as the input
file (outfile=TI:).

Utility Error Messages

E-13

utl -- EMPTY FILE

INDE}{ AREA

Descrl ptlon

The utility attempted to read records from an Indexed file using an
Alternate index without entries. This condition can occur:
• when records written into the file are too short to contain the
Alternate Key
• when occurrences of the Alternate Key field contain its null key
value
Suggested Action

Type the command string again, specifying the Primary index or
another Alternate index.

utl -- ENTER FILE IN DIRECTORY ERROR ON filespect ERROR CODE= fipcode

Description

The specified file-name could not be entered in the account.
Suggested Action

Refer to the appropriate manual.

utl -- ERROR IN FILE PROLOGUE - filespec

Description

The utility encountered an error in the Prologue of the indicated file.
The utility bypasses the file and continues processing.
Suggested Action

Use RMSDEF to define and create a file, then use RMSIFL or
RMSCNV to populate the new file from the corrupted one.

2. Use the last back-up copy of the file with the RMSRST utility.

E-14

Utility Error Messages

?utl --

ERROR IN TEMPORARY FILE

Description

The wild card processor called by the utility encountered an error
while creating a temporary file for resolution of wild cards in a file
specification.
Suggested Action

Type successive command strings to achieve the result wild cards
would produce.

?utl - -

ERROR INSERT TOO BIG t ERROR CODE= rmscode

Description

The utility cannot print the text of an error message on your terminal.
Suggested Action

Submit an SPR to DIGITAL documenting the conditions under
which the error occurred.

?utl - -

ERROR WITH LOG I CAL UN IT

Description

The utility assigned a logical channel erroneously.
Suggested Action

Recall the utility and try the command string again. If this message
recurs, submit a Software Performance Report as described in Section 9.0.2.6.

?utl - -

ERROR WI TH WI LDCARDS

Description

The wild card processor returned an error to the utility during resolution of wild cards in a file specification.
Suggested Action

Type the command string again. If the condition recurs, use successive command strings to achieve the result wild cards would
produce.

Utility Error Messages

E-15

?utl - -

EH TEND ERROR ON OUT PUT FI LE

Description

The output file you provided was not large enough to contain the
input records. The utility explicitly extended the file, but the operation failed.
Suggested Action

Determine why the file could not be extended (such as, disk full,
contiguous on RSTS/E, and so on). Allocate a larger output file and
type the command string again.

?utl --

FAILURE TO CREATE CONTIGUOUS FILE - filespec

Description

You specified the CO switch, but there is insufficient contiguous
space on the disk to allocate the specified file.
Suggested Action

Type the command without the CO switch. You could also delete
unneeded files from the disk and reorganize the empty blocks into a
contiguous space using the DSC utility.

?utl --

FAILURE TO OPEN COMMAND FILE

Description

An 1/0 or file system error, such as a privilege violation, an attempted write to a locked unit, parity error, and so on, occurred
when the utility attempted to open an indirect command file.
Suggested Action

Correct the problem and retry the command.

?utl --

FATAL RMS ERROR F I LE = files pee

PC= nnnnnnt STS= rmscodet STl,,I= rmscodet

Description

A fatal error occurred in the utility's interface with RMS-11 file and
record handling routines.
Suggested Action

Appendix B.2 describes the STS and STV codes.

E-16

Utility Error Messages

?utl - -

FI LE ACCESS ERROR ON filespec t ER RO RN CODE= fipcode

Description

The specified file could not be accessed.
Suggested Action

Refer to the appropriate manual.

?utl - -

FI LE ALREADY EH I STS - filespec

Description

The utility attempted to create a file that already exists.
Suggested Action

Type the command line using a new or corrected filespec or delete
the existing file and type the original command string again.

?utl -- FILE t filespec t CORRUPTED

Description

The internal structure of an RMS-11 file is corrupted, and RMS-11
cannot read or write records in the file.
Suggested Action

1. Use RMSDEF to define and create a file, then use RMSIFL or

RMSCNV to populate the new file from the corrupted one.
2. Use the last back-up copy of the file with the RMSRST utility.

?utl -- FILE LOCKED - filespec

Description

The specified file was improperly closed the last time it was accessed
and remains in a locked state.
Suggested Action

Use the UN switch with PIP to unlock the file.

Utility Error Messages

E-17

?utl - -

FI LE NAME SYNTAX ERROR - filespec

Description

The name portion of the file specification does not conform to the
syntax rules.
Suggested Action

Type the command string with the correct syntax.

?utl - -

FI LE NOT At.IA I LABLE - filespec

Description

The indicated file is being accessed for exclusive use by another job.
Suggested Action

Retry the command until the file is released.

[?]utl --

FILE NOT FOUND - filespec

Description

The indicated file was not found in the specified account and device.
Suggested Action

Verify the file specification and type the command string again.

?utl - -

FI LE POSIT I ON LOST - filespec

Description

The utility lost its position within a container file on magnetic tape
while rewinding or backspacing. The error may be caused by hardware failure.
Suggested Action

Determine from the output account or summary listing file the extent of the processing that was completed prior to the error. Type
the command string eliminating file specifications of files successfully processed. Use a new tape volume and/or a different tape drive.

E-18

Utility Error Messages

?utl - -

FI LE READ ERROR

Description

The utility encountered a hardware read error on an input or output
device.
Suggested Action

Use CTRL/Z to terminate the utility. Check input and output devices for hardware problems.
?utl - -

FI LE NAME TOO LONG

Description

You typed a file specification longer than 120 characters.
Suggested Action

Type the command string with a shorter filespec.
?utl --

FILE TYPE SYNTAX ERROR - filespec

Description

The extension of the file specification does not conform to the syntax
rules.
Suggested Action

Type the command string with the correct syntax.
?utl - -

FI LE t.JERS I ON SYNTA>< ERROR - filespec

Description

The version portion of the file specification does not conform to the
syntax rules.
Suggested Action

Type the command string with the correct syntax.
utl - -

FI ND FI LE ERROR ON filespec t ERROR CODE= fipcode

Description

The indicated file could not be found.
Suggested Action

Refer to the appropriate manual.

Utility Error Messages

E-19

?utl - -

ILLEGAL DEIJ I CE - device

Description

The device does not exist.
Suggested Action

Type the command string with a correct device specification.

?utl --

ILLEGAL FILE SPECIFICATION - filespec

Description

The file specification does not conform to the syntax rules.
Suggested Action

Type the command string with the correct syntax.

?utl - -

ILLEGAL IN PUT 0 PT ION - switch

Description

The indicated switch is not a legal infile switch for the utility.
Suggested Action

Type the command string with the correct switch or without the
indicated switch.

?utl - -

ILLEGAL MAH I MUM RECORD SIZE - filespec

Description

The Maximum Record Size attribute stored in the indicated file
exceeds the RMS-11 maximum. The file is corrupted.
Suggested Action

Move the disk pack to another device and try the command string
again. If the utility processes the file normally, there was a read
error on the first attempt. However, if the utility prints this error
message again, you must recreate the indicated file:
1. Use RMSDEF to define and create a file, then use RMSIFL or
RMSCNV to populate the new file from the corrupted one.
2. Use the last back-up copy of the file with the RMSRST utility.

E-20

Utility Error Messages

?utl - -

ILLEGAL NUMBER OF IN PUT FI LES

Description

You specified multiple input files for a utility that accepts only a
single input file.
Suggested Action

Type the command specifying a single input file.

?utl - -

ILLEGAL 0 PT ION VALUE - switch

Description

The value specified in the switch is illegal.
Suggested Action

Type the command string with the correct switch value.

?utl - -

ILLEGAL OUT PUT 0 PTI ON - switch

Description

The switch is not a legal outfile switch.
Suggested Action

Type the command string with the correct switch or without the
indicated switch.

utl -- ILLEGAL RMS FILE ORGANIZATION - value

Description

The switch value does not represent a legal RMS-11 file organization.
Suggested Action

Type the command with the correct switch value.

?utl - -

ILLEGAL RMS RECORD FORMAT - value
I

Description

One of the following conditions exists:
• The indicated switch value does not represent a legal RMS-11
record format.

Utility Error Messages

E-21

• The indicated record format is not permitted with the file
organization.
Suggested Action

Type the command with the correct record format.
?utl - -

ILLEGAL RMS RECORD SIZE

Description

One of the following conditions exists:
• You specified a record size greater than 16,383 bytes.
• You failed to specify a Maximum Record Size greater than zero
when creating a Relative file.
• The specified record size is too large for the buckets of the file.
Suggested Action

Type the command with the correct record size.

?utl --

ILLEGAL USE OF /AP, /BL, /MA DR /SU WITH FILE ORGANIZATION

Description

You combined one of the switches illogically with a file organization,
as follows:
• AP or SU with a non-Sequential file
• BL with a disk file
• MA with a non-Indexed file
Suggested Action

Type the command string without the switch.

?utl --

ILLEGAL USE OF /LO WITH FILE ORGANIZATION

Description

You specified the LO switch in an RMSCNV command string. However, the output file is not an Indexed file.
Suggested Action

Type the command string without the LO switch.

E-22

Utility Error Messages

?utl --

ILL USE OF /PD WITH OUTPUT FILE'S RECORD FORMAT

Description

You specified the PD switch in an RMSCNV command string, but
the output file contains variable-length records.
Suggested Action

Type the command string without the PD switch.

?utl - -

ILLEGAL USE OF I WF WITH RECORD FORMAT

Description

You specified the WF switch in an RMSCNV command string, but
neither the input nor the output file contains variable-with-fixedcontrol (VFC) records.
Suggested Action

Type the command string without the WF switch.

?utl - -

ILLEGAL USE OF WILD CARDS IN SUMMARY LISTI NG FI LE

Description

You used asterisks in a file specification as the argument of an SL
switch. RMS-11 does not permit wild cards in this context.
Suggested Action

Type the command string without wild cards in the summary listing
file specification.

?utl --

ILLEGAL USE OF WILD CARDS ON INPUT FILE

Description

The utility does not permit asterisks in an infile specification.
Suggested Action

Type the command string without wild cards m the input file
specification.
?utl - -

ILLEGAL USE OF WI LO CARDS ON OUT PUT FI LE

Description

The utility does not permit asterisks in the outfile specification.

Utility Error Messages

E-23

Suggested Action

Type the command string without wild cards in the outfile
specification.

?utl --

ILLOGICAL DEl.JICE - device

Description

The indicated device is not permitted in the context of the command string.
Suggested Action

Type the command string with an appropriate device specification.
?utl -- ILLOGICAL USE OF OPTION - switch

Description

The indicated switch is illogical in the context of the command
string.
Suggested Action

Type the command string with appropriate switches.

utl - -

INCOMPLETE OUT PUT FI LE REMA I NS - filespec

Description

The error condition described by a prior error message caused the
termination of processing before the indicated output file could be
completed and properly closed.
Suggested Action

None.

?utl -- INCORRECT FILE ORGANIZATION - filespec

Description

The organization of the file does not match the organization you
specified.
Suggested Action

Type the command line with the correct file organization.

E-24

Utility Error Messages

?utl - -

I NDEHED OUT PUT FI LE NOT EMPTY

Description

The output file you specified contains records.
Suggested Action

Specify or create another, empty output file.

?utl - -

IN PUT AND OUT PUT F U<ED CONTROL HEADER SIZES DO NOT CORRESPOND

Description

You attempted to write records from one file to another. Both files
contain variable-with-fixed-control (VFC) records, but the sizes of
the fixed areas are different.
Suggested Action

Redefine the output file and type the command again.

?utl - -

IN PUT

AND

OUT PUT RECORD

SIZES DO NOT CORRESPOND

Description

You attempted to write records from one file to another. However,
one of the following conditions exists:
• Both files have fixed-length records, but the record sizes differ.
• Both files have variable-length records, but the Maximum Record
Size of the input file is greater than the Maximum Record Size of
the output file.
Suggested Action

If you are using RMSCNV, use the TR or PD switch or both (see
Section 9.2.4.3). Otherwise, redefine the output file and retry the
command.

utl -- INPUT FILE IS NOT BACK-UP FILE - filespec

Description

The utility requires the input file to be a back-up file. You have
specified a file not in back-up format.
Suggested Action

Type the command string with the correct file specification.

Utility Error Messages

E-25

utl - - INPUT RECORDS NOT IN ASCEND I NG ORDER

Description

You specified the MA switch to RMSCNV, but your input file was
not sorted in ascending order by the output file's Primary Key value.
Suggested Action

Sort the input file or type the command again without the MA
switch.

utl -- INTEGRITY CHECK TABLE FULL. CONTINUE <YtN>?

Description

The table allocated internally by the utility to monitor input read
errors exceeded its capacity.
Suggested Action

Type "Y" to continue processing. Type "N" to terminate processing.

?utl -- INVALID /KR VALUE

Description

You specified a key of reference greater than 9. The utility cannot
process an Indexed file with more than ten defined keys.
Suggested Action

To process an Indexed file with more than ten keys, you must write
an application program.

utl - - I I 0 ERROR ENCOUNTERED ON IN PUT FI LE filespec t
1,IBN vbnl TO vbn2. CONT I NUE ( Y t N)?

Description

The virtual blocks cannot be read from the file. If the file is input to
RMSBCK, the utility cannot back up the data records within the
blocks. If the file is input to RMSRST, the utility cannot restore the
data within the blocks.
Suggested Action

Type "Y" if you want the utility to continue d~spite the error. Type
"N" if you want the utility to terminate.

E-26

Utility Error Messages

utl --· I/O ERROR ENCOUNTERED ON OUTPUT FILE - filespec

Description

One of the following conditions exists:
• The device is not on-line.
• The device is not mounted.
• The hardware failed.
• The volume is full.
Suggested Action

Correct the problern and type the command string again.
?utl - -

I I 0 ERROR ON COMMAND LI NE IN PUT

Description

The utility cannot read:
• the command string you typed at the terminal
• a line within an indirect file
Suggested Action

Type the command line again or recreate the indirect file and retry
the command.
?utl -- KEY OPTION COMBINATION CHG-NODUP ILLEGAL

Description

You attempted to create an Indexed file with an illegal combination
of characteristics for one or more keys.
Suggested Action

Type the command with legal key characteristics.
?utl --· /KR NOT ALLOWED FOR SEQUENTIAL DR RELATil.lE FILE

Description

The KR switch is permitted for Indexed files only. You either specified the wrong file or used the KR switch in the wrong context.
Suggested Action

Type the command with the correct file specification or without the
KR switch.
Utility Error Messages

E-27

?utl - - LABEL ERROR - device

Description

The magnetic tape volume does not have ANSI-standard labels.
Suggested Action

Mount the appropriate ANSI-labeled volume.
?utl - - MA}< I MUM RECORD EXCEEDED - filespec

Description

No more records can be written into the indicated Relative file because the file's Maximum Record Number has been reached.
Suggested Action

Create a Relative file with the RMSDEF utility or an application
program: specify an appropriate Maximum Record Number. Rerun
the utility.
utl -- MISSING OPTION t,IALUE - switch

Description

You did not supply a value with the indicated switch.
Suggested Action

Type the command specifying a value for the indicated switch.
?utl - - NO DE FAUL TS ALLOWED

Description

You did not specify a value for a quantity the utility cannot supply.
Suggested Action

Type the command string specifying all required values.
?utl - - NO KEY DECLARED FOR I NDE><ED FI LE

Description

You attempted to create an Indexed file without defining a Primary
Key.
Suggested Action

Type the command string with an appropriate Primary Key
definition.

E-28

Utility Error Messages

?utl - - NO RENAME ALLOWED

Description

The RMSBCK and RMSRST utilities require output file-names and
extensions be the same as the corresponding input file-names and
extensions. Only magnetic tape container files can have names that
do not correspond to an input name.
Suggested Action

Type the command string specifying wild cards for the name and
extension components of the output file specification.

utl -- NO SUCH FILE

Description

No file in the account you specified fits the wild cards in a file
specific a ti on.
Suggested Action

List the files in the account. Type the command string again with a
valid file specification.

?utl -- NO SUCH KEY FOR FILE - value

Description

The key of reference value you specified represents a key not defined
in the specified Indexed file.
Suggested Action

Type the command with a valid key of reference value.

?utl - - NOT A DI RECTORY DEV I CE - device

Description

You issued an account-oriented command for a device (such as a
printer) that does not have directories (accounts).
Suggested Action

._.,..

Type the command string without an account .

Utility Error Messages

E-29

?utl - - NOT A SHARABLE DEV I CE - device

Description

The command string you typed requires the sharing of a nonsharable
device.
Suggested Action

Type the command string with a correct device specification.

?utl -- NOT ENOUGH MEMORY AVAILABLE

Description

The memory required by the utility is greater than the job's memory
size maximum.
Suggested Action

Submit an SPR to DIGITAL documenting the conditions under
which the condition occurred.

?ut l - - 0 NL Y C* ,* J 0 R [ X , Y J I S LEGAL AS DEST I NAT ION

Description

You typed an outfile specification with an invalid account number.
The only acceptable forms are:
• Wild cards as [*,*]
• Two numbers separated by commas as [x,y]
Suggested Action

Type the command string with an account specification of[*,*].

?utl -- OUTPUT FILE MISSING

Description

You did not specify an output file in the command string.
Suggested Action

Type the command string with an output file specification.

E-30

Utility Error Messages

?utl - - OUT PUT FI LE MUST BE AN I NDEHED FI LE

Description

You specified a Sequential or Relative file as an outfile.
Suggested Action

Specify or create another, Indexed output file.

?utl -- OUTPUT IJOLUME CORRUPT - filespec

Description

While accessing the indicated file, the utility determined that the
output volume is corrupted.
Suggested Action

Mount another volume or use a different device and type the command with appropriate device specifications.

?utl - -

PR I MARY KEY CAN NOT CHANGE

Description

You specified that the Primary Key of an Indexed file can change.
RMS-11 does not permit Primary Keys to change.
Suggested Action

Type the command without specifying change for the Primary Key.

?utl - -

"-"''

PR I lJI LEGE l,l I OLAT I ON - filespec

Description

You do not have the privileges necessary to access the indicated file.
Suggested Action

Ask the owner of the file to change the protection code.

?utl - -

PROLOGUE l,JERS I ON NUMBER TOO HIGH

Description

The utility opened an RMS-11 Indexed file and checked the version
number of the file's prologue. That number indicates that the file
was created with RMS-11 attributes that the utility cannot handle.

Utility Error Messages

E-31

Suggested Action

Use a newer version of the utility or write an application program
using the latest RMS-11 software to perform the processing.
utl - - READ AFTER WR I TE ERROR ON OUTPUT FI LE filespec
l.JBN ubnl TO ubn2. CONTINUE CY,N>?

Description

This diagnostic error message prins on your terminal if Query Mode
is enabled and the RA switch (read after writing) is specified. During
data integrity checking, the utility encountered a read error while
attempting to read virtual blocks vbnl through vbn2 of the indicated
file.
Suggested Action

Type "Y" if you want the utility to stop processing the indicated file,
but process the rest of the command string. Type "N" if you want
the utility to stop processing the command string.
uU

READ ERROR, INTEGRITY CHECK TABLE AND REWRITE DATA MAY
HAVE BEEN LOST

Description

The utility encountered an error while attempting to read formatting
information in a back-up or container file.
Suggested Action

Retry the command, specifying the RC switch. If the same error
occurs, determine from the summary listing produced when the
back-up or container file was created, which files cannot be completely restored. If you have other back-up copies of these files, use
them with the command.
utl - - READ ERROR ON FI LE ATTRIBUTES - filespec

Description

The volume is corrupted or you do not have the necessary privileges
to access the indicated file.
Suggested Action

Use RMSDSP to determine the file's protection:

• If you are forbidden access to the file by its protection code, request that its protection be changed.
• If you are allowed access to the file by its protection code, then use
the VFY /CLEAN utility to check the volume for corruption.

E-32

Utility Error Messages

utl - - READ ERrWR ON FI LE PROLOGUE - filespec

Description

The utility cannot read the Prologue of the indicated file. The utility
bypasses the file and continues processing.
Suggested Action

Type the command specifying only the indicated file. If the same
error occurs, move the disk pack to another device and try the command string again. If the utility processes the file normally, a read
error occurred on the first two attempts. However, if the utility
prints this error message again, you must recreate the indicated file:
1. Use RMSDEF to define and create a file, then use RMSIFL or

RMSCNV to populate the new file from the corrupted one.
2. Use the last back-up copy of the file with the RMSRST utility.
utl - - READ ERROR OR I NCO NS I STENT DATA. MAY HAl,JE LOST FI LES.

Description

The utility encountered an error while reading a back-up or container file.
Suggested Action

Retry the command, specifying the RC switch. If the same error
occurs, determine, from the summary listing produced when the
back-up or container file was created, which files cannot be completely restored. If you have other back-up copies of these files, use
them with the command.
?utl - - RECORD TOO BIG - filespec

Description

A record from the input file exceeds the Maximum Record Size
specified for the output file.
Suggested Action

Use the RMSDEF utiHty to create a file with an appropriate Maximum Record Size.
?utl - - RELAT II.IE OUT PUT FI LE NOT EMPTY

Description

The utility requires that a Relative output file not contain any
records. However, the utility found one or more records in the output
file.
Utility Error Messages

E-33

Suggested Action

Use the RMSDEF utility to create an empty Relative file.
?utl - - RELEASE CORE ERROR

Description

The utility could not release task address space to the operating
system.
Suggested Action

The utility finished processing the current command string. If you
want to continue using the utility, you should terminate it with
CTRL/Z (if necessary) and then recall it.
utl - - REQUESTED RECORD LOCKED. CONT I NUE ( Y t N)?

Description

The utility is sharing a file with another user. That user is accessing
a bucket required by the utility.
Suggested Action

Type "Y" if you want the utility to wait for the record to be released,
then continue processing. Type "N" to terminate the utility. Since
this condition can occur repeatedly while the file is shared with
another user, you can terminate the processing and retry the command later.
?utl -- REWIND OR SPACE ERROR ON FILE - filespec

Description

The utility encountered an error while rewinding or backspacing on
magnetic tape.
Suggested Action

Retry the command. If the condition occurs again, mount the volume on another drive and retry the command.
?utl - - SELECT EPROR - device

Description

One of the following conditions exists:
• The device is not on-line.

E-34

Utility Error Messages

• The device is not mounted.
• The hardware failed.
Suggested Action

Rectify the condition and type the command string again.
utl - -

SOME DUPLICATE RECORDS

NOT

WRITTEN

Description

The utility encountered at least one input record that would cause
duplicate values in an output Indexed file key field that does not
allow duplicates. The utility does not insert the ;record into the output file, but prints this message on your terminal.
Suggested Action

Type the command again, specifying a summary listing file (see
Section 9.2.4.2). The utility writes into the file the first 72 characters
of each exception record. See the error message "DUP RCD=" for
suggested action.
?utl - -

SYNTAX ERROR

IN COMMAND

STRING

Description

The command string you typed contained a syntax error.
Suggested Action

Type the command string correctly.
?utl - -

SORTS ERROR

CODE

IN OCTAL :nnnnn

Description

The SORT-11 modules in RMSIFL encountered an error, where
nnnnn is a sort error code.
Suggested Action

Refer to the PDP-11 SORT Reference Manual for a description of
the indicated error code. Correct the problem and type the command
string again.
?utl - -

SW ITCH

/TR

I PD OR BOTH ARE NEEDED FOR TH IS CONVERT

Description

The record size for the input file is different from the record size for
the output file.

Utility Error Messages

E-35

Suggested Action

Read "Outfile Switches," Section 9.2.4.3, for the use of the TR and
PD switches. Then, type the command again using one or both of the
switches.
?utl - - SYN TA}{ ERROR string

Description

The format of the command string does not conform to the syntax
rules. The utility prints the command string from the point of the
error to the end of the line.
Suggested Action

Refer to the appropriate utility description in Chapter 9. Type the
command string with the correct syntax.
?utl - - TI LUN ASSIGNMENT ERROR

Description

The utility cannot communicate with your terminal.
Suggested Action

Consult with a DIGITAL Software Support specialist.
?utl -- TOO MANY KEYS - filespec

Description

The specified Indexed file has more than ten keys defined. The utility cannot process an Indexed file with more than ten keys.
Suggested Action

You must write an application program to process an Indexed file
with more than ten keys.
?utl

TOO

MANY SELECTIIJE FILE SPECIFICATIONS

Description

More than ten file specifications appear as switch values of the SE
switch.
Suggested Action

Type the command line specifying no more than ten file specifications as switch values of the SE switch. Use multiple command
strings to select additional files.

E-36

Utility Error Messages

utl - - UNABLE TO RESTORE SPEC I AL ATTRIBUTES - filespec

Description

The utility cannot restore the file with one or more of the following
attributes from the original file:
• one or more of its original date attributes
• its original protection specification
• contiguity
• placement control
• areas
Suggested Action

Use the RMSDSP utility to determine which attributes of the file
were not restored.
utl - - UNCORRECTABLE ERROR t 1.JBN vbnl TO vbn2
REASON= message

where:
message is one of the following:
READ CHECK ERROR
WRITE CHECK ERROR
INPUT ERROR
OUTPUT UNCHECKABLE

Description

This diagnostic error message prints on your terminal if Query Mode
is enabled. During automatic retry of errors found during data integrity checking (RA or RC switch) or while reading a file, the utility
encountered an uncorrectable error.
Suggested Action

Record the ranges of erroneous virtual blocks in the file.
?utl -- UTILITY INPUT ERROR

Description

A severe, unrecoverable error occurred during the execution of the
indicated utility.

Utility Error Messages

E-37

Suggested Action

Type the command again. If the same message appears, contact a
DIGITAL Software Support specialist.

?utl - - WRITE ERROR ON ATTRIBUTES OF FI LE - filespec

Description

The volume is corrupted or you do not have the necessary privileges
to write the file.
Suggested Action

Verify access to file.

?utl - - WRITE ERROR ON CREATE OF OUT PUT FI LE - filespec

Description

One of the following conditions exists:
• The device is not on-line.
• The device is not mounted.
• The hardware failed.
• The volume is full.
Suggested Action

Rectify the condition and type the command string again.

?utl - - WR I TE ERROR ON INTEGRITY CHECK TABLE ON OUT PUT FI LE files pee

Description

The utility cannot write internal data integrity checking tables in
the output back-up file.
Suggested Action

If the output medium is magnetic tape, use a different tape volume
and retry the command. If the output medium is disk, rename the

output file so that the utility will not attempt to use the space and
retry the command.

E-38

Utility Error Messages

Appendix F
Magnetic Tape Handling

HMS-11 supports only the magnetic tape structure defined by the American
National Standards D (ANSI-D) format. This standard is described
"Magnetic Tape Labels and File Structure for Information Interchange,"
ANSI X3.27-1969.
RMS-LL enforces only one part of the ANSI-D standard; the operating
system's file processor enforces the rest, primarily the tape labeling formats.
As a result, magnetic tape handling is largely transparent to RMS-11 and the
RMS-LL user.
This appendix contains the few RMS-11 processing options that differ with
magnetic tape.
For documentation about using magnetic tapes on your system, including
creating ANSI-standard tapes and tape label formats, see:
• For IAS, the !AS System Management Guide and the IAS/RSX-11 1/0
Operations Reference Manual
• For RSTS/E, the RSTS/E System User's Guide, RSTS/E System Manager's
Guide, and RSTS/E Programming Manual
• For RSX-llM, the RSX-llM Operator's Manual and the JAS/RSX-111/0
Operations Reference Manual

F.1

General Magnetic Tape File Processing
Magnetic tape is a sequential access, single-directory medium. Only one user
can access a given tape reel, called a volume, at a time. No more than one file
in a volume can be open at a time.

F-1

You can write files on tape volumes in the following combinations:
• Single file on a single volume
• Single file on more than one volume
• Multiple files on a single volume
• Multiple files on more than one volume

NOTE
When you have more than one tape volume that are processed together, the related volumes are called a volume set.
And when a file is stored on more than one volume, the part
of the file stored on each volume is called a file section.

The file processor distinguishes between volumes and between files with
labels, short sections of tape written to a defined format. The following labels
are used on ANSI-standard tapes.
• Volume label (VOLl)
• End-of-volume labels (EOVl) and (EOV2).
• File header labels (HDRl) and (HDR2).
• File trailer labels (EOFl) and (EOF2).
• User labels (not supported by RMS-11 or the file processors)
The following graphics illustrate volume structures supported by PDP-11
operating systems according to the ANSI-D standard.

NOTE
These conventions are used in the graphics:
• An asterisk indicates a tape mark.
• BOT indicates Beginning-of-Tape.
• A comma indicates a physical record delimiter.
Single File on a Single Volume
60TtVOL1 tHDR1 tHDR2*---DATA---*EOF1 tEOFZ**

Single File on Multiple Volumes
BOT ,tJOL1 tHDR1 tHDRZ*---DATA---*EOl.11 tE0'.12**
6QT,t.IQL1 tHDR1 tHDR2*---DATA---*EDF1 tEOF2**

F-2

Magnetic Tape Handling

Multiple Files on a Single Volume
BOT t 1..JOL1 tHDR1 tHDR2*---DATA---*EOF1 tEOF2*HDR1 tHDR2*---DATA--*

EOF1 tEOF2**

Multiple Files on Multiple Volumes
BOT tl.JOL1 tHDR1 tHDR2*--DATA--*EDF1 tEClF2*HDR1 tHDR2*---DATA--·M-E0l.J1 tEOt.12**

BOT t1.10L1 tHDR1 tHDR2*--DATA--*EDF1 tEOF2*HDR1 tHDR2*--DATA--*EOF1 tEOF2**

F.2 RMS-11 Magnetic Tape File Processing
RMS-11 requires the unused characters at the end of a block on magnetic
tape to be circumflexes ("):
• While reading a tape, RMS-11 returns the error code ER$ANI if it encounters any character other than a circumflex between the last record in a
block and the end of the block.
• When writing a tape, RMS-11 monitors the end of each tape block, whose
length is set when you create the file. RMS-11 checks each put operation to
see if the specified record fits in the tape block. If it does not, RMS-11
writes circumflexes to the end of the block and then starts a new block with
the specified record. Records do not span blocks in magnetic tape files.

F.2.1 Rewinding Tape Volumes
You can rewind a magnetic tape volume when a file on it is created, opened,
or closed. You specify one of the following options:
• Rewind-on-open as input to a file creation operation.
• Rewind-on-open as input to a file open operation.
• Rewind-on-close as input to a file creation operation.
• :Rewind-on-close as input to a file open operation.
• Rewind-on-close as input to a file close operation.

NOTE
If you specify the rewind-on-close option during file creation

or open, you cannot reset that requirement before you close
the file. However, if you do not request rewind-on-close
during the create or open, you can select the option before
initiating the close operation and the file processor rewinds
the tape after it closes the file.

Magnetic Tape Handling

F-3

When requested to rewind a tape volume, the file processor performs one of
the following steps:

• If you mounted the volume with a specific command, the file processor
rewinds the tape to Beginning-of-Tape (BOT). If the volume is not the first

volume in the set, the file processor rewinds the tape and then puts the
device off-line.
• If you have not mounted the tape and:

the rewind is requested during file creation or open operation, the file
processor requests the system operator to mount the volume on the device included in the file specification supplied for the operation.

the rewind is requested during file close, the file processor rewinds the
tape to BOT. If processing continues and the next volume in the set is
required, the file processor prints the mount request.

F.2.2 Positioning for the Next File
During file creation, you can specify that the file must be written immediately
after the file labels of the most recently closed file. The new file logically
deletes any files on the rest of the tape.
If you also specify the rewind-on-open option, the file processor creates the file

after the VOLl label on the volume. Files previously contained on the volume
are lost.
Normally, the file processor positions a new file at the logical end of the
volume (after the last file currently on the tape). While it is doing this, the file
processor does not check the volume for a file with the same name. Therefore,
you could add a new version of a file to a tape, but not be able to read it: each
time you open the file, the file processor locates the first physically occurring
(the older) version of the file.
Therefore, in most processing, you should explicitly position new files.
Example You perform single-file operations by specifying the rewind-on-open and rewind-onclose options during file creation or open. Using this approach, you use a scratch
tape as follows:
1. Create a file with rewind-on-open.

2. Write the data records and close the file with rewind-on-close.
3. Open the file again for input.
4. Read and process the data.
5. Close the file with rewind-on-close.
6. Create a second file with rewind-on-open.
7. Write the data records.
8. Close the file with rewind-on-close and perform any additional processing.

F-4

Magnetic Tape Handling

Example You create multiple files on a volume by creating, writing, and then closing a series
of files without specifying the rewind-on-open option. You sequentially process files
on a volume by not specifying the rewind-on-close option.
You use the following tape operations to create a multiple-file tape volume:
1.

Create a file for output with rewind-on-open.

2. Write data records and close the file without rewind.
3. Create the next file without rewind-on-open.
4. Write data records and close the file without rewind.
5. Repeat steps 3 and 4 as necessary. If you fill the volume:
• The RSTS/E file processor does not support volume sets. Therefore, if you
exhaust a tape in the middle of a file, you must monnt another ANSI-labeled
tape and start the data transfer from the beginning of the file. If you have the
Large Files option, you cannot copy a file larger than about 60,000 disk blocks
on a 2400-foot tape at 1600 bits-per-inch (approximately 24 blocks per foot).
• The IAS and RSX-llM file processors automatically provide multivolume
tape support. When the processor detects End-of-Tape, it generates the proper
labels, rewinds the volume, and requests the system operator to mount another
volume. Once a new tape is ready, your program continues processing.
Tape files can be opened in a nonsequential order, but this increases processing and
tape positioning time. Avoid nonsequentially accessing files in a multivolume set.

Magnetic Tape Handling

F-5

._.,.

Glossary

"-''
15-·bit Signed Integer (Key)

Same as Two-Byte Signed futeger.
16-bit Unsigned Binary (Key)

Same as Two-Byte Unsigned Binary.
31-bit Signed Integer (Key)

Same as Four-Byte Signed Integer.
32-bit Unsigned Binary (Key)

Same as Four-Byte Unsigned Binary.
Acc:ess
1.

The ability to read or write data in a file.

2. To :read or write data in a file.
Acc:ess by Record's File Address

A Record Access Mode where a program can randomly access a record by its Record's
File Address.
Acc:ess Declaration

Included in the information a program provides when it opens a file, the access
declaration indicates the record operations the program will perform on the file.
Contrast with Allow Declaration.

Glossary-I

Access Path

The sequence of steps RMS-11 performs to transform the parameters in the user's
record operation request into the requested record. See also Index, Key.

Access Time

The period between the initiation of a record operation and its completion. During
this period, RMS-11 reads, writes, and/or modifies portions of a file as required by
the record operation.
Active Page Register

A hardware register used by the operating system to map the virtual address space
used by a task onto physical memory.
Allocation

The disk blocks associated with a file; normally, the size of a file established when
the file is created.
Allow Declaration

Included in the information a program provides when it opens a file, the allow
declaration indicates the type of record operations the program allows other programs
to perform on the file. Contrast with Access Declaration.
Alternate Index

A structure providing a secondary logical access path to records stored in an RMS-11
Indexed file. Each Alternate index is based on a related Alternate Key field in the
user data record. See also Primary Index.
Alternate Key

A series of bytes in a data record that can be used to identify the record for access.
Alternate Key value does not affect the position of the user data record in the file.
Contrast with Primary Key. See also Segmented Key.
Application Program

A program that processes data for an end user of a computer system. Normally, any
program not part of the operating system. See also Program, Utilities.
APR

Same as Active Page Register
Area

A portion of an Indexed file treated independently by RMS-11 for initial allocation,
extensions, placement, and bucket sizes.

Glossary-2

Area Descriptors

Data used by RMS-11 to maintain areas in an Indexed file. RMS-11 stores this data
in the file's Prologue. See also Key Descriptors.

"-11'

Assembler

Software that converts assembly-language mnemonic instructions to object code. The
assembly language for the PDP-11 is called MACR0-11. See also Compiler.
Asynchronous Record Operation

A record operation where RMS-11 may return control to your program before
the operation is finished. The program continues processing while the physical
transfer(s) of data between disk and memory is carried out. Contrast with Synchronous
Operations.
Bac:k up

1. The process of copying data to a medium that is preserved in case the original
data is lost or destroyed.
2. Result of that process.
3. RMSBCK utility.
See also Restore.
Bimuy Search Technique

A searching technique applicable only to items ordered by the value in the search
argument field. A binary search proceeds as follows:
1. Locate an item at the middle of the items searched.
2. Divide the items into three parts: low part, high part, and the item found.
3. If the item found does not satisfy the search, repeat this procedure on the low or
high part depending on a comparison of the search value with the item found.
Bit

The smallest storage location recognized by PDP-11 hardware. A bit is a hardware
location (in physical memory, on a disk or tape surface, and so on) that can assume
one of two recognized values, conventionally designated "O" and "1". See also Byte,
Block, Track, Cylinder.

1. A logical unit of disk storage, containing 512 bytes.
2. A logical unit of magnetic tape storage, containing from eight through 8192 characters. See also Virtual Block, Logical Block Number, Sector.

Glossary-3

Block 1/0

The mode of file access that bypasses RMS-11 record processing. Contrast with
Record Processing. See also Undefined Record Format.
Block Spanning

You must specify whether you allow records to span blocks when you create Sequential files. If you allow block spanning, records can cross block boundaries and are not
restricted by block size. In Relative and Indexed files, records can automatically
span blocks if a bucket contains more than one block; however, records cannot span
buckets.
Bootable Volume

A disk containing a bootstrap loader program in Logical Block 0. Since this boot
block is not a part of any file, RMSBCK and RMSRST do not read and write it.
Bucket

The I/O and disk storage unit for Relative and Indexed files.
Bucket Header

Fifteen bytes of control information that RMS-11 uses in each Indexed file bucket.
See also Record Header, Record Format Overhead.
Bucket Locking

A process administered by the operating system between files and by RMS-11 between Record Access Streams.
RMS-11 activates system-level bucket locking for a Relative or Indexed file when the
first program to open it allows write sharing. During record operations, RMS-11
requests the operating system to lock specific virtual blocks comprising the buckets
being used.
RMS-11 maintains its own locked-block list when a program connects more than one
Record Access Stream to a Relative or Indexed file.
When one Record Access Stream locks a bucket, no other program or stream can
access it. See also Sharing Specifications in Accessing Programs.
Bucket Pointer

In this manual, part of an index record that indicates the bucket whose High-Key
Value is contained in the rest of the index record.
Bucket Size

The number of blocks in a bucket. Bucket size is an attribute of a Relative file or an
Indexed file area and cannot be changed for the life of the file.

Glossary-4

Bu<:ket Splitting

The process of formatting a new bucket in the file and moving the high portion of the
target bucket into the new bucket. Bucket splitting occurs during put and update
operations when both of the following are true:

.'-".

• RMS--11 must insert a record into the target bucket to preserve ascending key value
sequence or expand a record already in the target bucket.
• The record will not fit in the target bucket because there is not enough free space.
Buffer

A part of the virtual memory dedicated to a task. A buffer is a holding area for data
moved to and from other storage areas. See also I/0 Buffer, User Buffer.
Byte

A hierarchical unit of disk and memory structure, containing eight bits. See also
Sector, Track, Cylinder.
Ca<:he, Caching

The process of storing blocks in memory against future need; used to minimize
physical transfer of data between mass storage devices and memory.
Ca<:he Cluster

A single- or multi-block unit of the RSTS/E data cache. The caching software imposes a grid of cache clusters on the logical structure of a disk. An I/O operation that
reads blocks to be cached is broken into disk access requests by cache cluster. This
mechanism requires careful alignment of cache, pack, and file clusters.
Carriage Control

A type of forms control carried as a file attribute. When records from such a file are
written directly to a unit record device, the device driver puts a line feed character
before the record and a carriage return character after the record before passing it to
the device.
Cell

Same as Record Storage Cell.
Change Key (Characteristic)

A flag associated with an Alternate Key. The flag indicates whether you allow or do
not allow the key field to change values during an update record operation. See also
Duplicate Key, Null Key.
Check-After-Write

An optional data integrity check performed by the RMSBCK and RMSRST utilities.
See also Read-After-Write.

Glossary-5

Close (File Operation)

RMS-11 terminates access to a file and releases control structures associated with the
file. Contrast with Open.

Cluster

A sequence of logically contiguous blocks treated as a unit. See also Cache Cluster,
Device Cluster, File Cluster.
Clustersize

The number of blocks in a cluster.
Command File

Same as Indirect File.
Command String

A series of characters, terminated with the RETURN key, with which you instruct a
utility to perform processing. A command string normally includes one or more file
specifications and· optional switches.
Compiler

Software that converts higher-level-language instructions to object code. See also
Assembler.
Concatenate

To string separate entities together end-to-end. Especially, overlay segments. See
also Contiguity.
Connect (Record Operatio91)

RMS-11 makes the records of a file available to your program for a stream of operations. Contrast with Disconnect.
Container File

A file on magnetic tape containing one or more RMS-11 files in back-up format.
When the output medium is tape, each command string to RMSBCK produces one
container file.
Context

The position of a Record Access Stream within a file, consisting of Current Record
and Next Record.
Contiguity

The property of having all parts touching. A contiguous file contains a continuous
series of logical blocks; all logical blocks between the first and the last ones, inclusive,
belong· to the file.

Glossary-6

Con1tinuation Bucket

An Indexed file bucket containing part of a series of data records with the same key
value. Duplicate series can exist only in Level 0 buckets. When a duplicate series
grows beyond a single bucket, RMS-11 continues it in a separate Continuation
Bucket that is not represented by an entry in Level 1. All records in a Continuation
Bucket contain the same key value.
Con1trol Structure

A part of virtual memory used by RMS-11 routines to communicate with the program and with each other. See also 1/0 Buffer, User Buffer.
Convention

A method or structure of presentation that enables easier communication.
Convert
1. To use the RMSCNV utility to copy data from one file to another. Since the files

may have different attributes, the data must be converted from its input form to
the proper output form.
2.

RMSCNV utility.

Cra:;h Dump

The output from an RMS-11 utility when it cannot recover from an error condition.
Create (File Operation)

RMS-11 passes all necessary information to the file processor, then requests it to
create a file as specified. Contrast with Erase. See also Open.
Creation Time

That time when you actually instruct the file processor to crea@a file.
Cunent Record

Part of the context of a Record Access Stream. The Current Record is:
• established by a successful find or get operation
• the target of a delete, get (if immediately preceded by a find), truncate, or update
operation.
Since only get or find operations set the Current Record, one of these operations must
precede an update or delete operation. Other operations leave the stream without a
Current Record. RMS-11 rejects any update or delete operation attempted without a
Current Record.
See also Next Record.

Glossary-7

Cylinder

The tracks at a single radius on all platters of a disk. See also Bit, Byte, Sector,
Track.

Data Bucket

A bucket in an Indexed file that contains either user data records or SIDRs. See also
Index Bucket.
Data llntegrity Check

An optional process performed by RMSBCK or RMSRST to validate the data that
has been transferred from one volume to another. See also Check-After-Write, ReadAfter-Write.
Data level

The lowest level of an index, containing data buckets. See also Index Level, Lowest
Index Level.
Data Record

Same as User Data Record. Also a SIDR in Alternate indexes. See also Index Record.
DCN

Same as Device Cluster Number.
Default Extension Quantity

The number of blocks that RMS-11 requests the file processor to add to a file when
RMS-11 must extend the file automatically to complete a record operation. See also
Allocation, Extend.
Default Value

The value supplied for an operation parameter when you do not provide an explicit
value.
Deferred Write

An RMS-11 I/0 technique. RMS-11 does not write the data in an I/0 buffer to disk
until it must use that buffer for other data.
Delete (Record Operation)

RMS-11 marks a record in a file, indicating that the record is no longer valid.
Depth

The number of the Root level in an index.

Glossary-8

DEC~

Same as Default Extension Quantity.
Desiign

The process of considering the goals of an application and its files and the environment RMS-11 provides, and then planning both for optimal performance.
Deviice Cluster

A single·· or multi-block unit of disk storage assignment whose size is characteristic of
a type of device.
Device Driver

Software written for a specific type of hardware device that instructs devices of that
type during data transfer and other operations. See also File Processor.
'-"'

Dire·ctory

A file on a mass storage device that describes the layout of the data on that device in
terms of file-names, lengths, locations, last date of access, protection codes, and so
on.
Directory Caching

A RSTS/E technique. Directory blocks are cached in memory.
Disconnect (Record Operation)

RMS--11 terminates a stream of record operations, making the buffers assigned to the
stream available for other operations. Contrast with Connect.
Disk~ Head

The electro-mechanical device that reads and writes information on disk platters.
Disk-Resident Overlay (Structure)
1.

A segment of a task that resides on disk until required by the memory-resident
portion of the task.

2. The tree-like structure that relates the segments. Contrast with MemoryResident Overlay.
Display
1.

To print the attributes and structural data of a file on a terminal or to a file with
the RMSDSP utility.

To make file attributes and structural data available to a program with the
RMS-11 $DISPLAY file operation.

3. RMSDSP utility.

Glossary-9

Duplicate Key (Characteristic)

A flag associated with a Primary or Alternate Key. The flag indicates whether you
allow or do not allow more than one record in a file to contain the same value in a
specific key field. See also Change Key, Null Key.

Duplicate Pointer Array

A series of record pointers stored in each SIDR for an Alternate Key where duplicates
are allowed. Each pointer indicates a user data record containing the key value
represented by the SIDR.

Dynamic Access

The ability to change Record Access Mode with each record operation. In PDP-11
COBOL, you must declare dynamic access when you open a file. RMS-11 does not
make this requirement.

End-of-Fiie (Attribute)

The location of the end of useful data stored in a file. The location indicated is not
necessarily the last block in the file.
Equal Access Times

All records in an RMS-11 Indexed file can be accessed by Primary Key with the same
number of I/O operations, regardless of how long they have been in the file.

Erase (File Operation)

RMS-11 requests the file processor to delete the file from the device directory and
release its blocks for re-use. Contrast with Create.
Exclusive Access to a Fiie

A type of file access that prohibits concurrent access to a file by any other task. Not
available through RMS-11.
Executable Task

Same as Task.
Executive
In IAS and RSX-UM, the part of an operating system that coordinates the other
components. See also Monitor.

Glossary-IO

Expc>ne ntial Variance

"-"

Extend (File Operation)

RMS-11 requests the file processor to add blocks to a file's allocation. See also
Default. Extension Quantity.
Extended Buffer Pool

In RSTS/E, a portion of a computer's physical memory that is reserved for use by the
operating system. Message send/receive, the DECnet/E package, the RSTS/2780
package, the FIP buffering module, and the data caching module use this reserved
memory; if none is available, they use small buffers.
Extended Diagnostic Messages

A feature of RMSBCK and RMSRST: the utility prints a detailed error message on
your terminal, allowing you to decide whether to continue processing. Also called
Query mode.
Exte1nt

A portion of a file containing contiguous blocks that is not recognized by the file
processor as being contiguous with other portions of the file. See also Allocation.
F11ACP

Same as FILES-11 Ancillary Control Processor.

Glossary-11

FAB

Same as File Access Block.
FCB

Same as File Control Block.
FCS-11

Same as File Control Services.
Fifo

Same as First-In, First-Out.
File

A logical container of data, individually maintained by the file processor without
regard to the nature of its contents.

File Access Block

An RMS-11 control structure that represents a file during file operations.
File Attributes

Characteristics of a file stored in the file directory and Prologue. They include file
organization, record format, and so on.
File Cluster

A single- or multi-block unit of disk storage assignment. File cluster is the minimum
unit used by the file processor during file allocation and extension. File clustersize is
a file attribute.
File Control Block

In IAS and RSX-llM, a memory-resident structure used by the operating systems to
coordinate access to a file opened by an active task.
File Control Services

A file management software product available on IAS and RSX-llM operating
systems.
File Id

An identifying notation optionally returned when a file is created or opened by file
specification; can be used to open file.
File Operation

A file-level function, including create, open, close, extend, and erase. See also Record
Operation.

Glossary-12

File Organization

Method of arranging records within a file. File organization establishes the availability of access methods, record operations, and other ways of using the file.
File !Processor

1. A component of the operating system that maintains the structure and integrity
of data storage on file-structured devices.
2. In RSTS/E, the file processor.
See also FILES-11 Ancillary Control Processor.

File :Specification

A string of identifiers that specify a particular file or class of files.

FILES-11 Ancillary Control Processor

In IAS and RSX-llM, the file processor.

FILES-11 ODS·-1

On-Disk Structure Level One.
FILES-11 ODS·-2

On-Disk Structure Level Two.
File spec

Same as File Specification.

Fill Number

The number of bytes in an Indexed file bucket that you want used to store records.
Data level and index level fill numbers can be set for each index in the file, but are
honored by RMS-11 only during operations that so request. By default, RMS-11 uses
all free bytes in a bucket to store records.

Find (Record Operation)

RMS-11 locates the record specified, but does not move it from the I/0 buffer to the
user buffer.
FIP

Same as File Processor

Glossary-13

First-In, First-Out

Pertaining to a technique of storing serial values. The first value stored is the first one
found in a sequential access. RMS-11 uses this technique when it stores user data
records containing duplicate Primary Key values and the SIDR pointers to duplicate
Alternate Key values.

~"""'

Fixed-Length (Record Format)

Records in the file are the same length. The length is a file attribute enforced
by RMS-11 during write-type record operations. See also Stream, Variable-Length,
Variable-with-Fixed-Control, Undefined.
Flush (Record Operation)

RMS-11 writes all 1/0 buffers to a file if they haven't already been written.
Follow the Index

The procedure that RMS-11 uses when it must start at the Root and read down an
index to a specific record location.
Forms Control

Codes passed to a unit record device that control the printing or display of data on
the device. See also Carriage Control, FORTRAN Forms Control.
FORTRAN Forms Control

A type of forms control carried as a file attribute. The device driver interprets the
first byte of each record as a FORTRAN forms control character.
Four-Byte Signed Integer (Key)

A key data type that can represent the decimal integer values -2,147,483,648 through
+2,147,483,647. See also Two-Byte Signed Integer, Two-Byte Unsigned Binary, FourByte Unsigned Binary.
Four-Byte Unsigned Binary (Key)

A key data type that can represent the decimal integer values 0 through
+4,294,967,295. See also Two-Byte Unsigned Binary, Two-Byte Signed Integer, FourByte Signed Integer.
Free (Record Operation)

RMS-11 releases buckets locked by a Record Access Stream. See also Bucket
Locking.
Get (Record Operation)

RMS-11 locates the specified record and normally moves it from the I/0 buffer to the
user buffer.

Glossary-14

Hashing
A programming technique where a key value is converted to a relative record number
used for random access.
H El.P Message
A message printed by an RMS-11 utility in response to a question mark (?) or the
word "HELP." The message helps you use the utility.
High-Key Value
The key value of the last record in an Indexed file bucket, where:

• The last record in a bucket has an equal or higher key value than any other record
in the bucket.
• The last record in a bucket has a lower key value than the first record in the next
bucket in the chain, neglecting Continuation Buckets if present.
Both data and index buckets have High-Key Values. The key value
used in the Primary Level 0 is the Primary Key value.
High Portion

Those records in an Indexed file bucket with key values higher than the record being
inserted or updated. During a bucket split, the high portion of the data is moved to a
newly created bucket.
Higher Level language

In this manual, BASIC-PLUS-2, DIBOL, PDP-11 COBOL, and RPG II.
Highest Index Level

Same as Root.
Higlllest Possible Key Value

A logical maximum whose form depends on key data type; RMS-11 uses the highest
possible key value to mark the logical end of an index level.
1/0 Buffer

A portion of a task's virtual address space used to store data meant for or arriving
from peripheral devices. See also Control Structure, User Buffer.
1/0 Operation

The process of requesting a transfer of data from a peripheral device to memory, or
vice-versa, the actual transfer of the data, and the processing and overlaying activity
to make both of those happen.

Glossary-15

1/0 Techniques

Programming techniques used to improve the performance of record operations. See
also Deferred Write, Mass illsert.
1/0 Unit

The data moved in and out of a task during an I/0 operation. For disk Sequential
files, the I/0 unit is one or more blocks, depending on the MBC value; and for
Relative and Indexed files, the 1/0 unit is the bucket.
Incremental Reorganization

The process of inserting each data record where it logically belongs in Level O and
updating the upper levels of an index.

Index

The structure for the logical access path to a data record. An Indexed file contains
one index for each key defined for the file. Each index contains index records that
guide RMS-11 through the index levels to the data records in Primary Level 0.
Index Bucket

A bucket in an Indexed file that contains index records. See also Data Bucket.
Index Descriptor

A memory-resident copy of a Key Descriptor's fields that are required for normal
operations.
Index Level

The higher levels (1 +) of an index, containing index buckets. See also Data Level,
Lowest Index Level.
Index Record

A record maintained by RMS-11 in Levels 1 + of all indexes. The record contains a
High-Key Value for a bucket in the next lower level and a pointer to that bucket. See
also Data Record.
Indexed Fiie

An RMS-11 file created with the Indexed file organization.
Indexed Fiie Load

An RMS-11 utility that bypasses normal access methods to load an Indexed file with
records from a file you designate. RMSIFL optimizes the structure of all indexes in
the file it populates. See also RMSCNV.

Glossary-16

Indexed Fiie Organization

The method of organizing records in a file so that the records are sorted in ascending
Primary Key order and one or more indexes are built that provide access path(s) to
those records. See also Sequential File Organization, Relative File Organization.
Indirect Fiie

A file containing a sequence of commands that can be interpreted by a single task,
usually a system task such as a utility. These commands appear in the indirect. file
exactly as they are entered from your terminal. As such, they can be command
strings or answers to questions.
I nflle

A convention for the term "input file specification."
lnltl.allze (Bucket)

When RMS-11 allocates a bucket for a Relative file, during an explicit extend file
operation, for instance, RMS-11 initializes the bucket by setting all bits to zero.
However, RMS-11 does not format Indexed file buckets until it must use the bucket
to store a record.
Input Fiie

The source of records for processing by a utility. Contrast with Output File.

Key

Identifier for a record during random record operations, used especially for Indexed
files. A key is the contents of one or more specific portions of each user data record;
the combination of these portions is called a key field. For each key defined for a file,
RMS-11 constructs an index in the file. See also Segmented Key.
Key Characteristics

Features of individual keys in an Indexed file, including duplicates, changes, and null
key.
Key Data Type

Feature of a key indicating to RMS-11 how to interpret the data in the associated key
field. Supported types include string, two- and four-byte integer and binary, and
packed decimal.
Key Descriptors

Data used by RMS-11 to maintain keys in an Indexed file. RMS-11 stores this data
in the file's Prologue. See also Area Descriptor.

Glossary-17

Key Fleld

Concatenation of one or more portions of a user data record. The key field contains
the value for the associated key. See also Segmented Key.

Key of Reference

A number indicating the index RMS-11 should follow during a random record operation. A key of reference of 0 indicates the Primary index, while a key of reference of 1,
2, and so on, indicates the First Alternat.e index, Second Alternate index, and so on.
Key Value

Value of the data in the field associated with the key.
Least Significant Byte

The last byte examined when RMS-11 performs key value comparisons. Key data
type determines whether the lowest-addressed byte (binary and integer data types) or
highest-addressed byte (string and packed decimal data types) is the least significant. Contrast with Most Significant Byte.
Level

A chain of buckets in an Indexed file. All buckets in a level have the same conceptual
function: Level 0 buckets contain data records, either user data records or SIDRs;
Level 1 buckets contain index records that point to Level 0 buckets and indicate their
High-Key Values; Level 2 buc}sets contain index records that point to Level 1 buckets
and indicate their High-Key Values; and so on to the single-bucket level called the
Root.
Library

In this manual, same as RMS-11 Resident Library.
Locate Mode

A Record Transfer Mode where RMS-11 allows a user program to manipulate data in
the J/0 buffer instead of the user buffer. Contrast with Move Mode.
Loglcal Access Path

Same as Access Path.
Loglcal Block Number

The number the file processor assigns to each logical block on a disk, starting with 0.
See also Logical Structure, Virtual Block.
Logical Structure

A method of making the file processor device-independent. Each disk is considered to
be composed of a logically contiguous series of data units called blocks. The disk
driver translates the Logical Block Number to a physical location.

Glossary-18

Lowest Index Level

The chain of index buckets whose entries point to data buckets. Also called Level 1.
See also Data Level, Index Level.
Mass Insert

RMS--11 I/0 technique valid for Indexed files only. With Mass Insert, during sequential put operations, RMS-11 can extend the Primary Level 0 bucket by bucket,
packing records into the buckets in the order they are written. As each bucket gets
full, RMS-11 creates a new one, beginning with the next record inserted, and notes
its existence in a Primary Level 1 index bucket.
Mass Insert significantly improves file population performance for single-key Indexed
files. The percentage of improvement lessens with each additional key defined in the
file.
See also Deferred Write.
Maximum Key Value

Same as Highest Possible Key Value.
Ma>timum Record Size

An attribute for files containing VFC or variable-length records. If Maximum Record
Size is not zero, RMS-11 checks each record specified in a put or update operation for
size. If the record is longer than the maximum associated with the file, RMS-11
returns the error code ER$RSZ.
The record size for a file of fixed-length records is also stored as the Maximum Record
Size attribute.
Ma>timum Record Number

RMS-11 will not put a record into a Relative file with a relative record number
greater than the assigned Maximum Record Number (MRN)--unless MRN is zero.
In that case, RMS-11 makes no check on relative record numbers during put operations. MRN is an attribute of Relative files.

'-""'

MBC

Same as Multi-Block Count.
MBF

Same as Multiple Buffers.
Medium

A file storage device. Relative and Indexed files can be stored only on disks, while
Sequential files can be accessed on disks, magnetic tapes, and unit record devices.

Glossary-19

Memory-Resident Overlay

An overlay structure or segment maintained in physical memory. The overlay is
executed by the task via Active Page Registers. See also Disk-Resident Overlay.
Monitor

In RSTS/E, the part of the operating system that coordinates and controls the other
components. See also Executive.
Most Significant Byte

The first byte examined when RMS-11 performs key value comparisons. Key data
type determines whether the lowest-addressed byte (string and packed decimal data
types) or highest-addressed byte (binary and integer data types) is the most significant. Contrast with Least Significant Byte.
Move Mode

A Record Transfer Mode where RMS-11 always moves a record between the I/O
buffer and the user buffer. The user program can manipulate the data only in the
user buffer. Contrast with Locate Mode.
MRN

Same as Maximum Record Number.
Multi-Area File

An Indexed file with more than one area. See also Single-Area File.
Multi-Block Count

The number of blocks that RMS-11 moves in and out of the 1/0 buffer during each
I/0 operation for a Sequential file. See also I/0 unit.
Multiple Buffers

An I/0 technique where you assign to a Relative or Indexed file more than the
minimum number of I/0 buffers required by RMS-11. If the file is not being writeshared, RMS-11 uses the extra buffers to cache buckets in memory.
Multiple Record Access Streams

RMS-11 allows each program to use from one to 255 streams on a Relative or Indexed
file.
Next Record

Part of the context of a Record Access Stream. The Next Record is the target of a
sequential get not immediately preceded by a find, a find, or a put operation (put
operations on Sequential and Relative files only). See also Current Record.

Glossary-20

Nonoverlaid

Pertaining to a task that does not include overlay segments. All of a nonoverlaid task
must fit in the available virtual address space (64KB) and is brought into memory
when the task is run. Contrast with Overlay.
Null Key (Characteristic)

A flag associated with an Alternate Key. The flag indicates whether you defined a
null key value for the key. For binary, integer, and packed decimal key data types,
the null key value is zero expressed in that data type. For the string key data type,
you must specify a byte value from 000(8) through 377(8) including ASCII codes.
During a put or update operation, RMS-11 makes no entry for a record in an Alternate index if:
• The null key flag is set, meaning that a null key value has been defined for the
associated Alternate Key field.
• If the associated key field is filled with the defined null key value.

See also Change Key, Duplicate Key.
Ob,iect Code

An intermediate form of executable code where external references are not resolved.
Assemblers and compilers normally produce modules of object code. Each module
consists of relocatable machine code, relocation information, and a symbol table
listing the use and definition of symbols within the module.
The Task Builder reads object modules, resolving such references between them and
producing an executable task.
Ob,iect Module Library

A method of storing object modules so they can be used easily by the Task Builder.
Specifically, the RMS-11 object module library named RMSLIB.OLB.
ODL

1. Same as Overlay Description Language. 2.
See ODL File.

File containing ODL statements.

ODL File

A file containing ODL statements that describes all or part of the overlay structure
for a task. ODL files are control files for the Task Builder.
,ODS-1

Same as FILES-11 ODS-1.
ODS-2

Same as FILES-11 ODS-2.

Glossary-21

Open (File Operation)
RMS-11 initiates access to a file; also occurs as last step in the create file operation.
Contrast with Close.

Operating System
The collection of tasks that administer the computer system by scheduling and
controlling user and system tasks, performing VO and various utility functions, and
allocating resources for efficient use of the hardware. See also File Processor, Monitor, Executive, Device Driver.
Outfile

A convention for the term "output file specification."
Output File
The destination for records processed by a utility. Contrast with Input File.
Overlay
The technique of repeatedly using the same area of memory during different stages of
a program. When one routine is no longer needed in memory, another routine can
replace all or part of it.
Overlay Description Language

A set of statements and syntax that the Task Builder can interpret to build the
overlay structure for a task.
Overlay Segment

A section of a task treated as a unit that can overlay and be overlaid by other overlay
segments.
Overlay Structure
A system of overlays defined by one or more ODL files. The structure consists of a
root segment and optionally, one or more overlay segments.
Pack Cluster

See Device Cluster
Packed Decimal (Key)

An Indexed file key data type where a half-byte represents a decimal digit. See also
Two-Byte Signed Integer, Four-Byte Signed Integer, String, Two-Byte Unsigned
Binary, Four-Byte Unsigned Binary.

Glossary-22

Paddling

Characters added to data to achieve a required length. In disk Sequential files,
RMS-11 adds a null byte (all bits 0) to user data records that are an odd number of
bytes long. Additionally, when you do not allow records to span disk blocks, RMS-11
writes null bytes in the gap between the last record in a block and the end of a block.
However, on tape, RMS-11 uses circumflexes (A) to pad blocks. See also Word
Aligned.
P atclh Level

A number specifying the number of times an RMS-11 utility has been patched.
Physical Block

Same as Sector.
Physical 1/0 Operation

The actual transfer of data between memory and disk, involving the positioning of
the disk heads and the electrical flow of data. See also I/0 Operation.
Placiement Control

The ability to specify a Logical Block, Device Cluster, or Virtual Block Number
where (or near where, for VBN) the file processor should begin an area. On
VAX/VMS, you can also specify a related file as a guide to placement.
~·

Platter

One of the circular pieces of metal covered with a ferromagnetic substance that make
up a disk. Data is stored on platters. See also Bit, Byte, Track, Cylinder, Sector.
Pointer Array

Same as Duplicate Pointer Array.
Population

In this manual, the process of inserting a large burst of records into a file after it has
been created and before it is made available for normal processing. You can populate
a file with RMSIFL, RMSCNV, or an application task.
Position of Key in Record

The number of the byte(s) where segment(s) of the key begin. Numbering begins with
O; therefore, the first byte in a record is referred to as byte 0. See also Key Field,
Segmented Key.
Primary Index

The structure providing the logical access paths for the Primary Key. See also Alternate Index.

Glossary-23

Primary Key

The portion of each user data record whose value determines the position of that
record within Level 0 of the Primary index. RMS-11 constructs the Primary index
from the Primary Key values of records inserted into the file. You must define a
Primary Key for each Indexed file. Contrast with Alternate Key. See also Segmented
Key.
Primary ODL File

The ODL file referenced by the indirect file you supply to the Task Builder. You can
also supply the name of the primary ODL file directly in a Task Builder command
line. See also Secondary ODL File.
Primary Root

The single bucket that is the entry point for processes following the Primary index.
See also Root.

Program

1. The source file that must be processed by a compiler or assembler and the Task
Builder before it can be run. Contrast with Task. 2. That part of a task written by a
user, as opposed to the RMS-11 routines in another part of the task.
Prologue

The first blocks of a Relative or Indexed file where RMS-11 maintains file attributes
that cannot fit in the file directory.
Prototype ODL File

A heavily commented version of the ODL file for the RMS-11 9KB overlay structure.
You can modify this prototype to select only those RMS-11 functions required by
your program and to optimize the overlay structure. See also Standard ODL File.
Pyramid

Figurative way of representing an index.
Query Mode

Same as Extended Diagnostic Messages.
Random Access Mode

A mode of record access where you, rather than the organization of the file, establish
the order in which records are processed. You must specify a record identifier with
each random access. Each record access is independent of the previous record used.
Successive operations in the Random Access Mode can identify and access records
anywhere in the file. Supported for Relative and Indexed files only. See also Sequential Access Mode, Access by Record's File Address.

Glossary-24

Read Ac:cess

To open a file to perform find and/or get operations only. Contrast with Write Access.
Read-After-Write

An optional data integrity check performed by the RMSBCK and RMSRST utilities.
See also Check-After-Write.
Read Sharing

More than one task opens a file for read access and allows other tasks to perform
read-type operations.
Read-T~rpe (Record Operation)

F'ind or get. See also Write-Type.
Record

A discrete group of data items whose form is repeated throughout the data. Each
group represents an entity, and a file consists of a series of such entities. A record is
normally the unit of data interchange between a user program and an RMS-11 file.
Record Access Mode

The method used to locate the position of a record within a file for purposes of
inserting a :record there or retrieving the record already there. See also Access By
Hecord's File Address, Random Access Mode, Sequential Access Mode.
Record Access Stream

A channel between your program and a file through which record operations pass.
Record Format

The shape or form by which RMS-11 recognizes and processes each record in a file.
Format dictates how RMS-11 determines a record's size. See also Fixed-Length,
Stream, Variable-Length, Variable-with-Fixed-Control, Undefined.
Record Format Overhead

The control bytes added to your data that RMS-11 requires to support a record
format. The overhead is zero for fixed-length records and equals two for variablelength and VFC records. See also Bucket Header, Record Header.
Record Header

Seven bytes of control data that RMS-11 appends to every user data record in an
Indexed file. This header is converted to an RRV and left in the record's place when
the record is moved from its original bucket during a bucket splitting operation; the
record assumes a new header in its new location. See also Bucket Header, Record
Format Overhead.

Glossary-25

Record-Length Field

A unit of data appended to every VFC or variable-length record, regardless of file
organization. The field contains the length of the record to which it is attached.

Record Operation

A processing technique involving a Record Access Stream, including:
Connect
Delete
Disconnect
Find
Flush
Get
Put
Rewind
Truncate
Update
Re co rd Processing

Processing using records as the logical units of data interchange. Contrast with Block

I/0.
Record Size

The number of bytes in a record. All fixed-length records have the same size, which is
stored as the Maximum Record Size file attribute. The size of variable-length and
VFC records is indicated by the attached record-length fields. Files containing variable-length and VFC records can have a Maximum Record Size as an attribute.
Record Storage Cell

A fixed-length logical division of a Relative file. Records are stored in cells. However,
not all cells must contain records.
~
-="!

Record Transfer Mode

A method of performing record operations indicating whether the user program can
manipulate record data in the I/0 buffer or in the user buffer. See also Locate Mode,
Move Mode.
Record's File Address

A unique identifier for every record within a disk file. This Record's File Address
(RFA) remains valid for that record alone for the life of the file. If a record is deleted,
its RFA is not reused. See also Access by Record's File Address.
Record's Reference Vector

A copy of a record header that is left in the record's original position when the record
is moved during a bucket splitting operation. The RRV preserves Access by Record's
File Address and facilitates Alternate Key access.

Glossary-26

Relative File

An RMS-11 file created with the Relative file organization.
Relative File Organization

A method of organizing records in a file so that each record is stored in a fixed-length
cell and can be accessed randomly. See also Sequential File Organization, Indexed
File Organization.
Rela1tive Record Number

Each record in a Relative file is stored in a cell. Each cell can be addressed randomly
by a number relative to its distance from the beginning of the file. This number is
called the relative record number. For instance, the sixteenth cell has relative record
number 16. The cell's number is associated with the record in the cell.
Resident Library

A set of memo1y-resident routines which can be shared by multiple tasks, but are
part of none of them. When a task uses a routine in the Library, the operating system
maps the Library segment containing the routine into the task's address space with
Active Page Registers.
Restore

1. The process of copying back-up data to recover the contents of a file.
2.

RMSRST utility.

Retrieval Pointer

Data associated with a file that specifies blocks on disk. From the structure and
content of a retrieval pointer, the file processor can equate virtual blocks to logical
blocks. See also Window.

Rewind (Record Operation)

RMS-11 resets the context of a stream to the logical beginning of the file.
RFA

Same as Record's File Address.

RMSBCK
The RMS-11 back-up utility. See also RMSRST.

RM SC NV
The RMS-11 utility that converts one RMS-11 file into another. See also RMSIFL.

Glossary-27

RMSDEF

The RMS-11 utility that helps you define an RMS-11 file during an interactive
process and then creates that file.

RMSDSP

The RMS-11 utility that prints the attributes and structural data of any RMS-11
file.
RMSIFL

An RMS-11 utility that loads records into an RMS-11 Indexed file. RMSIFL bypasses normal RMS-11 record processing and optimizes the structure of all indexes. See
also RMSCNV.
RMSRST

The RMS-11 utility that restores data backed up by RMSBCK.
Root

The highest level in an index. The Root is a single-bucket entry point to the index for
random accesses using the associated key.
Root Caching

An I/O technique. You supply virtual address space for more than the two I/O buffers
required by RMS-11 for an Indexed file. If the file is not being write shared, RMS-11
uses the extra buffers to cache the Root buckets of indexes it uses for random access
operations. See also Multiple Buffers.
RRV

Same as Record Reference Vector.
Secondary Index Data Record

Records occupying Level 0 of each Alternate index. SIDRs contain an Alternate Key
value and one or more pointers to user data records in the Primary Level 0 that
contain the key value.
Secondary Key

Same as Alternate Key.
Secondary ODL File

An ODL file indirectly referenced by the primary ODL file. Especially, the RMS-11
standard ODL files and your modified version of the RMS-11 prototype ODL file.

Glossary-28

A physical division of a disk, containing 512 bytes on most disks supplied by
DIGITAL. See also Bit, Byte, Track, Cylinder.
Segmented Key

A Primary or Alternate string key that consists of separate sections, or segments, in
different parts of the record. RMS-11 concatenates the specified segments before it
performs any operations that involve key value comparisons. The concatenated version of the key is also stored in index records and SIDRs (for Alternate Keys).
Sequential Access Mode

A mode of record access where the organization of the file establishes the order where
records are processed. Each record access depends on the previous record used. Successive operations in the Sequential Access Mode access records in their logical order
according to the file organization. Supported for all file organizations. See also Access
by Record's File Address, Random Access Mode.
Sequential File

An RMS--11 file created with the Sequential file organization.
Sequential File Organization

A method of organizing records in a file by their order of insertion. The records are
virtually contiguous. See also Relative File Organization, Indexed File Organization.
Sequential Write Operations

Put operations in Sequential Access Mode must be performed within file organization
restrictions.
Shared Access

More than one task, File Access Block, or Record Access Stream maintain simultaneous
access paths to the same file. See also Exclusive Access to a File.
Sharing Specifications In Accessing Programs

When a program opens a file, it must declare the record operations it intends to
perform on the file and the type of record operations it will allow other programs to
perform on the file.
SIDR

Same as Secondary Index Data Record.
Signed Integer (Key)

See Two-Byte Signed Integer and Four-Byte Signed Integer. See also Packed Decimal, String, Two-Byte Unsigned Binary, Four-Byte Unsigned Binary.

Glossary-29

Single-Area File

An Indexed file with one area, whether by default or definition. Sequential and
Relative files also consist of one area for purposes of Placement Control. See also
Multi-Area File.

Standard File Structure and Interface

RMS-11 operates on most PDP-11 operating systems as well as VAX/VMS. The files
it creates are identical and can be used on any system within the confines of on-disk
structures. The RMS-11 interface with user programs is also standard and identical
on all systems (subject to system-specific limitations). VAX/RMS uses the same disk
structure for files as RMS-11 and an interface that parallels RMS-11 's.
Standard ODL File

Secondary ODL files provided by DIGITAL to define 8KB, 9KB, and 12KB RMS-11
overlay structures. See also Prototype ODL File.
Stream

Same as Record Access Stream.
Stream (Record Format)

Each record in the file consists of a series of contiguous characters. RMS-11 detects
the end of a stream record only by the presence of specific terminators. See also
Fixed-Length, Variable-Length, Variable-with-Fixed-Control, Undefined.
String (Key)

An Indexed file key data type where each byte is interpreted by its binary contents.
Permissible values are not limited to valid ASCII codes. See also Packed Decimal,
Two-Byte Signed Integer, Four-Byte Signed Integer, Two-Byte Unsigned Binary,
Four-Byte Unsigned Binary.
Summary Listing

A series of messages that serve as an audit trail of the processing performed by an
RMS-11 utility. Normally, a summary listing can be printed on your terminal or
written into a file you specify.
Switch

A mnemonic, preceded by a slash (/), that qualifies the processing requested by a
command string you supply to a utility.
Synchronous Operations

Any operation where RMS-11 returns control to your program only after all processing associated with the operation is finished. File operations are always synchronous.
Record operations can be asynchronous.

~
i

Glossary-30

System Protection (Code)

A file attribute that dictates how the operating system restricts access to a file based
on account number and type of operation.
Targ1et Bucket

The data bucket where RMS-11 determines it should perform the specified operation.

The executable version of a program, especially while running in memory. See also
Task Image.
Tas~t Builder

A system utility that converts and combines object modules into a task image. The
Task Builder also arranges task segments according to an overlay structure defined in
an ODL file.
Tasl< Image

The disk file containing the executable code that comprises the task. When you use
RMS-11, a task image is the product of the Task Builder utility. When a task image
is executed, it is called a task.
Trac:k

The sectors at a single radius on one disk platter. See also Bit, Byte, Sector, Cylinder.
Truncate (Record Operation)

For Sequential files only, RMS-11 logically deletes the Current Record and all records following it in the file by establishing a new end-of-file position at the first byte
of the Current Record.
·~

Two-Byte Signed Integer (Key)

A key data type that can represent the decimal integer values -32,768 through
+32,767. See also Four-Byte Signed Integer, Two-Byte Unsigned Binary, Four-Byte
Unsigned Binary.
Two-Byte Unsogned Binary (Key)

A key data type that can represent the decimal integer values 0 through +65,535.
See also Four-Byte Unsigned Binary, Two-Byte Signed Integer, Four-Byte Signed
Integer.
Undefined (Record Format)

No records are defined for the file. RMS-11 reads only blocks, and your program
must interpret the contents of each block. Used with Block 1/0 only. See also FixedLength, Stream, Variable-Length, Variable-with-Fixed-Control.

Glossary-31

Unit Record Device

A peripheral device capable of handling one record at a time. Terminals and line
printers are unit record devices. See also Medium, Forms Control.

·~

Unsigned Binary

See Two-Byte Unsigned Binary and Four-Byte Unsigned Binary. See also Packed
Decimal, Two-Byte Signed Integer, Four-Byte Signed Integer, String.
Update (Record Operation)

RMS-11 replaces a record in the I/0 buffer with a revised version from the user
buffer.
User Buffer

Memory allocated within your program's portion of a task to store one record. This
buffer is available to your program and the data in it can be manipulated. RMS-11
can also access this buffer.
User Data Record

The record your program provides in its user buffer plus RMS-11 overhead: the
logical unit of data actually stored in a file. See also Index Record.
Utilities

Tasks, provided by DIGITAL, that use RMS-11 to accomplish standard file-related
jobs. See also Application Program, Operating System.
Value Match

A criterion used by RMS-11 during a random read-type operation on an Indexed file.
Your program must specify whether the key value in a record must be greater than or
equal to (or either) the value specified when the operation was initiated.
Variable-Length (Record Format)

The records in the file can be any length, up to the Maximum Record Size associated
with the file. For each record, RMS-11 maintains a record-length field specifying the
number of data bytes in the record. See also Fixed-Length, Stream, Variable-withFixed-Control, Undefined.
Variable-with-Fixed-Control (Record Format)

Each record in the file consists of a fixed control area whose length is the same for all
records and a variable area that can vary from zero to the Maximum Record Size
associated with the file. See also Fixed-Length, Stream, Variable-Length, Undefined.
Virtual Block (Number)

The file processor treats each file as a device containing virtually contiguous blocks,
numbered serially from 1. See also Logical Block.

Glossary-32

Vir1tual-to-Logical-Block Mapping

The file processor translates the Virtual Block Number supplied by RMS-11 to the
Logical Block Number the processor must provide to the device driver during a disk
access operation. See also Retrieval Pointer, Window.
VFC

Same as Variable-with-Fixed Control.
Window

The set of retrieval pointers the operating system maintains in memory for each open
file for mapping Virtual Block Numbers to Logical Block Numbers.
Window Turning

The process of changing the retrieval pointers in memory until the window contains
pointers covering the specified virtual blocks.
Weird Aligned

Each record in a Sequential file starts and ends on a word boundary; that is, each
record is stored as an even number of bytes. RMS-11 uses this convention to maintain structural compatibility with FCS-11 sequential files.
Wriite Access

To open a file to perform put, update, or delete operations, as well as get or find
operations.
Wrlite Sharing

More than one task opens a file for write access and allows other tasks to perform
write-type operations.
Wriite-Type (Record Operation)

Delete, put, or update. See also Read-Type.

XBUF
Same as Extended Buffer Pool.

Glossary-33

Index
• Primary discussion of an item is indicated by the page number in bold.
• Page number followed by a 't' indicates that the item appears in a table.
• Page number followed by an 'f' indicates that the item appears in a figure.
A
@, 9--6

Access, 1-2, 1--4. See also
Indexed, Relative, and Sequential file
controlled, 1-2
declaration, 2-7. See also file-sharing
exclusive, 2--9. See also file-sharing
files, to critical, 1-36
modes. See Record Access Modes
path, 1-19, 6-3
randomly, 1--5
records in unvarying order, 1-6
routines, 1-28
sequentially, 1--4
shared, 2-5, 2-7. See also file-sharing
speed, 3-3
time, 5-1, 5--13
Access by RF A, 1-19, l-·23, 1-23, 1-25, 4-9
find operation, 3-8, 4-8, 7-2
get operation, 3-9, 4-10, 7-4
Account and/or volume, choice of restoration,
9-67
Active Page Hegister, 8-3f, 8-4
Activity overhead, 6-18
Address space, G-16, 8-2
Allocation, 1--35, 1-39, 2-1, 2-3, 3-4,
4-4, 4-5, 6-24
and areas, 6-10
calculation, 3--4, 4-4, 6-24
contiguity, 3-6, 4-6, 6-28
initialize Relative file buckets, 4-4
more efficient when create file, 3-4, 4-4
Primary Key, ascending order of, 6-24
putting Maximum Record into file, 4-5
quantity, 3--3, 4-3, D-2
quantity rounded up to bucket size, 4-4
in RMS DEF, 9-43
value, minimum, 6-26
window tutning optimization, 8-29
Alphanumeric character, 1-12
ALq, 3-4, 4-4
Alternate index, 5-11, fi-14, 6-7, 9-20.
See also Alternate Key
bucket size for, 6-21
bucket splitting, 5-13

Alternate index, (cont.)
built by RMSIFL, 9-60
optimization, 9-57
and update operation, 5-15
Alternate Key, 1-9, 1-18, 5-2, 5-11,
5-16, 6-3, 6-9, 9-26, 9-59, D-1.
See also Primary Key
access preserved by RRV, 5-12
and ANSI COBOL, 5-1
bucket size calculations, 6-21
buckets loaded inefficiently, 6-22
cost and application needs, 6-3
definition by RMSDEF, 9-38
definition by RMSDFN, D-3
duplicate values, 1-38
duplicates allowed, 6-8
duplicates allowed during put operation, 5-11
null key, 6-10
optimal at end of record, 6--7
in Prologue, 6-26
record operation, cost in, 5-18t
record placement, no affect on, 1-19
Secondary Index Data Record, 5-6
sequential access, performance on, 5-1
value change during update operation,
1-38, 5-15, 6-9
value in SIDR, 5-6
American National Standards D format, F-1.
See also ANSI, magnetic tape
American Standard Code for Information
Interchange, 1-12. See also ASCII
ANSI
COBOL, 5-1
labels and RMSDSP, 9-53
utilities compliance with, 9-10
X3.27-1969, F-1
Application, 6-3
design, 2-1, 8-1
environment affects VO time, 2-3
files on disk, 6-14
program, 1-11, 2-1, 2-14, 8-24. See
also program
program used for population, 6-30
record size, 6-2
Approximate match. See match criteria

Index-I

Area, 1-35, 1-38, 2-14, 5-2, 5-3~ 5-4~
6-10, 6-llf, 6-12f
0, 6-lfi
1+, 6-15
and allocation, 6-10
and bucket size, 6-10, 6-16, 6-19
buckets, allocation of, 5-2
contents, 1-39
contiguity, 6-28
file creation, 6-13
file extension, 6-10
fill number, 1-39
1/0 time, primarily saves, 6-11
Level 0, for, 6-10
Level 1, for, 6-10
Levels 2 +, for, 6-10
multiple and contiguity, 6-13
numbering, 6-13
placement control, 6-10, 6-28
in RMSDEF, 9-41, 9-45, 9-46
Virtual Block Numbers, continuous range of,
6-10
window turning optimization, 8-28, 8-29
Area Descriptor, 1-36, 5-2, 5-12, 5-18t, 6-26
Array, pointer, 5-6, 5-11, 6-9
ASCII, 1-12, 2-16, 2-17, 6-4, 6-10
ASCII file, stream, A-3
Assembler, 1-28, 8-1
Assessment routine, damage, 9-8. See
also utilities
AST, 1--33, A-2, A-4, A-5
Asynchronous record operation, 1-33
Asynchronous System Trap, 1-33, A-2, A-4, A-5
At sign, 9-6
Attributes. See file attributes

B
Back up
data, 1-10, 1-26, 9-10. See also RMSBCK
date, 9-73
file, 9-10, 9-12, 9-16, 9-67, 9-69,
9-72, 9-73. See also RMSRST
format, 1-26
medium, 9-12, 9-13
BASIC-PLUS, A-3
Record 1/0, 2-1
BASIC-PLUS-2, 2-2, 3-5, 4-5, 4-7, 4-16,
6-29, 6-32, 7-9, B-1. See also
higher level language
ACCESS APPEND clause in OPEN
statement, 3-7
BUCKETSIZE clause in OPEN statement,
4<3, 6-23

Index-2

CONTIGUOUS clause in OPEN statement,
3-6, 4-6, 6-28
FILESIZE clause in OPEN statement, 3-4,
4-4, 6-28
FREE statement, 2-13
SEQUENTIAL and BUCKETSIZE clauses in
OPEN statement, 3-15
UNLOCK statement, 2-13
WINDOWSIZE keyword, 8-28
Binary key, 6-3. See also key
Binary search technique, 5-10
Bit, 1-12
Bit map, disk free-block, 3-5
BKS, 4-4, 4-6, 6-17
Block, 1-14
count, 1-35. See also Multi-Block Count
file reused, 1-37
1/0, 1-40
magnetic tape size, 9-23
records span boundaries, 1-34f, 1-34.
See also block spanning
unused, 9-19. See also RMSBCK
Block Number
Logical. See Logical Block Number
Virtual. See Virtual Block Number
Block spanning, 1-34, 1-34f, 1-38,
2-18t, 3-1, 3-2, 4-1, 9-37
Bootable volume
and RMSBCK, 9-18
and RMSRST, 9-75
BPL, 6-17
Bucket, 1-34, 5-15, 5-17
in Alternate indexes, loaded inefficiently,
6-22
blocks as a unit, 1-34
cache, 2-19
chains called levels, 5-4
Continuation, 5-6, 5-18t, 6-21
creates another, 5-12
data in Level 0, 5-5
during Deferred Write, 4-15, 7-8
delete operation, use in, 7-2
file cluster, align with, 5-2
file, fixed within, 1-34
file, pieces of, 5-9
during find operation, 4-9
formatting, 1-34, 2-5, 5-4, 5-4f
during get operation, 4-11
header, 2-5
high portion, 5-12
and 1/0 buffer, 4-16, 7-9
and 1/0 operation, 1-34, 5-9
1/0 unit, as, 1-34, 4-2
index, 5-6

·~·

Bucket, (cont.)
Indexed files, unit of access for, 6-16
Level 0 properties, 5-5
and Multiple Buffers, 4-16, 7-9. See
also Multiple Buffers
overhead, 6-2
pointer, 5-6, 6-17
during put operation, 4-12
record size, do not adjust to fit, 6-2
records cannot cross boundaries, 1-34, 6-2
records reorganized in, 5-12
Relative file allocation, unit of, 4-1
Relative files, initialized in, 4-1
Root, 2-19, 5-5
size. See bucket size
speed, critical to, 6-16
split, 6-2, 6-3. See also bucket splitting
target data, 5-10, 5-12
during update operation, 7-6
virtual address space, critical to, 6-16
Bucket locking 2-9, 2-lOf, 2-llf, 2-12.
See also file-sharing
activated, 2-9
cost, 2-12
and program, 2-9, 2--12
release explicitly, 2-13
unlocking, 2-9, 2-12
Bucket size, 1-38, 2-3, 2-14, 4-2, 4-4,
4-6, 6-16, D-2
and activity overhead, 6-16
allocation quantity rounded up to, 4-4
Alternate indexes, calculations for, 6-21
and areas, 6-10
blocks, whole number of, 6-22
and data record size, 6-16
default value, 6-23
different, advantage of, 6-19
and file clustersize, 6-17
and file population, 6-16
and higher level languages, 6-22
and 1/0 buffers, 6-16
maximum, 1--35, 6-16
pointer length, 6-21
power of two, 5-2
Primary index, calculation for, 6-17
program syntax, 6-22
in RMSDEF, 9-44
selection, 6-18
Sequential access, affects, 4-3
space, affects, 4-2
speed, affects, 4-2
task size, affects, 4-3
Bucket splitting, 5-12. See also bucket
costs, 5-13
1

Bucket splitting, (cont.)
frequency increases with packing efficiency
factor, 6-30
in index, 5-13
during put operation, 5-12
routines, 8-8
Buffer, 3-12, 8-3f. See also l/0
buffer, user buffer
size, 2-5
Byte, 1-12
-aligned record, 4-1, 5-4
offset, 3-2

c
Cache, 2-19, 4-16, 5-8, 7-9
Caching
data, 8-30
directory, 8-28
Carriage control, 1-39
in RMSDEF, 9-37
Carriage return, 1-39, 2-17
line feed, with, 2-16
Cell, 1-17. See also record storage cell
Changes, 6-7. See also key characteristics
during update operation, 5-15, 5-16
Character, alphanumeric, 1-12
Check byte, 5-4f
Checkpointing, 8-30
Circumflex, 3-2, F-3
Clustersize. See file clustersize
Co-tree, 8-6
COBOL. See PDP-11 COBOL
Command interpreter, 9-12, 9-21, 9-32,
9-52, 9-61, 9-68
Command string. See individual utility
continuation, 9-7
Compiler, 1-28, 8-1
Completion code, B-4
Concatenate, 8-2
Concatenated factor or module, C-2
Connect record operation, 1-26, 1-31,
3-7, 4-8, 7-1
and Current Record, 3-7, 4-8, 7-1
and Next Record, 3-7, 4-8, 7-2
records available to program, make, 1-26
stream to a file, more than one, 1-32
Container file, 9-13
list contents, 9-49, 9-54
multiple, 9-13
rewind, 9-16, 9-72
in RMSBCK, 9-13
in RMSDSP, 9-53
in RMSRST, 9-69
select files from in RMSRST, 9-73

Index-3

Contents-addressable record access, 5-1
Context. See Record Access Stream
Continuation Bucket, 5-6, 5-18t, 6-21
Control blocks, 2-2
Control flags, 5-4f
Control structure, 1-30
Controlled access, 1-2
Convention
documentation, 9-9
module names, RMS-11, 8-14
utilities, 9-4
Cost
bucket splitting, 5-13
clustersize on RSTS/E, 8-29
contiguity on RSTS/E, 8-29
I/Os in record operation, 5-17
key characteristic, change, 6-10
key characteristic, duplicate, 6-8
key characteristic, null, 6-10
key size, 6-3
key, four-byte signed integer, 6-5
key, four-byte unsigned binary, 6-6
key, number of, 6-2
key, packed decimal, 6-6
key, position in record, 6-7
key, string, 6-4
key, two-byte signed integer, 6-5
key, two-byte unsigned binary, 6-5
CR, 9--10
Crash dump, 9-8. See also utilities
Creation date. See file, date
Cross block boundaries, 3-1. See also
block spanning
(CTRL/CJ, 9-10
CTRL/Z, 2-16, 2-17, 3-2
terminate utility, 9-12, 9-21, 9-32,
9-52, 9-61, 9-68
(CTRL/ Z), 9-10
Current keyboard monitor, 9-12, 9-21,
9-32, 9-52, 9-61, 9-68
Current Record, 1-31, 3-7, 4-8, 5-16, 7-1
connect operation, affected by, 3-7, 4-8, 7-1
find operation, affected by, 3-8, 4-9, 7-3
find or get operation, established by, 1-31
get operation, affected by, 3-10, 4-11, 7-5
get operation, use in, 3-9, 4-10
put operation, affected by, 3-11, 4-13, 7-5
record operation failure, affect of, 3-7, 4-8, 7-1
record operation, cost in, 5-18t
record operation, target of, 1-31
and truncate operation, 3-11
and update operation, 3-12, 4-13, 5-15, 7-6
CYL, in RMSDEF, 9-43

Index-4

Cylinder, 1-12, 1-35, 6-14
disk heads can access without changing
position, 1-12
in RMSDEF, 9-43
tracks at a single radius on all platters, 1-12
Cylinder/track/sector, 1-16. See also
physical block
D

Damage assessment routine, 9-8. See
also utilities
Data, 2-1, 5-2. See also application
back up, 9-10. See also RMSBCK
buckets, 5-2, 5-5
duplicate of, 1-10. See also back up
form repeated, 1-3. See also record
in index, 5-5
integrity, 2-9. See also file-sharing
integrity checks, 9-11, 9-68
numeric, 1-12
organization, 1-10
restricted, 1-2
security, 9-3
segregation, 1-1, 1-2
size, 3-3t, 4-2t, 6-1
storage devices, mechanical, 5-8
storage space, 2-5. See also space
structure, 1-11, 1-12
transfer between disk and memory, physical,
1-33
valuable, 1-10. See also back up
written to disk, ensure all is, 1-26.
See also Flush record operation
Data base key concept, 5-1
Data bucket, 5-5. See also bucket
Data caching, 8-30
Data processing, 1-11, 1-25
Data record, 5-12, 5-18t, 6-3
and bucket size, 6-17
file, number in a, 6-18
pointer to, 5-5
Decimal radix, 9-29
Declaration. See file-sharing
access, 2-7
allow, 2-7
Default Extension Quantity, 1-37, 1-39,
2-3, 3-3, 3-·5, 4-3, 4-5, 6-26, 6-29
bucket size, multiple of, 6-29
calculation, basis for, 3-5, 4-5, 6-29
in RMSDEF, 9-44, 9-45
in RMSIFL, 9-59
run-time value, 6-29
zero, effect when, 3-5, 4-5, 6-29
Defaults, 2-2

Deferred Write, 2-17, 4-8, 4-12, 5-1
benefit, little or no, 4-16, 7-9
delete operation, effect in, 7-2
file-sharing and, 4-16, 7-8
Indexed files and, 7-8
Mass Insert, used with, 6-33
performance, improves, 4-16, 7-8
and Relative files, 4-15
RMSCNV, used by, 9-28
run time, specifying at, 4-16, 7-9
and Sequential files, 2-17, 3-14
update operation, effect in, 7-6
Delete record operation, 1-25, 4-8, 7-2
Alternate Keys, cost of, 6-3
cost, 5-18t
and Current Record, 1-31, 4-8, 7-2
Deferred Write, affected by, 7-2
and deleted records, 5-17, 6-7
duplicate key values, effect of, 6-9
find or get operation, preceded by successful,
1-25, 5-16
I/0 buffer in, 1--30
index structure, does not modify, 5-17
Indexed files, effect on, 5-16
and Next Record, 4-8, 7-2
Primary Key, position of, 6-7
record header, flag, 7-2
record no longer exists, 1-25
Sequential files, illegal on, 1-26.
See also Truncate record operation
space reused, 1-25, 1-30
Deletion overhead, 6-18
Density, optimal of records, 3-2
Depth, index, 5-fi, 6-18. See also index
DEQ. See Default Extension Quantity
Design, 2-3
application, 2-1
consider yourself and program's maintainer,
2-13
considerations, 8-1
ease of, 2-13
employing com;iderations, 2-14
examples of bad, 2-1
file organization, selection of, 2-14
files off-line, 1-40
keep things simple, 2-· 13
optimizations, applying, 2-14
performance, if you 're not happy with, 2-2
process, 2-14
Relative files, 4-2. See also file design
Sequential files, 3-3. See also file design
space requirements, reduce, 2-4
suiecess, important to, 2-14
task, 3-6, 4-7, 7-1. See also task design

Design, (cont.)
when to, 2-2
before you write the application, 2-2
- Device. See hardware
data storage, 5-8
directory, 1-37
driver, 1-12
head, 3-6, 4-6
sort work, reassign, 9-65
Device Cluster Number, 6-15
RMSDEF, 9-42, 9-43
Diagnostic message, 9-14, 9-71
Diagnostic messages, extended, 9-11, 9-67.
See also Query mode
DIBOL, 2-13, 3-4, 3-6, 3-7, 4-3, 4-5, 4-6,
4-7, 4-16, 6-23, 6-28, 6-29, 6-32, 7-9,
B-1. See also higher level language
PROC statement, 3-15
DIGITAL Command Language, 9-1
Directory
caching, 8-28
device, 1-37
Disconnect record operation, 1-26
context, affect on, 3-8, 4-8, 7-2
stream of operations, terminate, 1-26
Disk, 1-11, 1-17, 2-18t, 3-2, 3-6,
4-6, 6-14, 8-4, 9-13, 9-69
bit map, free-block, 3-5
drive, 6-1~ 8-30, 8-31
driver, 1-16, 1-33
head movement, 3-6, 4-6, 6-11, 6-14
initialize volume, 8-27
memory and, 1-33
mount volume, 8-27
performance, impact on, 8-30
prime storage device, 1-11
usage, 2-1, 8-29, 8-30
Disk-resident overlays, 8-3f, 8-4,
8-24, 8-25, C-1
structure, 1-28
structure illustrations, C-1
Display, 1-10, 1-27. See also RMSD SP
DO, 6-18
Driver
device, 1-12
disk, 1-12
DSC, 8-31
Duplicate
factor, 6-21
key values, 5-6
overhead, 6-22
pointer array, 5-11, 5-18t, 6-9
Duplicates, 6-7. See also key
characteristics

Index-5

End-of-file
attribute, 3-2
indicators, 3-2t
Sequential files, 3-2, 3-11
truncate operation, 3-11
Environment
asynchronous, A-2, A-4, A-5
synchronous, A-2, A-4, A-5
Equal match. See match criteria
ER$ANI, F-3
ER$CUR, 3-11, 3-12, 7-6
ER$DEL, 4-9, 4-11, 7-2, 7-4
ER$EOF, 3-8, 3-10, 3-11, 4-9, 4-11, 7-2, 7-4
ER$IOP, 3-11, 3-12, 3-16
ER$NEF, 3-11
ER$REX, 4-12
ER$RFA, 3-8, 3-10, 4-9, 4-11, 7-2, 7-4
ER$RLK, 2-12, 2-13
ER$RNF, 7-2, 7-4
ER$RSZ, 3-12
Error code mapping, B-4. See also
higher level language
Error messages, 9-5. See also utilities
command utilities, E-9
RMSDEF, E-1
utilities, E-1
Error, hardware read, 9-16, 9-72
Escape, 2-16
Exception record, 9-59. See also RMSIFL
Executive, 1-14. See also monitor
Exponential variance, 5-7, 6-18
Extended Buffer Pool, 8-28, 8-30
Extended diagnostic messages, 9-11, 9-67.
See also Query mode
F

FllACP, 1-33, 8-28
FAB, C-2
Factor
duplicate, 6-21
or module concatenated, C-2
name, C-2
packing efficiency, 6-22, 6-30
Fast delete, 5-18t
FCB, 8--27
FCS-11, 3-2, A-1, A-3
FIFO, 5-6, 5-11, 6-8
File, 1-1, 1-2f, 1-15
access to, initiate, 1-36. See also
file, opening
append records to a Sequential, 3-7
attributes, 1-2f, 1-3, 1-36, 1-37,
1-39, 1-40, 3-1, 3-2

File, (cont.)
attributes and RMSDSP, 9-49
attributes controlled with RMSDEF, 9-29.
See also RMSDEF
attributes identify a file, 1-3
attributes, change, 1-10, 1-40. See
also RMSCNV
attributes, display, 1-27. See also RMSDSP
attributes, list, 1-10. See also RMSDSP
back-up, 9-10. See also back up,
RMSBCK, RMSRST
blocks, unused, and RMSBCK, 9-19
closing, 1-37
closing disconnects stream, 3-16, 4-17, 7-10
closing magnetic tape, 3-16, 4-17, 7-10, F-3
cluster with buckets, 5-2, 6-17
collection of, 9-11, 9-67
container, as a, 1-1, 1-15. See
also Container file
contiguity, 3-4, 3-6, 4-3, 4-5, 8-31, D-2
contiguity and multiple areas, 6-13
contiguity as a window turning optimization,
8-29
contiguity in RMSDEF, 9-44
contiguity in RMSRST, 9-75
contiguity, cost on RSTS/E, 8-29
contiguity, maximize, 8-28
convert, 1-10, 1-27, 9-19. See also RMSCNV
corruption, 5-2
creating, 1-10, 1-27, 1-36, 1-39,
9-29. See also RMSDEF
creating and areas, 6-13
creating and DEQ, 3-5, 4-5
creating and Indexed file attributes, 7-11
creating and initial allocation quantity,
1-36, 1-39, 3-16, 4-17, 7-11
creating and placement control, 1-36,
3-16, 4-17, 7-11
creating and Relative file attributes, 4-17
creating and Sequential file attributes,
3-16
creating for Block I/0, 1-40
creating magnetic tape, F-3, F-4
creating with file specification, 3-16,
4-17, 7-11
creating with total allocation more efficient,
3-4, 4-4, 6-24
current size, 1-39
data records in a, number of, 6-18
date, back-up, 9-73
date, creation, 9-17
date, creation of back-up, 9-73
date, revision, 9-17
defining, 1-10, 1-27. See also RMSDEF

Index-6

File, (cont.)
design, 3-3, 3-4, 4-2, 4-3, 6-1, 6-2
design RMS-·11 files off-line, 1-40
different extents, 1-15
directory, 1-:33, 1-39, 3-1, 3-2, 3-5, 4-1, 8-26
erasing, 1-37
erasing and magnetic tape or unit record
devices, 3·-16
erasing causes delete from directory, 1-37
erasing releases blocks for reuse, 1-37
erasing while other programs are accessing,
1-37
extending, 1-35, 1-36, 1-36
extending and magnetic tape or unit record
devices, 3-16
extending explicitly by program, 1-36
extending implicitly by RMS-11, 1-37
extension and areas, 6--10
extension, automatic, 1-39, 3-4, 3-11,
4-12, 5-12
extension, time required for, 3-5
header, 8-26
ID, 8-31
indirect, 9-6. See also utilities
list, 1-10. See also display
load, 6-30. See also file, population
load an indexed, 1-27. See also RMS/FL
load, indexed. See indexed file load
longest record actually stored in a, 1-38
on magnetic tape, F-3. See also
magnetic tape
management, 1-33, 3-1
managers, compatibility with other, 3-2,
A-1, A-3
manipulated by RMS·-11, 1-33
medium, 1-37. See also medium
-name, 9-12, 9-13, 9-16, 9-72
next, positioning for the, F-4. See
also magnetic tape
ODL, 8-8. See also ODL, file
opening, 1-36, 1-36
opening for Block 1/0, 1-40
opening magnetic tape, 3-16, F-3
opening with File ID, 3-16, 4-17, 7-11
opening with file specification, 1-40,
3-16, 4-17, 7-11
operate on whole, 1-10. See also utilities
operation, 1-36, 1-37, 2-3, 3-6,
3-15, 4-7, 4-17, 7-10, 8-25
organization, 1-5, 1-6, 1-7, 1-11, 1-16,
1-17, 1-37, 2-12, 2-14, 2-18t, 2-19t,
2-20t, 9-19, D-2
organization and record formats, 1-37t
organization selection, 2-14

File, (cont.)
organization, change, 1-10. See also
RMSCNV
population, 6-16, 6-29, 6-30, 6-31,
9-3, 9-19, 9-29, 9-57
population techniques, 6-29, 6-30,
6-31, 6-32, 6-33. See also
RMSCNV, RMSIFL
processing environment, 1-33
processor, 1-12, 1-14, 1-16, 1-33, 1-36, 3-5
protection, 1-37. See also operating
system, protection code
records in a, 1-3f
Relative file buckets initialized, 4-4
restricted group of people, available to a, 1-2
RMSRST, restored by, ·9-67
section, F-2. See also magnetic tape
segregates data, 1-1
selection, explicit and implicit, 9-11, 9-67
sharing. See file-sharing
size, 2-3, 3-12, 4-6
sort work, 9-58
specification, 1-37, 1-40, 9-11,
9-12, 9-13, 9-16, 9-21, 9-31, 9-53,
9-61, 9-62, 9-67, 9-69, 9-72, 9-73
specification, comply with system
conventions, 1-37
standard interface, 1-28
stream ASCII, A-3
structure, 1-11, 1-28
structure and integrity, 1-12
structure, conceptual, 2-14, 4-2, 5-4
structure, physical, 2-14, 4-1, 5-2
structure, why this, 5-8
summary listing, 9-12. See also
summary listing
term applies to the data as well, 1-1
timely access to critical, 1-36
unique record address for life of, 5-1
versions of, 9-5
virtual device, as, 1-40
waste space in Sequential, 3-1
write-shared, sequentially reading, 7-10
File Access Block, C-2
File Control Block, 8-27
File Control Services, A-1, A-3.
See also FCS-11
File Processor, 1-33. See also FIP
File-sharing, 1-36, 2-3, 2-5, 2-12,
2-18t. See also shared access
access declaration, 2-7
access, exclusive, 2-9
allow declaration, 2-7
bucket locking, 2-9. See also bucket locking

Index-7

File-sharing, (cont.)
bucket, unlock, 2-12. See also Free
record operation
controlled by programs and order of access,
1-36
and file organization, 1-36
file organization restrictions, 2-12
IAS/RSX-llM exception, 2-8
lock-list, 2-12
permission, levels of, 2-5
read-type operations, for, 1-36
Record Access Streams, multiple, 2-12
RSTS/E caution, 2-7, 2-8
shared access criteria, 2-7
sharing among programs, 2-7
sharing among Record Access Streams, 2-12
specifications in programs, 2-5
system protection code, 2-5. See
also operating system, protection code
write-shared files, reading, 7-10
write-sharing and read-only access to
Relative file, 4-3
write-type operations, for, 1-36
Files--11 Ancillary Control Processor,
1-33. See also FllACP
Files--11 ODS-1, 1-14
Files--11 ODS-2, 1-14
Fill number, 6-31. See also file,
population techniques
and RMSCNV, 9-24
in RMSDEF, 9-46
and RMSIFL, 9-65
Find record operation, 1-25
block-spanning record and, 3-8
cell, empty during, 4-9
context, set, 5-14
and Current Record, 1-31, 3-8, 4-9, 7-3
and delete operation, 1-30, 1-31, 5-16, 7-2
deleted record during, 4-9, 7-2
errors depend on access mode, 3-8, 4-9, 7-2
get operation, use instead of, 3-9, 4-10, 7-3
I/0 buffer in, 1-30
Indexed files, effect on, 5-13
and key of reference, 7-3
and Next Record, 1-31, 3-8, 4-9, 7-3
Next Record in Sequential Access Mode,
use of, 5-17
random operation, cost in, 5-18t
record does not exist, 7-2
record, valid during, 4-9
Relative files, effect on, 4-8
RF A, returns, 3-8, 4-9
Sequential files, affect on, 3-8
sequential operation, cost in, 5-18t

Index-8

Find record operation, (cont.)
stream records, processing, 2-16
and update operation, 1-30, 1-31, 5-14, 7-6
use, 3-9, 4-1~ 7-3
user buffer, data not moved to, 1-30
FIP, 1-33
buffering, 8-28
First-in, first-out, 5-6, 5-11, 6-8
Fixed control area, 1-38, 2-16. See
also record format, VFC
and RMSCNV, 9-25
in RMSDEF, 9-36
and RMSIFL, 9-66
Fixed-length records, 1-37. See also
record format, fixed
Fixed record format, 2-15. See
also record format, fixed
Flag byte, 5-16
Flush record operation, 1-26
context, does not effect, 3-9, 4-10, 7-4
data written to disk, ensure all, 1-26
I/O buffer in, 1-30
Follow the index, 5-8
Form feed, 2-16, 2-17
Format, record. See record format
Forms control, 1-39
carriage control, 1-39
FORTRAN, 1-39
RMSDEF, 9-37
Four-byte signed integer key, 6-5. See also key
Four-byte unsigned binary key, 6-5.
See also key
Fragmentation, eliminating, 8-31
Free-block bit map, disk, 3-5
Free record operation, 2-12, 2-13
FSS, B-4
FSZ, 4-6
G

Generic match. See match criteria
Get record operation, 1-25
block-spanning record and, 3-10
context, set, 5-14
and Current Record, 1-31, 3-10, 4-11, 7-5
and delete operation, 1-30, 1-31, 5-16, 7-2
errors depend on access mode, 3-10, 4-11, 7-4
I/0 unit in, 1-30
Indexed files, effect on, 5-13
and Locate Mode, 3-14, 4-15, 7-7
and Move Mode, 3-13, 4-14, 7-7
and Next Record, 1-31, 3-10, 4-12, 7-5
Next Record in Sequential Access Mode,
use of, 5-17
random operation, cost in, 5-18t

Get record operation, (cont.)
read-only access and allow write declarations,
4-11
Relative files, effect on, 4-10
RFA, returns, 3-10, 4-11
Sequential files, effect on, 3-9
sequential operation, cost in, 5-18t
stream records, processing, 2-16
and update operation, 1-30, 1-31, 5-14, 7-6
user buffer in, 1-30, 1-31
Global references, 8-1, 8-20, 8-22
Greater-than, 5-11. See also match criteria
Greater-than-or-equal, 5-8. See also
match criteria
H

Hardware, 9-15, 9-72
data structure, 1-·11
devices, 1-10
driver written for specific, 1-12
mass data storage, 1-1
operating system, 1-11, 1-12
read error, 9-16, 9-72
Header
bucket, 2-5. See also bucket, formatting
record, 2-5, 5-12, 7-2
HELP message, 9-5
High-Key Value, 5--5, 5-6, 5-11
High portion, 5-12. See also bucket splitting
Higher level language, 1-26, 1-40, 5-15,
5- rn, 6-22, 6--32
error code mapping, B-4
and ODL files, 8--6
RMS-11 features available, subset of,
B--1, B-2t, B-3t
routines separate from RMS-11, 8-1
stream and connect, not provided by, 1-31
Highest possible key value, 5-6, 5-8

1/0 buffer, 1-29, 2--17
and bucket size, 6-16
and Deferred Write, 2-17
during delete operation, 1-30, 4-8.
See also Delete record operation
file, closing a, 1-37
find and get operation, use in sequential, 5-17
during find operation, 1·-30, 3-8, 4-9
during flush oper.ation, 1-30
during get operation, 1-30, 3-10, 4-11
and Locate Mode, 3-14, 4-14, 7-7
and Multi-Block Count, 2-19
and Multiple Buffers, 2-19, 4-16, 7-9. ·
See also Multiple Buffers

1/0 buffer, (cont.)
during put operation, 1-30, 3-11, 4-12
in record operation, 1-30
during update operation, 1-30, 3-12,
4-13, 7-6
and user buffer, 1-29
1/0 operation, 1-34, 2-3, 3-5
Alternate index search, one extra during,
5-12
Alternate index, in update of, 6-3
application environment effects time, 2-3
areas, time saved by, 6-11
blocks, allocation of, 6-24
bucket size, effect of, 6-16
bucket, one required to access, 5-9
cost in bucket splitting, 5-13
cost of performing record operations, 5-17
disk drives, fastest, 8-30
file processing, slowest part of, 5-8
keys, significant portion caused by, 6-2
overhead, minimize request, 8-31
performed by file processor, 1-33
physical, 2-17
Primary Key, number to locate a record by,
5-10
put operation, failure of, 6-8
during put operation, minimum, 5-10
requests made by RMS-11, 1-33
RMS-11 operations, number of, 4-3
time, 2-3, 2-4f, 3-6, 4-5, 8-30
1/0 techniques, 2-17, 2-18t
Deferred Write, 2-17, 3-14, 4-15,
7-8. See also Deferred Write
Mass Insert, 2-19. See also Mass Insert
Multi-Block Count, 2-19, 3-15.
See also Multi-Block Count
Multiple Buffers, 2-19, 3-14, 4-16,
7-9. See also Multiple Buffers
record operation, asynchronous, 3-14,
4-15, 7-8
used by RMSCNV, 9-28
1/0 unit, 1-29, 1-34, 2-18t, 4-2.
See also bucket
during delete operation, 1-30. See
also Delete record operation
during get operation, 1-30
during put operation, 1-30
size depends on file organization and design,
1-29
during update operation, 1-30
IAS, 1-14, 1-33, 1-35, 1-39, 3-1, 3-3,
3-1~ 4-1, 4-18, 6-28, 7-11, 8-2, 8-21,
8-23, 8-26, 8-27' 8-30, 9-5
file managers, compatibility with other, A-1

Index-9

IAS, (cont.)
file-sharing exception, 2-8
record operation, asynchronous, A-2
restrictions on RMS-11, A-1
RMS-11 restrictions, A-1
users, note to, 9-1
Incremental reorganization, 5-13
Index, 1-8, 1-9, 5-2, 5-6, 5-17, 5-18t, 6-3
bucket, 5-2, 5-6, 6-11, 9-46
depth, 5-5, 6-18, 6-30
entry point, 5-5
find or get operation, for search during, 5-13
follow the, 5-8
level, 2-14, 5-6, 9-46
pyramid, as a, 5-5f
record, 5-~ 5-13, 6-3, 6-17, 6-21
structure not modified during delete
operation, 5-17
structure, in conceptual, 5-4
updated during put, 5-10
updating as process, 5-13
Index Descriptor, 5-8, 6-11. See also
J(ey IJescriptor
Indexed file, 1-18. See also Indexed
file organization
Indexed file load, 1-10, 1-27. See also RMS/FL
Indexed file organization, 1-7, 1-8f,
1-9f, 1-18, 1-37
advantages, 2-19t
allocation, 6-24, 6-28
Alternate Keys at end of record, 6-7
areas, 1-39, 5-3f, 6-10, 6-13. See also area
bucket formatting, 5-4, 5-4f. See
also bucket, formatting
bucket size, 1-38. See also bucket size
characteristics and capabilities, 2-18t
data storage space, 2-5
Default Extension Quantity, 6-29
Deferred Write, 7-8
delete operation, 5-16, 5-17
DEQ is zero, effect if, 6-29
design goals, 5-1
disadvantages, 2-20t
disk only, 1-37
duplicate key effects, 6-9
duplicate keys and RMSCNV, 9-27
file design, 6-1
file-sharing, restrictions on, 2-12
fill numbers, RMSCNV honors, 9-24
fill numbers, RMSIFL honors, 9-65
find operation, effect of, 5-13
find operation, key of reference changes in,
7-3
get operation, effect of, 5-13

lndex-10

Indexed file organization, (cont.)
High-Key Value, 5-5
I/0 unit, 1-34
index not modified during delete operation,
5-17
indexes, 1-18
keys, 1-38, 6-2. See also Key
keys by RMSCNV, limit on number of, 9-26
keys by RMSIFL, limit on number of, 9-62
logical Next Record located, 7-3
Multiple Buffers, 7-9. See also
Multiple Buffers
overhead, 2-5
placement control, 6-14. See also
placement control
Primary Key, 6-7, 6-24
Prologue, 5-2, 6-26
put operation, effect of, 5-10
Random Access Mode, 1-21
random access using, 5-8
Record Access Streams, multiple, 7-10
and record formats, 1-37
record placement criterion, 1-18.
See also Primary Key
record size, 6-1, 6-2
records from another file, inserting,
9-19. See also RMSCNV
records logically adjacent, 1-21
records, only fixed and variable, 6-1
reorganize, time to, 5-2
RMSCNV uses Mass Insert, 9-24
RMSIFL optimizes indexes, 9-57
Root caching, 7-9
sequential access, fast, 1-19
sequential same as random insert, 1-21
structure, 5-2, 5-4, 5-5, 5-8
update operation, effect of, 5-14
write-shared files, reading, 7-10
Indirect file, 9-6, 9-29. See also utilities
Indirect reference to RMS-11 ODL file, 8-7
Infile, 9-12, 9-13, 9-21, 9-53, 9-61, 9-69
Information retrieval, flexibility in, 6-3
Initialize
bucket, 4-1
disk volume, 8-27
Input file, 9-19, 9-20
Insertion sequence, regardless of, 5-5
Installation process, 8-4
Integer key, 6-3. See also key
Intratask interface, 2-2
K
Key, 1-5, 1-8, 1-38, 2-1, 5-2
Alternate. See Alternate Key

Key, (cont.)
changes. See key characteristics
characteristics, 6-2, 6-7. See
also key characteristics
cost, 6-2, 6-3, 6-4, 6-5, 6-6, 6-7
data base concept, 5-1
data type, -1-38, 6-2, 6-3t, 6-4, 9-38
duplicates. See key characteristics
field, 1-2.2, 5-1, 5-15
four-byte signed integer format, 6-5
four-byte unsigned binary format, 6-5
1/0 operation, cause significant portion of,
6-2
index, 1-19, 6-2
name in RMSDEF, 9-39
null key value. See key characteristics
number, 1-19, 1-22, 1-38, 2-3, 6-2
numeric key least significant byte, 6-4, 6-6
packed decimal format, 6-6
position, 1-38, 6-2, 6-6, 9-38
position in record, benefits from, 6-7
Primary. See Primary Key
RMSIFL, number limited by, 9-62
segmented string keys, 6-7, 9-38, 9-39
selection, 6-2
signed integer value, two-byte, 6-5
size, 1-38, 6-2, 6-3, 6-3t, 6-3, 6-9, 9-39
string, 6-4
two-byte signed integer format, 6-4
two-byte unsigned binary format, 6-5
unsigned binary value, four-byte, 6-6
value, 1--22, 1-38, 5-13
value, highest possible, 5-6, 5-8
value, specific, 1-9
Key characteristics, 6-7. See also key
changes, 6-7, 6--9, 9-40
combinations, 6-8t, D-3t
cost of changes, 6-10
cost of duplicates, 6-8
cost of null key value, 6-10
definition by RMSDFN, D-3
duplicate key effects, summary of, 6-9
duplicate values, 1-38, 2-3, 5-6
duplicates, 6-7, 6-8, 9-39
null key value, 1-38, 6-7, 6-10, 9-40
restrictions, 6-8t
Key Descriptor, 1-36, 5-2, 6-26
Key of reference, 5-13, 9-20. See also index
changes during find operation, 7-3
limit in RMSCNV, 9-26
limit in RMSIFL, 9-57

L
Label, F-2. See also magnetic tape
Language, programming. See higher
level language
LB:, 8-4
LBN, 6-28, 9-42. See also Logical
Block Number
Least significant byte
four-byte signed integer, 6-5
four- byte unsigned binary, 6-5
two-byte signed integer, 6-4
two-byte unsigned binary, 6-5
Level, 5-4, 5-5f
data, 5-5. See also Level 0
horizontal chain of buckets, 5-4
index, 5-4f, 5-6
number, 5-5
Level 0, 5-5, 5-10, 5-11, 5-13, 5-14,
5-16, 6-3, 6-11
an area for, 6-10
bucket properties, 5-5
Continuation Buckets, 5-6
Secondary Index Data Record, 5-6
updated, 5-13
Level 1, 5-4, 6-10, 6-11
Level 2, 5-5
Levels 1+, 6-17
Levels 2+, 6-10, 6-11
LF, 9-10
(ill ' 9-10
Librarian utility, 8-14
Library. See Resident Library
Library, object module, 8-1
Line feed, 1-39, 2-16, 2-17
Listing, summary. See Summary Listing
Load file, 6-30. See also file, population
Load, indexed file. See indexed file load
Locate Mode. See Record Transfer Mode
Lock-list, 2-12
Logical access path, 1-19
Logical block, 1-14, 1-15
map virtual block to, 2-4
Logical Block Number, 1-15, 1-33, 8-26
cylinder, starting for, 6-14
RMSDEF, 9-42, 9-43
track, starting for, 6-14
Logical data structure, 1-14f
Logical device, 1-14, 8-4
Logical end-of-file, 9-24
Logically contiguous, 1-14, 1-15. See
also block
Lowest-addressed byte, 6-5, 6-6

Index-11

M
MACR0-11, 2-2, 5-18t, B-1
$CONNECT, 3-15
$CREATE, 3-4, 3-5, 3-6, 4-3, 4-4, 4-5,
4-6, 4-7, 6-23, 6-27, 6-28, 6-29, 6-32
F$MRN, 4-7
FAB ALQ field, 3-4, 4-4, 6-27
F AB BKS field, 4-3, 6-23
F AB DEQ field, 3-5, 4-5, 6-29
FAB FOP field, 3-6, 4-6, 4-16, 6-28,
6-31, 7-9
FAB MRN field, 4-7
F AB RTV field, 8-28
FB$CTG, 3-6, 4-6, 6-28
FB$DFW, 4-16, 6-31, 7-9
$FREE, 2-13
$INIT, 8-6
ODL file, write your own, 8-6
RAB MBC field, 3-15
RAB ROP field, 6-31, A-2, A-4, A-5
RB$ASY, A-2, A-4, A-5
RB$EOF, 3-7
RB$MAS, 6-31
RB$SEQ, 6-31
RMS-11 features available, all, B-1,
B-2t, B-3t
$SET, 3-6, 3-7, 4-6, 4-16, 6-28, 7-9
$STORE, 3-4, 3-5, 3-15, 4-3, 4-4, 4-5,
4-7, 6-23, 6-27, 6-29, 6-32
XAB ALQ field, 3-4, 4-4, 6-27
XAB AOP field, 3-6, 4-6, 6-28
XAB BKZ field, 4-3, 6-23
XAB DEQ field, 3-5, 4-5, 6-29
XAB DFL field, 6-32
XAB IFL field, 6-32
XB$CTG, 6-28
Magnetic tape, 2-18t
allocation information for files by RMSDSP,
9--57
ANSI-D format, F-1
ANSI labels, 9-2, 9-18
block size in RMSBCK, 9-18
block size in RMSCNV, 9-23
block size in RMSDEF, 9-38
block size, maximum, 9-23
block size, minimum, 9-23
data storage, secondary, 1-11
end-of-file indicators, 3-2
file attributes shown by RMSDSP, 9-51
file processing, F-3
file section, F-2
format, operating system enforces rest of, F-1
handling largely transparent to RMS-11, F-1

Index-12

Magnetic tape, (cont.)
label, F-2
next file, positioning for the, F-4
processing, general, F-1
processing, RMS-11, F-3
record size, minimum fixed-length, 2-15
RMSBCK, rewind by, 9-16
RMSRST /BD switch, cannot use, 9-73
RMSRST, rewind by, 9-72
volume, F-1
volume handling on RSTS/E, A-3
volume set, F-2
volume, mounted, F-4
volumes, rewinding, F-3
'Magnetic Tape Labels and File Structure for
Information', F-1
Map, 8-14, C-1. See also Task Builder
disk free-block bit, 3-5
name, C-1
virtual block to logical block, 2-4
Mapping, error code, B-4
Mapping, virtual-to-logical-block. See
virtual-to-logical-block mapping
Mass data storage, 1-1. See also hardware
Mass Insert, 2-19, 6-33, 9-24. See
also file, population techniques
and Deferred Write, 6-33
Match criteria, 1-21, 1-22, 5-13
during find or get operation, 5-14
on Indexed files, 5-13
on Relative files, 4-8, 4-10
Maximum Record Number, 1-38, 1-38,
2-18t, 4-2, 4-6. See also Relative
file organization
allocating a Relative file by putting, 4-5
maximum, technical, 9-37
in RMSDEF, 9-36
Maximum Record Size, 4-2, 6-2
MBC, 2-19. See also Multi-Block Count
MBF, 2-19. See also Multiple Buffers
Medium, 1-17, 2-18t
back-up, 9-18
depends on file organization, 1-37
magnetic tape, F-1. See also magnetic tape
Memory, 1-33
Memory Allocation Map, C-1. See
also map
Memory-resident overlays, 1-28, 8-3f,
8-4, 8-20, 8-24
I/0 operation, eliminate, 1-29
maintained separately in memory, 8-4
shared among programs, 1-28
Modes, access. See Record Access Modes

Module
concatenated factor or, C-2
library, object, 8-1
name conventions, 8-14
Monitor, 1-12
Most significant byte
four-byte signed integer, 6-5
four-byte unsigned binary, 6-5
packed decimal, 6-6
string keys, 6-4
two--byte signed integer, 6-4
two·· byte unsigned binary, 6-5
Move Mode, 3--13, 4-14, 7-7. See also
Record Transfer Mode
MRN. See Maximum Record Number
Multi-Block Count, 2-19, 3-15
and Relative and Indexed files, 2-19
RMSCNV, used by, 9-28
run time, set at, 3-15
sequential processing, improves, 3-15
task size, increases, 3-15
Multiple Buffers, 2-19
and Indexed files, 7-9
' and Relative files, 4-16, 4-17
and Sequential files, 3--14
sequential operation, no benefit during,
4-17, 7-9
N
NDRBK, 6-17, 6-·18, 6-22, 6-24, 6-30
Next file, positioning for the, F-4.
See also magnetic tape
Next Record, 1-31, 3-7, 4-8, 7-1
and connect operation, 3-7, 4-8, 7-2
findl and get operation, use in sequential,
fi-17

and find operation, 3-8, 4-8, 4-9, 7-3
and get operation, 3-9, 3-10, 4-10,
4-12, 7-5
Indexed files, logical sequence enacted in,
7-3
and put operation, 3-11, 4-13, 7-5
record operation failure, affect of,
~~- 7' 4-8' 7-1
record operation, affected by, 1-32
record operation, cost in, 5-18t
record operation, target of, 1-31
NIRBK, 6-17, 6-18, 6-21, 6-22, 6-30
Nonoverlaid RMS-11, l-·28, 8-2, 8-3f, 8-24
/Noqu, 9-15, 9-71
NRBKT, 4-4
NRF, 3-4, 4-4, 6--18, 6-22, 6-24, 6-30
NRI, 6-18
NUL, 2-16, 2-17

Null byte, 2-5, 3-2. See also
Sequential file organization
Null key characteristic, 5-18t, 6-7.
See also key characteristics
Numeric data, 1-12
0
Object
code, 1-28, 8-1
module library, 8-1
ODL, 8-4. See also Overlay Description
Language
file, 8-4, 8-5f, 8-6, C-6
optimization examples, 8-14
ODS. See On-Disk Structure
ODS-1, 1-14
ODS-2, 1-14
On-Disk Structure, 1-14, 3-3
Operating system, 1-11, 1-28, 1-33, 1-37,
2-5, 3-1, 5-12, 8-1, 8-27
and ANSI-D format, F-1
areas invisible to, 1-35
blocks, allocation of, 6-24
contiguity handled differently, 3-4, 4-3
disk, system, and RMSBCK, 9-18
disk, system, and RMSRST, 9-75
failure, during write-type record operation,
5-2
failure, preserve state of processing despite,
5-1
memory layout, 1-27f
monitor coordinates components, 1-12
overlay segment, reads, 8-4
protection code, 2-5, 2-6f. See
also file-sharing
protection code changed by RMSRST, 9-71
protection code definition by RMSDFN, D-4
protection code, comply with conventions,
1-37
resources, more to a task, 8-29, 8-30
RMS-11 as part of program, 1-27
RMS-11, interface with, A-1
Optimal density, 1-34
Optimizations
apply one by one, 2-14
other, 8-29
Outfile, 9--12, 9-21, 9-53, 9-61, 9-69, D-1
Output file, 9-19, 9-20, 9-21, 9-61, 9-65
Overhead
activity, 6-18
bucket, 5-4, 6-2. See also Indexed
file organization
deletion, 6-18
duplicate, 6-22

Index-13

Overhead, (cont.)
1/0 requests, minimize, 8-31
record format, 4-2, 4-4, 4-7, 5-4, 6-1,
6--2, 6-18
Sequential files, almost none in, 3-1
Overlaid RMS-11, 8-2
Overlay, 1-28, 2-1, 2-3, 2-4, 3-5,
8-2, 8-3f, 8-4, C-2
disk-resident, 1-28
memory-resident, 1-28
segment, 8-4, 8-7
structure, 1-28, 8-4, 8-5f, 8-7,
8--8, 8-12, 8-13f, 8-24, C-1
types, deciding between, 8-23
Overlay Description Language, 8-4, 8-6

Performance, (cont.)
Deferred Write on Relative files, improved by,
4-16
design, 2-1, 2-2
disk devices, impact of, 8-30
evaluate, 2-14
file size, affected by, 2-3
fill numbers improve, 6-31
and Mass Insert, 6-33
mass storage devices and, 2-3
and placement control, 1-35
program development, 8-24
program, affected by, 2-3
search time curves, 5-9f
sequential access via Primary Key, 5-13
write-shared files, improved while reading,
p
7-10
Physical block, 1-16
Pack, Ei-14
PIP, 6-28
Packed decimal key, 6-3, 6-6. See also key
/ENTER switch, 9-18
Packing efficiency, 6-22, 6-30
PKL, 6-17
Pad byte, 3-2
Placement control, 1-35, 2-14, 6-14
Patch level, 9-7, 9-14, 9-22, 9-54, 9-62,
allocation, nonzero, 9-43
9-70. See also utilities
and areas, 6-10, 6-28
PDP-11, 8-2
block calculations, starting, 6-14
PDP-11 COBOL, 2-2, 2-13, 4-7, 4-16, 6-32,
Device
Cluster Number, 6-15
7-9, B-1. See also higher level language
by
file
or
by areas, 1-35
/AL switch in ASSIGN clause, 3-4, 4-4,
file,
creating
a, 1-36
6-28
head
movement,
6-14
/AL switch in VALUE OF ID, 3-4, 4-4,
in RMSDEF, 9-42
6-28
virtual block specifications, 6-15
BLOCK CONTAINS clause in
Platter, 1-11
file-description-entry, 6-23
PLG, 4-4
BLOCK CONTAINS clause in
Pointer, 5-6, 5-8, 5-17
file-description-entry, 4-3
array, size of, 5-6
bucket sizes, different, 6-19
to data records, 5-5
changes and no duplicates, 5-15
length of bucket, 6-17, 6-21
/CO switch in ASSIGN clause, 3-6, 4-6,
retrieval, 8-26, 8-27, 8-28
6-28
Primary index, 5-7f, 5-10, 5-11, 5-14,
/CO switch in VALUE OF ID, 3-6, 4-6,
9-20. See also Primary Key
6-28
bucket size, 6-17
/EX switch in ASSIGN clause, 3-5, 4-5, 6-29
optimization, 9-57
/EX switch in VALUE OF ID, 3-5, 4-5, 6-29
EXTEND keyword in OPEN statement, 3-7
Primary Key, 1-9, 1-18, 5-2, 5-11,
key characteristic combinations, 6-8
5-12, 6-2, 9--20, 9-24, 9-26, 9-57, D-1.
OTS, 5-15
See also Alternate Key
SEQUENTIAL and RESERVE AREAS
ascending order and allocation, 6-24
clauses in OPEN state, 3-15
definition by RMSDEF, 9-38
Performance, 2-3. See also speed
definition by RMSDFN , D-3
and areas, 1-35, 6-10
duplicate values, 1-38
and bucket size, 1-38
duplicates allowed, 6-7, 6-8
contiguity and areas, 6-13
I/Os to locate a record, number of,
contiguity, impacted by, 3-4, 4-3
5-10
defaults, affected by, 2-1
length and bucket size, 6-17
Deferred Write on Indexed files, improved by, 7-8
null key, no, 6-10

Index-14

,,,

Primary key, (cont.)
optimal at beginning of record, 6-7
population and value, 6-30
in Prologue, 6-26
put operation, use in sequential, 5-17
record operation, cost in, 5-18t
records ordered by increasing value, 5-5
sequential access, performance on, 5-1
value and population, 6-30
value cannot change during update operation,
6-9
value changed during update operation, 5-15
Primary ODL file, 8-6. See also ODL, file
Primary Root, 5-13. See also Root
Prints vs. types, 9-10
Priorities, 8-30
Program, 1--11, 1-25, 1-40, 2-3, 8-3f.
See also application
during Block I/0, must interpret contents of
block, 1-40
buffer for record in, 1-29. See also
user buffer
considerations, 2-12
control returned before operation is done,
1-33. See also record operation,
asynchronous
development, 8-1, 8-24
executable form, 1-27
file processor, affected by, 1-33
file-sharing, 2-5, 2-7
file simultaneously, multiple update a, 2-9
file, explicitly extends a, 1-36
make records available to, 1-26. See
also Connect record operation
nonkey field, sorts by, 6-3
object code, convert to, 1-28
records, before it can access, 1-31
records, RMS-11 processes, 1-27
HMS-11 communicates with, 1-30
and RMS-11 routines, 1-28, 8-1, 8-2.
See also nonoverlaid RMS-11
syntax and bucket size, 6-22
and your computer system, 1-10
Programming language, 1-28, B-1. See
also higher level language
Prologue, 1--39 4-1, 5-2, 5-4, 5-12, 5-13, 5-18t
bucket size, extended to integral multiple of,
4-1, 5-2, 6-26
size calculation, 6-26
Prompt, 9-12, 9-21, 9·-52, 9-61, 9-68
Protection code, 2-2, 2-6f. See also
operating system, protection code
Prototype 0 D L file, 8-4
optimization, 8-8, 8-9
precautions, 8-9
1

Put record operation, 1-26
Alternate Keys, position of, 6-7
bucket splitting, 5-12
cost, 5-18t, 6-3
and Current Record, 3-11, 4-13, 7-5
duplicate key values, effect of, 6-8
duplicates not allowed, failure and, 6-8
file extension, automatic, 1-37, 3-11,
4-12, 5-12
I/0 buffer in, 1-30
Indexed files, 5-10
and Locate Mode, 3-14, 4-15, 7-7
and Move Mode, 3-13, 4-14, 7-7
and Next Record, 1-31, 3-11, 4-13, 7-5
Primary Key in Sequential Access Mode,
use of, 5-17
record in a file, store new, 1-26
records, compress deleted, 5-11
Relative file, allocating with Maximum
Record, 4-5
Relative files, effect on, 4-12
Sequential files, effect on, 3-11
stream records, processing, 2-17
truncate operation, immediately following,
3-12
user buffer in, 1-30, 1-32
Pyramid, 5-4, 5-5f
Q

/QU switch, 9-11, 9-67
Query mode, 9-11, 9-67. See also
extended diagnostic messages
R
Radix, decimal, 9-29
Random access, 1-11, 5-1, 5-8
Random Access Mode, 1-19, 1-21, 1-22,
1-23, 4-9
find operation, 7-2
get operation, 7-4
Read
access, 2-12. See also
file-sharing
error, 9-15, 9-16, 9-72
sharing. See file-sharing
-type operation, 2-12. See also
record operation, read-type
Reading program, 2-12. See also file-sharing
Record, 1-3. See also Indexed,
Relative, and Sequential file
in a file, 1-3f, 5-1, 6-18
access by contents, 5-1
access, randomly, 1-5
access, sequentially, 1-4
block spanning, 1-34, 1-34f

Index-15

Record, (cont.)
block, as a, 1-17. See also record
format, undefined
bucket spannin~ 1-34, 4-1, 6-2
buffer in program, 1-29. See also user buffer
byte-aligned, 4-1, 5-4
cell calculation, 4-2
consecutive, 1-19. See also
Sequential Access Mode
definition, 3-2, 4-2, 6-1
deleted, 4-9, 6-2
deleted and header flag, 7-2
deleted during find operation on Relative file,
4-9
deleted from Sequential file by truncate
operation, 3-11
deleted, compression, 5-11, 5-17, 6-7
deleted., during find operation, 7-2
deleted, effect of duplicates, 6-9
deleted, flag byte in, 5-16
deleted, not compressed during delete
operation, 5-17
duplicate, 1-10. See also back up
file to another, move from one, 9-19.
See also RMSCNV
during find operation on Relative file, 4-9
fixed and variable parts, 1-16. See
also record format, VFC
format. See record format
heade~ 2-5, 5-12, 7-2
identifier, 1-5, 5-4f
identifier, unique, 1-23. See also
Record 's File Address
index, 5-6. See also index, record
insert, 5-12
insertions, number of, 6-18. See
also Put record operation
key, 6-E)
length, 1-16, 1-34. See also record
format
-length field, 2-5, 2-15, 2-15f
location, physical, 1-5
longest actually stored in a file, 1-38
number, relative, 1-7
operation. See record operation
overhead, 2-18t
pointer, 5-5
and Primary Key, 5-5
processed by RMS-11, 1-33
processing environment, 1-27
during put operation, 5-12
retrieved every, 1-19
size, 1-37, 1-38, 4-4, 4-7, 6-1, 6-2,
6-9, 6-18, 9-36
storage, 3-1, 4-1

Index-16

Record, (cont.)
undefined, 1-17. See also record
format, undefined
and update operation, 5-15
Record Access Mode, 1-4, 1-5f, 1-11, 1-16,
1-19, 2-18t
Access by RFA, 1-19. See also Access
by RFA
changing, 1-23
file organization and medium must support,
1-25
Random Access Modes, 1-19. See also
Random Access Mode
Sequential Access Mode, 1-19. See
also Sequential Access Mode
Record Access Stream, 1-31, 5-15, 7-1
channel between program and file, 1-31
context, 1-4, 1-31
context and record operation, 1-32f
context and Sequential Access Mode, 1-32
context during find and get operation, 5-14
context, affect of record operation failure,
3-7, 4-8, 7-1
context, cannot resume after disconnect,
3-8, 4-8, 7-2
context, more than one for a file, 1-32
Current Record, 1-31. See also
Current Record
Indexed files, multiple and, 7-10
multiple, 1-32, 2-12
Next Record, 1-31. See also Next Record
record at a time, one, 1-31, 1-33
record operation, used for each, 1-31
Relative files, multiple and, 4-17
Sequential files, multiple and, 3-15
sharing among, 2-12
terminate, 1-26
Record format, 1-4, 1-4f, 1-11, 1-16,
1-37, 2-1, 2-14, 2-18t
defined by RMSDFN, D-4
end-of-file indicator for nonstream, 3-2
fixed, 1-16, 2-5, 2-15, 3-4, 4-2t,
6-1, 9-28, 9-67
fixed control area size, 2-16
fixed-length record size, 2-15
overhead, 4-2, 4-4, 4-7, 5-~ 6-1, 6-2, 6-18
padding fixed with RMSCNV, 9-24
padding fixed with RMSIFL, 9-65
record-length field, 2-5, 2-15, 2-15f, 2-16
and record size, 1-38
stream, 1-17, 2-16, 3-2, 3-12
undefined, 1-17, 2-17
variable, 1-16, 2-5, 2-15, 2-15,
3-4, 4-2t, 5-18t, 6-1, 9-27
variable area size, 2-16

"-"'

Record format, (cont.)
variable-length record size, 2-15
VFC, 1-H>, 2-16, 3-4, 4-2t, 9-25, 9-66
Record operation, 1-5, 1-11, 1-16, 1-25,
2-3, 2-18t, 3-6, 4-7, 8-25
asynchronous, 1-33, 3-14, 4-15, 7-8,
8-2, A--2, A-4, A--5
bucket size, effect of, 6-16
connect, 1-26. See also Connect
record operation
context, 1-32~ 3-7, 4-8, 7-1. See
also Record Access Stream, context
costs, 5-17, 5-18t
delete, 1-25. See also Delete record
operation
disconnect, 1-26. See also
Disconnect record operation
find, 1-25. See also Find record operation
flush, 1-26. See also Flush record
operation
get, 1-25. See also Get record
operation
Indexed fjies? effect on, 5-10, 7-1
put, 1-26. See also Put record
operation
random on Indexed file, 5-8
random, procedures for performing, 5-10
read-type, 1-25, 2-7, 2-12, 7-1
Hecord Access Stream, use, 1-31. See
also Record Access Stream
rewind, 1--26. See also Rewind record
operation
HSTS/E, asynchronous not on, A-3
on Sequential files, 3-7
sequential, procedures for performing, 5-17
stream records, processing, 2-16
stream, one asynchronous per, 1-33
synchronous, 1-33
update, 1--26. See also Update record
operation
write-type, 1--25, 2-7, 2-12, 5-2, 7-1
write-type and Deferred Write, 2-17
Record Reference Vector, 5-12
Record storage cell, 1-17, 1-18, 2-5, 4-1,
4-2, 9-20. See also relative record numbers
calculation of size, 4-2
control byte in, 4-8
during find operation, 4-9
numbering, 4-2
record per one, 4-2
size, 4-2
Record Transfer Mode, 1-33
changing at run time, 3-13, 4-14, 7-6
data integrity, 3-14, 4-15, 7-7

Record Transfer Mode, (cont.}
data transfers, eliminating, 3-14, 4-14, 7-7
and Indexed files, 7-6
Locate Mode, 3-14, 4-14, 7-7
Locate Mode, user buffer during, 3-14,
4-15, 7-7
Move Mode, 3-13, 4-14, 7-7
and Relative files, 4-14
and Sequential files, 3-13
Record 's address, 1-5, 5-1. See also
Record 's File Address
Record's File Address, 1-23, 3-2, 5-1
access preserved by RRV, 5-12
deleted record, J-23
find operation, returned during, 3-8, 4-9
get operation, returned during, 3-10, 4-11
get operation, use in, 3-9, 4-10
in RMSIFL, 9-59
Relative file, 1-17. See also Relative
file organization
Relative file organization, 1-7f, 1-7,
1-17, 1-18f, 1-37
Access by RF A, 4-2
advantages, 2-19t
allocated in bucket increments, 4-1
allocation by putting Maximum Record, 4-5
applications, 4-1
bucket size, 1-38. See also bucket size
buckets, initialization of, 4-1
cell size calculation, 4-2
cells, series of fixed-size, 4-2
cells, skip logically empty, 1-20
characteristics and capabilities, 2-18t
data sizes, maximum, 4-2t
data storage space, 2-5
Deferred Write, 4-16
deleted record control, 4-1
disadvantages, 2-20t
disk only, 1-37
file-sharing, restrictions on, 2-12
file size calculation, 4-6
find operation, effect of, 4-8
get operation, effect of, 4-10
I/O unit, 1-34
Maximum Record Number, 1-38, 4MRN in RMSDEF, 9-36
MRN, technical maximum, 9-37
Multiple Buffers, 4-17
placement control, 6-14. See also
placement control
Prologue, 4-1.fbiSee also Prologue
put operation, effect of, 4-12
random access, 1-18
Random Access Mode, 1-21, 4-2

Index-17

Relative file organization, (cont.)
Record Access Streams, multiple, 4-17
record deletion, 1-18
and record formats, 1-37
record number, if you want no checks on,
4-7
records from another file, inserting,
9-19. See also RMSCNV
records not necessarily adjacent, 1-20
Sequential Access Mode, 4-2
structure, 4-1, 4-2
update operation, effect of, 4-13
write-sharing and read-only access, 4-3
Relative record number, 1-7, 1-18,
1-21, 4-2, 4-10
Relative volume number, 9-42
Reorganization periods
eliminated, 5-13. See also
incremental reorganization
Reorganization, incremental. See
incremental reorganization
Resident Library, 8-1, 8-3f, 8-20, 8-21,
8-22, 8-23, 8-24, 8-31
RESLIB option, 8-22
Restoration account and/or volume, 9-67
Restore, 1-10, 1-27. See also RMSRST
~' 9-10
Revision date. See file, date
Rewind record operation, 1-26
context, affect on, 3-11, 4-13, 7-6
file, return to beginning of, 1-26
Rewinding tape volumes, F-3
RFA. See Record 's File Address
RFO, 4-4, 4-7, 6-18
RMSllS.ODL, 8-4, 8-7, 8-7, C-1, C-2,
C-3f, C-4f, C-5f
graphic illustration, 8-8, C-1
RMSllX.ODL, 8-4, 8-7, 8-7, C-1, C-3f,
C-4f, C-5f
graphic illustration, 8-8, C-1
RMS12X.ODL, 8-4, 8-7, 8-7
RMS16X.ODL, 8-4, C-6
RMS20X.ODL, 8-4, C-7
RMSALL, 8-6, C-4f
co-tree, as, 8-7
concatenated with user root, 8-7
overlay structure, in user, 8-7
in RMS16X.ODL, C-6
in RMS20X.ODL, C-7
RMSBCK, 1-26, 9-1, 9-10, 9-67
?, 9-13, 9-14t
ANSI, compliance with, 9-10
blocks, unused, 9-19
call and termination, 9-12

lndex-18

RMSBCK, (cont.)
cautions, 9-18
/CD, 9-14t, 9-17
check-after-write error with RC, 9-14
command string, 9-12, 9-12, 9-13, 9-17
container file, 9-13
data integrity check, 9-11, 9-14
default process, 9-14t
diagnostic message, 9-14
effect, 9-10
error messages, E-9
file selection, explicit and implicit, 9-11
global switches, 9-14, 9-14t
HELP, 9-13, 9-14t
/ID, 9-14, 9-14t
infile, 9-13
infile switches, 9-14t, 9-17
magnetic tape, 9-10, 9-18
/NOQU, 9-15
/NORA, 9-14t
/NORC, 9-14t
/NORW, 9-14t
/NOSU, 9-14t
outfile, 9-12
outfile switches, 9-14t, 9-15
prompt, 9-12
purpose, 9-10
/QU, 9-11, 9-14, 9-14t, 9-15, 9-16
Query mode, 9-14
query, answer N to, 9-15
/RA, 9-14t, 9-15
/RC, 9-14t, 9-16
/RD, 9-14t, 9-17
read-after-write error with RA, 9-14
read error on input file, 9-14
read errors, automatically retries, 9-11
/RW, 9-14t, 9-16
/SL, 9-14t, 9-15
/SU, 9-14t, 9-16
summary listing, 9-12, 9-15. See
also summary listing
using, 9-3
version number, current, 9-14
RMSCNV, 1-27, 6-30, 6-31, 8-29, 9-1, 9-9,
9-19, 9-29, 9-57, 9-75
? ' 9-21, 9-22t
/AP, 9-20, 9-21, 9-22t, 9-23
/BL, 9-22t, 9-23
call and termination, 9-21
cautions, 9-27
command string, 9-21, 9-21, 9-26
default process, 9-22t
duplicate key values, 9-27
effect, 9-19

RMSCNV, (cont.)
eriror messages, E-9
fill numbers, honor, 9-24
/FO, 9-22t, 9-23
global switches, 9-22, 9-22t
HELP, 9-21, 9-22t
I/0 techniques, 9-28
/ID, 9-22, 9-22t
Indexed file as input, 9-20
Indexed file as output, 9-20
Indexed file, limit to keys in, 9-26
infile, 9-21
in:file switches, 9-22t, 9-26
input records, truncate, 9-25
key of reference, 9-20, 9-26
/KR, 9-22t, 9--26
/LO, 9-22t, 9-24
/MA, 9-22t, 9--24
magnetic tape block size, 9-23
Mass Insert, use, 9-24
outfile, 9-21
outfile switches, 9-22t, 9-23
output file, specifying organization of, 9-23
pad input records, 9-24
padding and fixed-length records, 9-24
padding characters, 9--24
/PD, 9-22t, 9-24
prompt, 9-21
purpose, 9-19
records, fixed- and variable-length, 9-28
records, fixed-length, 9-28
records, source of, 9-21
records, variable-length, 9-27
Relative file as input, 9-20
Relative file as output, 9-20
Sequential file as input, 9-20
Sequential file as output, 9-20
Sequential file, append records to a, 9-23
Sequential file, supersede a, 9-25
/SL, 9-22, 9-22t
/SU, 9-20, 9-21, 9-22t, 9-25
summary listing, 9-22
/TR, 9-22t, 9-25
unit record devices, 9·-28
using, 9-3
version number, current, 9-22
version, creates next higher, 9-27
VFC records, handle, 9-25
/WF, 9-22t, 9-·25
RMSDEF, 1-27, 3-4, 3--5, 4-3, 4-4, 4-5,
4·-7, 6-23, 6-27, 6-28, 6-29, 6-32, 9-2,
9·-29, 9-511 9--75
? ' 9-29, 9-31
alignment, type of, 9-42

RMSDEF, (cont.)
allocation, data, 9-45
allocation, initial, 9-43
areas, 9-41
block spanning, 9-37
bucket size, 9-44
call and termination, 9-31
carriage return control, 9-37
caution, 9-35
changes, 9-40
cl ustersize, 9-48
command file, 9-29, 9-31, 9-32, 9-34, 9-48
command string, 9-31
contiguity, 9-44
data buckets, 9-45, 9-46
data structure, 9-35
Default Extension Quantity, 9-44
Default Extension Quantity, file, 9-45
duplicates, 9-39
effect, 9-29
error messages, E-1
file attributes, specify, 9-29
file creation, 9-48
file organization, 9-35
file specification, 9-32, 9-34
file structure, 9-41
file, supersede existing, 9-34
fixed control area, size of, 9-36
FORTRAN character control, 9-37
HELP, 9-31
IAS/RSX-llM, protection for, 9-47
/ID, 9-31, 9-32
index buckets, 9-46
index level, area number for lowest, 9-46
information requested, list of, 9-29
key, 9--38, 9-39
location, 9-43
magnetic tape, 9-35, 9-38
Maximum Record Number, 9-36
Maximum Record Size, 9-36
null key value, 9-40
placement control, 9-42, 9-43
placement, exact, 9-43
population, no, 9-29
processing flow, 9-33f
purpose, 9-29
record format, 9-36
RSTS/E, protection for, 9-48
using, 9-4
utility, terminating the, 9-32
version number, current, 9-31, 9-32
volume number for VAX only, 9-42
RMSDFN, 9-2. See also RMSDEF
I AL, D-2, D-2t

Index-19

RMSDFN, (cont.)
allocation quantity, D-2
Alternate Key definition, D-3
/BK, D-2, D-2t
bucket size, D-2
call and termination, D-1
cautions, D-6
/CO, D-2, D-2t
command string, D-1
contiguity, D-2
examples, D-5
file organization, D-2
/FO, D-2, D-2t
global switches, D-2t
/ID, D-2, D-2t
key characteristics, D-2
key definition, D-2
/KY, D-2, D-2t, D-3
/OP, D-2, D-2t, D-3
outfile, D-1
outfile switches, D-2t
/PR, D-2t, D-4
Primary Key definition, D-3
protection code, D-4
record format, D-4
record size, D-4
record size limit, D-6
/RF, D-2t, D-4
switches, D-1, D-2
version number, current, D-2
RMSDSP, 1-27, 1-40, 9-2, 9-9, 9-49
?, 9-53
ANSI labels, 9-53
areas, 9-54
/BP, 9-53t, 9-54
call and termination, 9-52
cautions, 9-57
command string, 9-52, 9-53, 9-54
container file, 9-53
container file, list contents of, 9-54
default process, 9-53t
disk Sequential files, 9-50
display, format of, 9-50
effect, 9-50
error messages, E-9
/FU, 9-53t, 9-54
HELP, 9-53
/ID, 9-53t, 9-54
Indexed files (full display), 9-51
Indexed files (short display), 9-51
Indexed files, full attributes of, 9-54
infile, 9-53
key definition, 9-54
magnetic tape files, 9-51, 9-57

Index-20

RMSDSP, (cont.)
outfile, 9-53
prompt, 9-52
purpose, 9-49
Relative files, 9-51
switches, 9-54
using, 9-4
version number, current, 9-54
RMSFAB, 8-3f, C-2
RMSFIL, C-2
RMSIFL, 1-27, 6-22, 6-30, 8-29, 9-2, 9-29,
9-57
? ' 9-62, 9-63t
call and termination, 9-61
cautions, 9-66
command string, 9-61, 9-61
/DE, 9-63t, 9-65
default process, 9-63t
disk space requirement, 9-58, 9-59, 9-60
effect, 9-57
/ER, 9-63, 9--63t, 9-65
error messages, E-9
exception record, 9-59, 9-63, 9-64
exception records, disallow, 9-65
fill numbers, honor, 9-65
global switches, 9-62, 9-63t
HELP, 9-62, 9-63t
/ID, 9-62, 9-63t
Indexed file as input, 9-58
Indexed file, existing, 9-61
Indexed file, populating an, 9-57
infile, 9-62
infile switches, 9-63t, 9-65
input file, requirements for, 9-62
input file, sorts, 6-30
input records, pad, 9-65
input records, truncate, 9-65
key of reference, 9-57, 9-66
/KR, 9-57, 9-63t, 9-66
/LO, 9-63t, 9-65
/NOER, 9-63t, 9-64, 9-65, 9-65
/NOSO, 9-57, 9-63t, 9-66
operation, phases of, 9-57
outfile, 9-61
outfile switches, 9-63, 9-63t
output file, delete, 9-65
output file, requirements for, 9-61
padding and fixed-length records, 9-65
/PD, 9-63t, 9-65
prompt, 9-61
purpose, 9-57
Record 's File Address, 9-59
records, fixed- and variable-length, 9-67
records, fixed-length, 9-67

RMSIFL, (cont.)
records, source of, 9-62
Relative file as input, 9-58
Sequential file as input, 9-58
SORT-11 modules, 9-58
SORT error, 9-58
sort phase, 9-57
sort phase, bypass, 9-66
sort work devices, reassign, 9-65
sort work files, 9-58
techniques not available, 6-30
/TR, 9-63t, 9-65
version number, current, 9-62
VFC records, ignores fixed control area in,
9-66
RMSLIB.OLB, 8-1
RMSRAB, 8-3f
RMSRES, 8-21
RMSRLS.ODL, 8-22
RMSRLX.ODL, 8-22
RMSROT, 8-H, C-3f
concatenated with user root, 8-7
in RMS16X.ODL, C-6
in RMS20X.ODL, C-7
RMSRST, 1-27, 9-1, 9-2, 9-10, 9-13, 9-67
'?' 9-69, 9-70t
account and/or volume, choice of, 9-67
account number, outfile, 9-69
account number, restore by original, 9-73
back-up date, restore on, 9-67
back-up files, restores, 9-67
/BD, 9-70t, 9-73
call and termination, 9-68
cautions, 9-75
check-after-write error with RC, 9-70
command string, 9--68, 9-69, 9-74
container file-name, 9-69
contiguous files, restoring, 9-75
data integrity check, 9-68, 9-70
default process, 9-70t
diagnostic message, 9-71
error messages, E-9
file selection, explicit and implicit, 9-67
files from container file, select, 9-73
/FR, 9-70t, 9-71
global switches, 9-70, 9-70t
HELP, 9·-69, 9-70t
/ID, 9-70, 9--70t
infile, 9-69
infile switches, 9-70t, 9-73
/NOQU, 9-71
/NORA, 9-70t
/NORC, 9-70t
/NOSU, 9-70t

RMSRST, (cont.)
/OA, 9-70t, 9-73
outfile, 9-69
outfile switches, 9-70t, 9-71
prompt, 9-68
protection codes, change, 9-71
purpose, 9-67
/QU, 9-67, 9-70, 9-70t, 9-72
Query mode, 9-70
query, answer N to, 9-71
/RA, 9-70t, 9-72
/RC, 9-70t, 9-72
read-after-write error with RA, 9-70
read error on input file, 9-70
read errors, automatically retries, 9-68
/RW, 9-72
/SE, 9-70t, 9-73
/SL, 9-70t, 9-71
/SU, 9-70t, 9-72
summary listing, 9-68, 9-71. See
also summary listing
using, 9-3
version number, current, 9-70
RMSSEQ, 8-21
Root, 1-9, 5-5, 5-8, 5-13, 5-18t, 6-3,
6-11, 6-16
cache, 2-19, 7-9
ROOT statement, 8-6
Round-robin scheduling, 8-30
Routines
Alternate Level 0 bucket, split, 8-8
data, that organize, 1-10
damage assessment, 9-8. See also utilities
file close, 8-8
file open, 8-7
index records, insert, 8-8
nonoverlaid, 8-2
overlaid, 8-2
Primary Level 0 bucket, split, 8-8
Primary Level 0, insert records into, 8-8
with program, combining, 8-1
Record Access Stream connect, 8-7
Record Access Stream disconnect, 8-8
Relative file record operation, 8-8
Sequential file record operation, 8-8
SIDRs, insert, 8-8
task, part of your, 8-2
RPG II, 2-2, 2-13, 3-6, 3-7, 3-15, 4-6,
4-7, 4-16, 6-28, 6-29, 6-32, 7-9,
B-1. See also higher level language
RPGASN utility, 3-4, 3-5, 4-3, 4-5,
6-23, 6-28
RRV, 5-12, 5-17, 5-18t, 6-30. See
also Record Reference Vector

lndex-21

RSTS/E, 1-14, 1-33, 1-35, 1-39, 3-1, 3-3,
3-16, 4-1, 4-18, 7-11, 8-23, 8-26,
8-27, 8-28, 8-30
file managers, compatibility with other, A-3
file-sharing caution, 2-7, 2-8
RA and RC switches not available, 9-14,
9-71
restrictions on RMS-11, A-3
RMS-11 restrictions, A-2
special account characters, support of, 9-2
RSX-UM, 1-14, 1-33, 1-35, 1-39, 3-1, 3-3,
3-16, 4-1, 4-18, 6-28, 7-11, 8-2, 8-21,
8-23, 8-2~ 8-27, 8-30, 9-5
Asynchronous record operation, A-4
command line, automatically terminates, 9-7
file managers, compatibility with other, A-3
file-sharing exception, 2-8
restrictions on RMS-11, A-3
RMS-11 restrictions, A-3
RSZ, 3-4, 4-4, 4-7, 6-18
Run time, 2-12, 3-13, 4-14, 4-16, 7-6, 7-9

s
Search time curves, 5-9f
Secondary Index Data Record, 5-6
format, 5-6f
Secondary key. See Alternate Key
Secondary ODL file, 8-6. See also ODL, file
Sector, 1-12, 6-14. See also logical block
Segmented key, 6-7. See also key
in RMSDEF, 9-38, 9-39
Segments that can run independently, task, 8-2
Sequential access, 1-11
Alternate Key, performance via, 5-1
Primary Key, performance and, 5-1, 5-13
Sequential Access Mode, 1-19, 4-9
context, importance of, 1-32
find operation, 3-8, 4-8, 7-2
get operation, 3-9, 4-10, 7-4
Indexed files, 1-21
key, retrieval sequence depends on, 1-21
next record depends on organization, 1-19
Primary Key sequence, insert in
nondescending, 1-21
put operation, 3-11, 4-12, 7-5
and record storage cell, 1-20
Relative files, 1-20
relative record number, sequence
determined by, 1-20
in RMSCNV, 9-20
Sequential files, 1-19, 1-20
Sequential file, 1-17. See also
Sequential file organization
append records to a, 3-7

Index·-22

Sequential file organization, 1-6, 1-6f,
1-17, 1-17f, 1-37, 1-40
advantages, 2-19t
allocation, calculation of initial, 3-4
applications, 3-1
and block spanning, 1-34, 1-38
characteristics and capabilities, 2-18t
data sizes, maximum, 3-3t
data storage space, 2-5
and Deferred Write, 2-17, 3-14
density, optimal, 1-34. See also
block spanning
disadvantages, 2-20t
end-of-file attribute, 3-2
end-of-file indicators, 3-2t
file-sharing, restrictions on, 2-12
file, creating one from another, 9-19.
See also RMSCNV
find operation, effect of, 3-8
get operation, effect of, 3-9
1/0 unit, 1-35
insertion, physical sequence of, 1-6
medium, 1-37
and Multi-Block Count, 2-19. See
also Multi-Block Count
Multiple Buffers, 3-14
overhead, almost no, 3-1
placement control, 6-14. See also
placement control
put operation, effect of, 3-11
put operation, locate end-of-file before, 3-11
Random Access Mode, no, 1-21
Record Access Streams, multiple, 3-15
and record formats, 1-37
record, deleted by truncate operation, 3-11
records are virtually adjacent, 1-19
records to file, append, 3-7
records, no spaces between, 1-6
space, unused flagged at end of block, 3-2
space, wasting, 3-1
stream record format limited to, 2-16
structure, 3-1, 3-2
truncate operation, size affected by, 3-12
update operation, effect of, 3-12
Sequential record operation, 5-17. See
also record operation
Shared access, 2-5, 2-7, 2-8t, 2-12.
See also file-sharing
Sharing, file. See file-sharing
SIDR, 5-6, 5-11, 5-14, 5-15, 5-16,
5-17, 5-18~ 6-3, 6-8, 6-9. See
also Secondary Index Data Record
Signed integer key, 6-3. See also key
Software data structure, 1-12

Software layers, 1-10, 1-33
SORT-11, 9-58
Sort work devices, reassign, 9-65
Sort work files, 9-58
Source-to-task sequence, 8-2f
Space, 2-4. See also address space
bucket size, 4-·2, 6-16
buffer size, 2-5
duplicate key values, impact of, 6-8
optimization, 8-1
requirement proportional to file organization,
2-5
task size, 2·-5
wasted in Sequential file, 3-1
Spanning, block, 1-38. See also block
spanning
Special account characters, support of, 9-2
Speed, 2-3, 2-4. See also performance
of access, 3-3
bucket size, 4--2, 6-16
hardware, 1-1
optimization, 8-1
SPR, 9-9
Standard ODL files, 8-4, 8-7
Status value, B--4
Storage cell, 1-17. See also record storage cell
Stream. See Record Access Stream
Stream ASCII file, A-3
Stream record format, 2-16. See
also record format, stream
String key, 6-3, 6-4. See also key
STS, 9-20
STV, B-4
Subfile, 1-35
Summary listing, 9-12, 9-68, 9-71. See
also RMSBCK, RMSRST
contents, 9-15, 9-22
data integrity checking, 9-12
error messages, 9-12, 9-68
in file, 9-12, 9-68
file processing, 9-12, 9-68
on terminal, 9-12, 9-68
Surface, 6-14
Swapping, 8--30
/switch, 9-12, 9--21, 9-53, 9-61, 9-69
Switch, 9-13, 9--53, 9-69, D-1
Symbol table file, 8-21
Syntax, description of, 9-9
System protection code, 2-5. See also
operating system, protection code
System, operating. See operating system
T
Tape, 1-17. See also magnetic tape
Target data bucket, 5-10, 5-12

Task, 8-1
design, 3-6, 4-7, 7-1
image, 8-4, 8-21
logical form, 1-29
resources, allocating more to, 8-29, 8-30
segments that can run independently,
8-2. See also overlay
size, 2-5
size, calculating changes in, 8-14
structure, 1-29f, 3-13f, 4-14f, 7-7f, 8-1
virtual address space, 6-16
Task Builder, 1-28, 8-1, 8-2f, C-1
command file, 8-6
concatenates RMS-11 routines with program,
8-2
considerations, 8-25
errors, possible, 8-14
map, 8-14, 8-25
MULTIPLY DEFINES SYMBOL, 8-14
object modules, combines, 8-1
and Resident Libraries, 8-22
Resident Library, speed with, 8-24
RESLIB option, 8-22
and RMS-11 routines, 8-1
UNDEFINED SYMBOLS error message, 8-8,
8-14
Terminal-format file, A-3
Time
application environment effects I/0, 2-3
equal access, 5-13
factors in I/0 operation, 2-4f
uniform random access, 5-1
Track, 1-12, 1-35, 3-6, 4-6, 6-14
disk head can access without changing
position, 1-12
sectors, 1-12
Truncate record operation, 3-11
and Current Record, 1-31, 3-12
end-of-file position, declares, 3-11
and Next Record, 3-12
Sequential file, size of, 3-12
Two-byte signed integer key, 6-4. See also key
Two-byte unsigned binary key, 6-5.
See also key
Types, prints vs., 9-10

u
/UIC switch, 9-6. See also utilities
Undefined record format, 1-40
Unit record device, 1-17, 1-39, 2-18t, 3-2
and RMSCNV, 9-28
Unlock bucket, 2-12
Unsigned binary key, 6-3. See also key
Update record operation, 1-26
Alternate indexes, revise, 5-.15

lndex-23

Update record operation, (cont.)
Alternate Key value changed, 5-15
cost, 5-18t, 6-3
and Current Record, 1-31, 3-12, 4-13,
5-15, 7-6
Deferred Write, effect of, 7-6
duplicate key values, effect of, 6-9
file extension, causes, 1-37
find or get operation, preceded by successful,
1-26, 5-14
I/0 buffer, 1-30, 7-6
Indexed files, effect on, 5-14
key value, check for change in, 6-9
and Move Mode, 3-13, 4-14, 7-7
and Next Record, 3-12, 4-13, 7-6
Primary Key value changed, 5-15
record with revised version, replace a, 1-26
Relative file, effect on, 4-13
Sequential file, effect on, 3-12
Sequential files, restrictions for, 3-12
stream records, processing, 2-17
user buffer, 1-30, 7-6
User
data record, 5-17. See also data
record, record
-defined words, 9-9
interface, utility assumes control,
9-12, 9-21, 9-52, 9-61, 9-68
User buffer, 1-29
during find operation, 3-9, 4-10, 7-3
during get operation, 1-30, 1-31, 5-14
and I/0 buffer, 1-29
during put operation, 1-30, 1-32
and Record Transfer Mode, 1-33, 3-13,
3-14, 4-14, 4-15, 7-6, 7-7
during update operation, 1-30, 7-6
Utilities, 1-10, 1-26, 1-26, 9-1
ANSI label on magtapes, require, 9-2
ANSI, compliance with, 9-10
command string continuation, 9-7
command vs. interactive, 9-5
conventions, 9-4
crash dump, 9-8
damage assessment routine, 9-8
documentation conventions, 9-9
error messages, 9-5, 9-7, E-1
fatal error messages, 9-8
HELP messages, 9-5
? in error messages, 9-8
indirect file, 9-6
installed, 9-12, 9-21, 9-31, 9-52, 9-61, 9-68
installed vs. uninstalled, 9-6
nonfatal error messages, 9-8
patch level, 9-7
prints vs. types, 9-10

Index-24

Utilities, (cont.)
RMS-11 functionality, provide, 9-1
RSTS/E special account characters,
support of, 9-2
syntax, description of, 9-9
tasks, independent, 9-1
/UIC switch, 9-6
uninstalled, 9-12, 9-21, 9-32, 9-52, 9-61, 9-68
user-defined words, 9-9
using, 9-2
Utilities, command, 9-5. See also utilities
Utilities, interactive, 9-5. See also utilities
Utility, installed, 9-6. See also utilities
Utility, uninstalled, 9-6. See also utilities

v
Valid record, 4-9
Value match. See match criteria
Variable-length records, 1-37. See
also record format, variable
Variable record format, 2-15. See
also record format, variable
Variable-with-fixed-control, 2-16.
See also record format, VFC
VAX, 1-14, 3-3
Asynchronous record operation, A-5
restrictions on RMS-11, A-5
RMS-11 restrictions, A-4
VBN. See Virtual Block Number
in RMSDEF, 9-42
Version, 9-12, 9-13, 9-16, 9-21, 9-27,
9-53, 9-61, 9-69, 9-72, 9-73
Version number, current, 9-14, 9-22, 9-54,
9-62, 9-70, D-2
Versions of files, 9-5
Vertical tab, 2-16, 2-17
VFC record format, 2-16. See also
record format, VFC
Virtual address space, 6-16. See also
address space
Virtual block, 1-15, 2-12, 3-1, 4-1, 9-16, 9-72
map to logical block, 2-4. See also
virtual-to-logical-block mapping
Virtual Block Number, 1-15,- 1-33,
1-34, 1-40, 3-2, 6-10, 6-15, 8-26, 9-42
Virtual device, 1-15, 1-40
Virtual-to-logical-block mapping, 1-15f,
3-6, 4-6, 8-1, 8-26
Virtually contiguous blocks, 1-15
Volume, F-1. See also magnetic tape
initialize disk, 8-27
mount disk, 8-27
relative number, 9-42
restoration account and/or, choice of, 9-67
surfaces on a, number of, 6-14

~·

w
Wild card characters, 9-11, 9-12, 9-13, 9-17,
9-21, 9-53, 9-61, 9--62, 9-67, 9-69, 9-73
/WIN switch, 8-27
Window, 8-26
retrieval pointers, maximum number of,
8-28
size adjustment on RSTS/E, A-3
size, increase on IAS/RSX-llM, 8-27
turning, 8-27
Word alignment, 2-5, 3-4

Words, user-defined, 9-9
Work files, sort, 9-58
Write access, 2-12. See also file-sharing
Write-shared files, sequentially reading, 7-10
Write sharing. See file sharing
Write-type operation, 2-12. See also
record operation, write-type

x
XBUF, 8-28, 8-30

Index-25

RMS-11 User's Guide
READER'S COMMENTS

AA-D538A-TC

NOTE: This form is for document comments only. DIGITAL will use comments submitted on this form at the
company's discretion. If you require a written reply and are eligible to receive one under Software
Performance Report (SPR) service, submit your comments on an SPR form.

Did you find this manual understandable, usable, and well-organized? Please make suggestions for improvement.

Did you find errors in this manual? If so, specify the error and the page number.

Please indicate the type of user/reader that you most nearly represent.
D
D
D
D
D
D

Assembly language programmer
Higher-level language programmer
Occasional programmer (experienced)
User with little programming experience
Student programmer
Other (please specify) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

Organization
Street
City ____________________

State - - - - - - Zip C o d e - - - - - - - - or
Country

I
-- - -

-Do Not Tear - Fold Here and Tape -

- - -

- -

- ---1

I-~

~amnoma

111111

No Postage
Necessary

if Mailed in the

United States

BUSINESS REPLY MAIL
FIRST CLASS PERMIT N0.33 MAYNARD MASS.
POSTAGE WILL BE PAID BY ADDRESSEE

I~.
I
I
I

ATTN: Commercial Engineering Publications MK1-2/2H3
DIGITAL EQUIPMENT CORPORATION
CONTINENTAL BOULEVARD
MERRIMACK N.H. 03054

I
I
I
I.~!

:__ -

Do Not Tear - Fold Here and Tape -

I
I

I
I~·

I
I

I~
I