IBM Z Performance and Capacity Analytics

Version 3 Release 1

IMS Performance Feature

Guide and Reference

IBM

SC28-3217-00
 Note
Before using this information and the product it supports, read the information in “Notices” on page
149.

This edition applies to version 3, release 1 of IBM Z Performance and Capacity Analytics (program number 5698-AS3)
and to all subsequent releases and modifications until otherwise indicated in new editions.
Last updated: October 2020
© Copyright International Business Machines Corporation 1994, 2017.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract with
IBM Corp.
© Teracloud S.A. 2018, 2020.
Contents

Figures................................................................................................................ vii

Tables.................................................................................................................. xi

Preface...............................................................................................................xiii
Who should read this book........................................................................................................................xiii
Publications............................................................................................................................................... xiii
Accessing publications online............................................................................................................. xiii
Accessibility............................................................................................................................................... xiii
Support information.................................................................................................................................. xiv
Conventions used in this book.................................................................................................................. xiv
Typeface conventions.......................................................................................................................... xiv
What's new in this edition (October 2020)................................................................................................ xv

Chapter 1. Introducing the IMS Performance Feature............................................. 1

Setting the IMS Application Tracing Facility (ATF) summary record type................................................. 1
Preparing IMS log data for collection.......................................................................................................... 1
Collecting IMS log data................................................................................................................................ 3
Selecting IMS Performance Feature components...................................................................................... 4
The collect components......................................................................................................................... 5
Choosing between standard and extended subcomponents................................................................5
Availability for IMS resources......................................................................................................................6
Updating other lookup and control tables...................................................................................................7
Using the IMS Performance Feature........................................................................................................... 8
How the IMS Performance Feature uses log and record procedures........................................................ 8
The log procedure...................................................................................................................................8
The record procedure...........................................................................................................................15

Chapter 2. Understanding data flow through the IMS Performance Feature........... 17

Log collector data flow.............................................................................................................................. 17
DRLSLOGP data flow..................................................................................................................................18
IMS CSQ light feature data flow................................................................................................................ 19
IMS Performance Feature object definitions............................................................................................ 20

Chapter 3. Administering the IMS Performance Feature........................................21

IMS logs in a shared-queue environment................................................................................................. 21
Using the log collector............................................................................................................................... 21
Input and output data sets.................................................................................................................. 21
Running the log collector..................................................................................................................... 23
Specifying the log collector parameters..............................................................................................24
Specifying log procedure parameters..................................................................................................25
Using DRLSLOGP........................................................................................................................................29
Input and output data sets.................................................................................................................. 29
Running DRLSLOGP..............................................................................................................................30
Specifying DRLSLOGP parameters...................................................................................................... 31
Using the IMS light feature........................................................................................................................ 32
Setting up a load library for the IMS light feature............................................................................... 32
Setting up the IMS light feature collect............................................................................................... 34
Using MAXFREE to allow more pending data to be held.......................................................................... 34

iii
 Using DRLSLOGP on remote systems....................................................................................................... 34
IEC130I issued for SQNLOGS+1............................................................................................................... 34
Using the light feature with IMS Application Tracing Facility (ATF) summary data................................ 35
Using the light feature with HALDB data...................................................................................................35
Using TABLEFLUSH to prevent storage shortage failures........................................................................ 35
Using TABLEFLUSH to prevent performance degradation....................................................................... 35
Using TABLEFLUSH to improve performance........................................................................................... 36
Using TABLEFLUSH to minimize virtual storage use................................................................................ 36
Recovering from abends during collect.................................................................................................... 37
Recovery using the log procedure checkpoint facility........................................................................ 37
Recovery without the checkpoint facility............................................................................................ 37
Additional capabilities............................................................................................................................... 37

Chapter 4. Data table reference........................................................................... 39

Log and record definitions......................................................................................................................... 39
Log definitions...................................................................................................................................... 39
Record definitions................................................................................................................................ 39
Composite and R2 record definitions.................................................................................................. 52
Composite record sections.................................................................................................................. 52
Data tables and lookup tables...................................................................................................................54
Naming standard for tables................................................................................................................. 54
Table descriptions................................................................................................................................ 55
Control tables....................................................................................................................................... 55
IMS Performance Feature data tables...................................................................................................... 55
IMS_AVAILABILITY_D, _W.................................................................................................................. 56
IMS_AVAILABILITY_T..........................................................................................................................57
IMS_CHKPT_IOSAM_T.........................................................................................................................58
IMS_CHKPT_POOLS_T......................................................................................................................... 58
IMS_CHKPT_REGION_T.......................................................................................................................59
IMS_CHKPT_STATS_T..........................................................................................................................59
IMS_CHKPT_VSAM_T........................................................................................................................... 59
IMS_HALDB_OLR_T, _H, _D, _W..........................................................................................................59
IMS_PSB_ACCOUNT_H, _D, _W.......................................................................................................... 60
IMS_PSB_ACCOUNT2_H, _D, _W, _M................................................................................................. 65
IMS_SYSTEM_TRAN_H,_D...................................................................................................................70
IMS_SYSTEM_TRAN2_H,_D, _M..........................................................................................................73
IMS_TRAN_H, _D, _W.......................................................................................................................... 77
IMS_TRAN_QUEUE_Q, _D.................................................................................................................... 82
IMS_TRAN_QUEUE_QV, _DV................................................................................................................84
KPM_IMS_TRAN_H,_D, _W..................................................................................................................85
KPM_IMS_ATF_T.................................................................................................................................. 89
KPM_IMS_ATF_EVT_T......................................................................................................................... 90
Lookup tables.............................................................................................................................................91
IMS_AVAIL_RESOURCE....................................................................................................................... 91
IMS_SYSTEM_NAMES..........................................................................................................................91
Mapping between table fields and CSQ records....................................................................................... 93
IMS_TRAN_x (H, D, W)......................................................................................................................... 93
IMS_TRAN_QUEUE_x (Q, D).................................................................................................................93
IMS_SYSTEM_TRAN_x (H, D)...............................................................................................................94
IMS_PSB_ACCOUNT_x (H, D, W).........................................................................................................94

Chapter 5. Report reference.................................................................................97

Report format and general description..................................................................................................... 97
Report title............................................................................................................................................97
Report ID.............................................................................................................................................. 97
Report group.........................................................................................................................................98
Source...................................................................................................................................................98

iv
 Attributes..............................................................................................................................................98
Variables............................................................................................................................................... 98
Report types......................................................................................................................................... 98
Standard report formats...................................................................................................................... 99
Samples of reports across non-CSQ and CSQ tables............................................................................. 100
IMS Detail reports....................................................................................................................................104
IMS Message Queue Pool Detail by Date report............................................................................... 104
IMS OSAM/ISAM Buffer Pool Detail by Date report.......................................................................... 105
IMS VSAM Buffer Pool Detail by Date report.....................................................................................106
IMS Availability reports........................................................................................................................... 107
IMS CSQ Subsystem Availability, Daily Trend report........................................................................107
IMS CSQ Region Availability, Daily Overview report......................................................................... 108
IMS CSQ Application Usage and Availability report.......................................................................... 109
IMS CSQ Resource Utilization, Daily Overview report...................................................................... 109
IMS CSQ Application Usage and Availability report.......................................................................... 111
IMS CSQ Resource Utilization, Daily Overview report...................................................................... 112
IMS Message Queue reports................................................................................................................... 113
IMS Message Queue Utilization, Date report.................................................................................... 113
IMS Msg Queue Utilization by Transaction, Date report................................................................... 115
IMS Message Queue Utilization Overview, Daily Report...................................................................115
IMS Transaction Arrival Rate and Msg Queue Usage, Daily Trend report........................................ 116
IMS CSQ Transaction Transit Time reports............................................................................................ 117
IMS CSQ Transit Time Analysis By Transaction Name report.......................................................... 117
IMS CSQ Transit Time Analysis By LTERM and Userid report.......................................................... 119
IMS CSQ Transit Time Analysis by Region report............................................................................. 121
IMS CSQ Utilization reports.....................................................................................................................123
IMS CSQ Resource Utilization, Daily Overview report...................................................................... 123
IMS HALDB OLR reports.......................................................................................................................... 125
IMS HALDB OLR Unit Of Reorganization, Details report................................................................... 125
IMS HALDB OLR Statistics report...................................................................................................... 126
Key Performance Metrics IMS reports.................................................................................................... 127
KPM IMS Processing Times by Transaction by Hour report............................................................. 127
KPM IMS Processing Times by PSB Name by Hour report............................................................... 128
KPM IMS Processing Times by Region Type by Hour report............................................................ 129
KPM IMS Average Enqueues by Transaction by Hour report............................................................130
KPM IMS DB Activity by Region by Program Name by Hour report.................................................. 131
KPM IMS ATF Transaction Statistics report...................................................................................... 132
KPM IMS ATF Transaction Events report.......................................................................................... 134
KPM IMS ATF Transaction Database Activity report.........................................................................136

Chapter 6. Customization reference................................................................... 139

Creating IMS log record DSECTs............................................................................................................. 139
Sample archive exit................................................................................................................................. 141
DFSLTMG0 log merge utility....................................................................................................................141
Controlling the log merge...................................................................................................................143
Tailoring example using MSGTEXT for IMS_TRAN_x tables.................................................................. 143
Accessing the extracted information.................................................................................................145

Appendix A. Support information....................................................................... 147

Contacting IBM Support.......................................................................................................................... 147

Notices..............................................................................................................149
Trademarks..............................................................................................................................................150

Bibliography...................................................................................................... 151
IBM Z Performance and Capacity Analytics publications...................................................................... 151
Glossary............................................................................................................ 153

v
 Index................................................................................................................ 157

vi
Figures

1. Preparing IMS log data for collection............................................................................................................2

2. Overview of data collection using the IMS Performance Feature................................................................3

3. Example of IMS Performance Feature log procedure parameter report...................................................12

4. Example of IMS Performance Feature log procedure node statistics report............................................ 13

5. Example of buffer usage statistics for the IMS 07 record buffer...............................................................13

6. Example of TABLEFLUSH processing statistics......................................................................................... 14

7. Example of WRITEPENDING processing messages.................................................................................. 14

8. Example of composite record statistics..................................................................................................... 15

9. IMS CSQ data flow through the log collector............................................................................................. 17

10. Data flow through DRLSLOGP...................................................................................................................18

11. Data flow through IMS CSQ light feature................................................................................................. 19

12. Data flow: Account and Availability subcomponent................................................................................ 20

13. Data flow: Transaction Transit Time subcomponent............................................................................... 20

14. Sample job for running the log collector within IMS CSQ........................................................................ 23

15. Sample job 1 for running DRLSLOGP........................................................................................................30

16. Sample job 2 for running DRLSLOGP........................................................................................................31

17. Tabular reports example...........................................................................................................................99

18. Graphic reports example........................................................................................................................ 100

19. IMSY01 Report Query (DRLQIY01 member)......................................................................................... 101

20. Example of Query from IMS_SYSTEM_TRAN_D CSQ table................................................................... 102

21. Example of mixed query......................................................................................................................... 103

22. Example of query input variables panel DRLDRSEL.............................................................................. 103

23. Example of IMS Message Queue Pool Detail by Date tabular report.................................................... 104

vii
 24. Example of IMS OSAM/ISAM Buffer Pool Detail by Date tabular report...............................................105

25. Example of IMS VSAM Buffer Pool Detail by Date tabular report......................................................... 106

26. Example of an IMS CSQ subsystem Availability, Daily Trend Report....................................................107

27. Example of an IMS CSQ Region Availability, Daily Overview report......................................................108

28. Example of an IMS CSQ application Usage and Availability report.......................................................109

29. Example of an IMS CSQ Resource Utilization, Daily Overview report................................................... 110

30. Example of an IMS CSQ application Usage and Availability report.......................................................111

31. Example of an IMS CSQ Resource Utilization, Daily Overview report................................................... 112

32. Example of IMS Message Queue Utilization, Date report......................................................................114

33. Example of IMS Message Queue Utilization by Transaction, Date report.............................................115

34. Example of an IMS Message Queue Utilization Overview, Daily Report............................................... 116

35. Example of an IMS Transaction Arrival Rate and Message Queue Usage, Daily Trend report.............117

36. Example of IMS CSQ Transit Time Analysis By Transaction Name, Daily report..................................118

37. Example of IMS CSQ Transit Time Analysis By LTERM and Userid, Daily report..................................120

38. Example of IMS CSQ Transit Time Analysis By Region, Daily report.....................................................122

39. Example of IMS CSQ Resource Utilization, Daily Overview report........................................................124

40. Example of an IMS HALDB OLR Unit of Reorganization, Details report................................................ 125

41. Example of an IMS HALDB OLR Statistics report...................................................................................126

42. Example of a KPM IMS Processing Times by Transaction by Hour report............................................ 128

43. Example of a KPM IMS Processing Times by PSB Name by Hour report.............................................. 129

44. Example of a KPM IMS Processing Times by PSB Name by Hour report.............................................. 130

45. Example of a KPM IMS Average Enqueues by Transaction by Hour report.......................................... 131

46. Example of a KPM IMS DB Activity by Region by Program Name by Hour report................................ 132

47. Example of a KPM IMS ATF Transaction Statistics report.....................................................................133

48. Example of a KPM IMS ATF Transaction Events report......................................................................... 135

viii
49. Example of a KPM IMS ATF Transaction Database Activity report....................................................... 136

50. Sample JCL for assembling IMS log record DSECTs..............................................................................139

51. Description from the sample IMS archive exit.......................................................................................141

52. DFSLTMG0 log merge utility................................................................................................................... 142

ix
x
Tables

1. Differences between standard and extended subcomponents...................................................................6

2. Parameter summary for the IMS shared-queue log procedure.................................................................25

3. FLUSHTYPE values...................................................................................................................................... 26

4. Parameter summary for DRLSLOGP........................................................................................................... 31

5. IMS record types and log records component record definitions............................................................. 40

6. Comparison of the IMS Performance Feature with other products for IMS record types........................ 51

7. Composite record sections......................................................................................................................... 53

8. IMS CSQ subcomponent data tables..........................................................................................................55

9. Descriptions of the IMS_SYSTEM_NAMES lookup table........................................................................... 92

10. IMS_TRAN_x (H, D, W) Description.......................................................................................................... 93

11. IMS_TRAN_QUEUE_x (Q, D) Description..................................................................................................93

12. IMS_SYSTEM_TRAN_x (H, D) Description................................................................................................94

13. IMS_PSB_ACCOUNT_x (H, D, W) Description..........................................................................................94

xi
xii
Preface
This book describes how to use IBM Z Performance and Capacity Analytics and the IMS Performance
Feature to collect and report performance data generated by Information Management System (IMS).
This book:
• Helps you understand general performance characteristics
• Describes performance issues and how they affect the level of services you can offer users
• Guides you through the component-selection, installation, and implementation processes
• Explores performance characteristics shown in IBM Z Performance and Capacity Analytics reports so
that you can analyze the characteristics of your system
The terms MVS™, OS/390, and z/OS are used interchangeably throughout this book.

Who should read this book

IMS Performance Feature Guide and Reference is for:
• Anyone who analyzes IMS performance
• Anyone responsible for establishing or meeting enterprise-wide service-level objectives for IMS user
groups
• Anyone who designs, monitors, or tunes IMS or the databases it uses
• IBM Z Performance and Capacity Analytics administrators (primarily as a reference to table and report
definitions)
• Users with various backgrounds who are interested in analyzing IMS performance data and improving
IMS performance
Use this book for guidance in collecting IMS-generated performance data and generating the reports
shipped with the IMS Performance feature. This book explains how to use IBM Z Performance and
Capacity Analytics reports to both understand and evaluate the performance of your systems. It helps you
identify any problems indicated by your data and offers suggestions about how you can monitor, analyze,
and improve IMS performance.

Publications
This section describes how to access the IBM Z Performance and Capacity Analytics publications online.
For a list of publications and related documents, refer to “IBM Z Performance and Capacity Analytics
publications” on page 151.

Accessing publications online

Publications for this and all other IBM products, as they become available and whenever they are
updated, can be viewed on the IBM Knowledge Center website from where you can also download the
associated PDF.
IBM Z Performance and Capacity Analytics V3.1.0
https://www.ibm.com/support/knowledgecenter/SSPNK7_3.1.0
IBM Knowledge Center
https://www.ibm.com/support/knowledgecenter

Accessibility
Accessibility features help users with a physical disability, such as restricted mobility or limited vision, to
use software products successfully. With this product, you can use assistive technologies to hear and

© Copyright IBM Corp. 1994, 2017 xiii

 navigate the interface. You can also use the keyboard instead of the mouse to operate all features of the
graphical user interface.
For additional information, refer to the IBM Accessibility website:
https://www.ibm.com/accessibility

Support information
If you have a problem with your IBM software, you want to resolve it quickly. IBM provides the following
ways for you to obtain the support you need:
• Searching knowledge bases: You can search across a large collection of known problems and
workarounds, Technotes, and other information.
• Obtaining fixes: You can locate the latest fixes that are already available for your product.
• Contacting IBM Software Support: If you still cannot solve your problem, and you need to work with
someone from IBM, you can use a variety of ways to contact IBM Support.
For more information about these ways of resolving problems, see Appendix A, “Support information,” on
page 147.

Conventions used in this book

This guide uses several conventions for special terms and actions, operating system-dependent
commands and paths, and margin graphics.
The following terms are used interchangeably throughout this book:
• MVS, OS/390®, and z/OS.
• VM and z/VM.

Typeface conventions
This guide uses the following typeface conventions:
Bold
• Lowercase commands and mixed case commands that are otherwise difficult to distinguish from
surrounding text
• Interface controls (check boxes, push buttons, radio buttons, spin buttons, fields, folders, icons, list
boxes, items inside list boxes, multicolumn lists, containers, menu choices, menu names, tabs,
property sheets), labels (such as Tip, and Operating system considerations)
• Column headings in a table
• Keywords and parameters in text
Italic
• Citations (titles of books, diskettes, and CDs)
• Words defined in text
• Emphasis of words (words as words)
• Letters as letters
• New terms in text (except in a definition list)
• Variables and values you must provide
Monospace
• Examples and code examples
• File names, programming keywords, and other elements that are difficult to distinguish from
surrounding text
• Message text and prompts addressed to the user

xiv Preface
 • Text that the user must type
• Values for arguments or command options

What's new in this edition (October 2020)

This edition of the IMS Performance Feature Guide and Reference is an update to the previous edition of
the same book, formerly known as the IMS CSQ Feature Guide and Reference.
Changes in this edition include:
• Updated IMS CSQ feature data table: “KPM_IMS_TRAN_H,_D, _W” on page 85 for APAR PH08411
• Updated the following sections for IMS v14 and v15 addressed by APAR PI70730 and PI37652:
– “Selecting IMS Performance Feature components” on page 4
– “The collect components” on page 5
– “Running the log collector” on page 23
– “Running DRLSLOGP” on page 30
– “DRLSLOGP parameter descriptions” on page 32
– Table 4 on page 31
– “Setting up a load library for the IMS light feature” on page 32
– “Setting up the IMS light feature collect” on page 34

Preface xv
xvi IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

Chapter 1. Introducing the IMS Performance Feature

IBM Z Performance and Capacity Analytics is a reporting system that collects performance data logged by
computer systems, summarizes the data, and presents it in a variety of forms for use in systems
management. IBM Z Performance and Capacity Analytics consists of a base product and several optional
features.
The base product includes:
• Interactive System Productivity Facility (ISPF) host reporting and administration dialogs
• The IBM Z Performance and Capacity Analytics log collector program
• Log and record definitions for all records used by the product features
Each feature provides:
• Update definitions for Db2® tables
• Table definitions
• Report definitions
IBM Z Performance and Capacity Analytics enables you to collect large volumes of data and keep the
space to store it at acceptable levels. The database stores all reporting data, which comes from several
sources. For example, logs from System Management Facilities (SMF), Resource Management Facility
(RMF), Customer Information Control System (CICS), and Information Management System (IMS) can be
consolidated into a single report. If you install all components of all IBM Z Performance and Capacity
Analytics features and set system and subsystem data-recording parameters as recommended for each
feature, you can ensure a steady supply of data about the operation of your entire computer center.
The IMS Performance Feature supports data from systems running IMS version 12 (C), release 1, and
later. You use the pre-merged or internally-merged IMS system log data set (SLDS) to generate data for
the predefined tables and reports in the IMS Performance Feature.

Setting the IMS Application Tracing Facility (ATF) summary record type
The IBM Z Performance and Capacity Analytics database collects IMS ATF summary records within the
Key Performance Metrics IMS component. The IMS ATF record type is user-defined. IBM Z Performance
and Capacity Analytics maps these records with a default record type of 160 (X'A0').
If your IMS ATF summary record type is not 160, follow these steps to customize the IBM Z Performance
and Capacity Analytics record definitions before you install the Key Performance Metrics IMS component:
1. Copy the following members to your LOCAL.DEFS data set:

DRLRIB1A (IMS Version 11)

DRLRIC1A (IMS Version 12)
DRLRID1A (IMS Version 13)
DRLRIE1A (IMS Version 14)

2. Edit each member by setting the variable RECTYPE_ATF to your IMS ATF summary record type.

Preparing IMS log data for collection

Before you collect IMS log data into IBM Z Performance and Capacity Analytics Db2 tables, you can
preprocess the log data in various ways (shown in Figure 1 on page 2):

© Copyright IBM Corp. 1994, 2017 1

Introducing the IMS Performance Feature

IMS systems IMS system

(shared queue environment) (non-shared queue environment)

1. IMS logs (SLDSs)

Optional pre-merge for
shared queue logs

2.
DFSLTMG0
(IMS log
merge utility)

(Optional for Remote)

3.
DRLSLOGP

Composite “R2”
records records

4. IBM Z Performance and Capacity Analytics IMS CSQ Performance Feature

Figure 1. Preparing IMS log data for collection

The following list relates to Figure 1 on page 2:

1. SLDS
Every IMS system produces a system log data set (SLDS) during the IMS archive process. The IMS
systems that produce these logs could be members of an IMS shared group, or could be stand-alone
IMS systems. Furthermore, these IMS systems could be local or remote to the z/OS system on which
IBM Z Performance and Capacity Analytics processing occurs.
2. DFSLTMG0
Logs produced by IMS systems in a shared group must be merged for collection. You can use the IMS
utility DFSLTMG0 to merge the logs prior to collection, or you can use the IBM Z Performance and
Capacity Analytics internal merge feature to merge the logs as part of the collection process (see
“Collecting IMS log data” on page 3 for more information).
3. DRLSLOGP
DRLSLOGP is a stand-alone batch program that calls the log and record procedures. It does not
update the Db2 tables. Instead, it produces composite records or R2 records as output.
DRLSLOGP is usually used to obtain a log (either DRLICOMP which is a composite record log, or
DRLIRPT2 which is an R2 record log) smaller than the original IMS log. You can use it as an input log in
a subsequent collect to populate the product tables.
Logs produced on a system that is remote to the z/OS system on which IBM Z Performance and
Capacity Analytics processing will occur must be transported or transmitted to the collection site.
DRLSLOGP can reduce the amount of data that needs to be sent. It gives you the choice of creating
either composite records containing all the information required by IBM Z Performance and Capacity
Analytics for IMS data collection, or R2 records containing transaction-oriented data only.
For more information go to the relevant topics.

2 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

• Data flow diagrams for the log collector and DRLSLOGP, see Chapter 2, “Understanding data flow
through the IMS Performance Feature,” on page 17.
• Running the log collector and DRLSLOGP, see Chapter 3, “Administering the IMS Performance
Feature,” on page 21.
• Collecting R2 records, see “Using the IMS light feature” on page 32.
4. IMS Performance Feature
The IMS CSQ feature can process the following:
• Raw IMS logs from shared groups or from stand-alone IMS systems.
• IMS logs merged by the IMS log merge utility DFSLTMG0.
• Composite, R2, and Transaction Level Statistics records produced at a remote site by the IBM Z
Performance and Capacity Analytics utility DRLSLOGP.

Collecting IMS log data

The process of collecting IMS performance data into Db2 tables is called a collect. Figure 2 on page 3
shows an overview of this process.
1. Input log data Unmerged
logs from
shared
queue Merged
environment log from Log from
shared non-shared DRLSLOGP
queue queue composite DRLSLOGP
environment environment records R2 records

IBM Z
Performance and
Capacity Analytics 3. Log
2. Internal
IMS procedure
merge
Performance (normal
Feature collect)

4. Composite
records

5. Record
procedure

6. R2
records

7. Record
definitions

8. Update
definitions

9. Data 9. Environment
tables information

10. Report 10. Report 10. Report

Figure 2. Overview of data collection using the IMS Performance Feature

The process is as follows:

1. IMS Performance Feature input log data
The IBM Z Performance and Capacity Analytics IMS Performance Feature can process the following:
• Raw IMS logs from shared groups or from stand-alone IMS systems.

Chapter 1. Introducing the IMS Performance Feature 3

Introducing the IMS Performance Feature

• IMS logs merged by the IMS log merge utility DFSLTMG0.

• Composite, R2, and Transaction Level Statistics records produced at a remote site by the IBM Z
Performance and Capacity Analytics utility DRLSLOGP.
All of these possible input sources are presented to IBM Z Performance and Capacity Analytics as
input log data.
2. Internal merge
If the input data to the IBM Z Performance and Capacity Analytics collect consists of logs from IMS
systems in a shared group, and you did not preprocess the logs with the IMS merge utility DFSLTMG0,
then you must use the internal merge feature to merge the logs at the time of collection. Activate the
internal merge by specifying SQNLOGS as an input parameter. For more information about this, see
“Running the log collector” on page 23.
3. IMS Performance Feature log definitions
The IMS CSQ feature provides a log definition for collecting raw IMS records using a log procedure.
The log procedure extracts information from selected log records related to the same event, such as
the processing of a set of related transactions, and produces composite records. Composite records
produced by DRLSLOGP are also processed by the log procedure.
4. Composite records
When the log procedure processes IMS log records, it produces a composite record for each
completed event. When the log procedure processes composite records produced by DRLSLOGP it
simply passes them on for further processing.
5. Record procedure
Composite records containing transaction-specific information are processed by a record procedure.
The record procedure breaks the information in the composite record up into one R2 record for each
transaction.
6. R2 records
Each R2 record contains data for the processing of a single IMS transaction. For a normal collect the
R2 records are created from the composite records produced by the log procedure as part of the
collection process. For a 'light' collect the R2 records produced by DRLSLOGP are used as input.
7. Record definitions
Composite records produced by the log procedure and not further processed by the record procedure
are described by record definitions provided by IBM Z Performance and Capacity Analytics . R2
records produced by the record procedure are also described by record definitions provided by IBM Z
Performance and Capacity Analytics .
8. Update definitions
Update definitions reference fields in the composite and R2 records using their related record
definitions, and update IBM Z Performance and Capacity Analytics data tables.
9. Data tables and environmental information
The IMS Performance Feature uses the composite and R2 records, along with user-supplied data in
lookup tables, to update the data tables. User-supplied data consists of IMS application names, and
period and shift descriptions. The IMS performance data is stored in a series of data tables that are
used when processing data and creating reports.
10. Reports
IBM Z Performance and Capacity Analytics creates reports from the information stored in the data
tables. In addition to the reports provided with the IMS Performance Feature, you can create your
own reports using, for example, the Query Management Facility (QMF) prompted query language.

Selecting IMS Performance Feature components

The IMS CSQ Feature components are:
• IMS 11.1 CSQ collect component
• IMS 12.1 CSQ collect component
• IMS 13.1 CSQ collect component

4 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

• IMS 14.1 CSQ collect component

• IMS 15.1 CSQ collect component
• Key Performance Metrics (KPM) - IMS collect component
In addition to the components specific to CSQ processing, the following collect components are available
for further analysis of IMS log data:
• IMS 11.1 log records component
• IMS 12.1 log records component
• IMS 13.1 log records component
• IMS 14.1 log records component
• IMS 15.1 log records component

The collect components

The collect components are divided into subcomponents. Each subcomponent collects data into Db2
tables and includes predefined reports. The subcomponents are:
Transaction Transit Time
This subcomponent collects information about transactions and BMPs. Information available includes
system response times, system transaction volumes, CPU and database utilization, and transaction
detail.
System Tran Transit Time
This subcomponent collects information about transactions and response times summarized by the
IMS system.
Extended System Transaction Transit Time
This subcomponent collects information about transactions and performance. It is an extension of the
System Tran Transit Time subcomponent with the addition of monthly aggregation. Information
available includes system response times, system transaction volumes, CPU and database utilization,
and transaction detail.
Accounting subcomponent
This subcomponent collects information about resource consumption of IMS systems, regions, and
applications.
Availability subcomponent
This subcomponent collects information about availability data of IMS systems, regions, and
applications.
Extended Accounting
This subcomponent collects information about resource consumption with the addition of monthly
aggregation. It is an extension of the Account part of the Account and Availability subcomponent
(IMS_PSB_ACCOUNT_x).
Statistics
This subcomponent records statistical information about buffer and pool usage.
HALDB OLR
This subcomponent collects information about the High Availability Large Database Online
Reorganization (HALDB OLR) process.

Choosing between standard and extended subcomponents

The main differences between the new extended subcomponents and the standard subcomponents are
as follows:

Chapter 1. Introducing the IMS Performance Feature 5

Introducing the IMS Performance Feature

Note: You cannot use both standard and extended subcomponents.

Table 1. Differences between standard and extended subcomponents

Standard Extended
The IMS_PSB_ACCOUNT_x table contains hourly The IMS_PSB_ACCOUNT2_x table contains
and daily statistics only. monthly statistics and additional key fields,
PERIOD_NAME (name of the period or shift in
which the activity occurred) and MVS_SYSTEM_ID.
The IMS_SYSTEM_TRAN_x tables do not contain Key fields have been reordered in the
the following columns: IMS_SYSTEM_TRAN2_x tables and the following
columns have been added to the key:
• MVS_SYSTEM_ID
• PROGRAM_NAME • MVS_SYSTEM_ID

• TRANSACTION_CLASS • PROGRAM_NAME

• PERFORMANCE_GROUP • TRANSACTION_CLASS
• PERFORMANCE_GROUP

The IMS_SYSTEM_TRAN_x tables do not contain The IMS_SYSTEM_TRAN2_x tables contain the
the PGM_CPU_APPROX column. PGM_CPU_APPROX column. This column
represents the sum of the approximate number of
CPU seconds of program execution time while
transactions are active.

For a description of these tables and columns, see “Data tables and lookup tables” on page 54.

Availability for IMS resources

Availability for IMS resources relates to the following types of resource:
• IMS subsystem
• IMS region
IMS availability is obtained by looking at the subsystem availability and at the real usage of the
applications.
For the IMS subsystem, the following IMS record types are used to track availability:
• Record type X'06': IMS subsystem Start/Stop
• Record type X'4001': IMS system checkpoint
For the IMS region, the following IMS record type is used to track availability:
• Record type X'47': IMS region checkpoint
The following tables are used:
• IMS_AVAIL_RESOURCE
• IMS_AVAILABILITY_T, _D, _W
The IMS_AVAIL_RESOURCE lookup table defines which IMS resources are used for tracking availability.
These values are used during the update of the IMS_AVAILABILITY_T table. It also contains the schedule
names and availability objectives to use for the different resources. These values are used in the
IMS_AVAILABILITY_D and _W tables.

6 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

Updating other lookup and control tables

About this task

The IMS Performance Feature uses the DAY_OF_WEEK and PERIOD_PLAN control tables, which are also
used by other IBM Z Performance and Capacity Analytics features. Check these tables and update them
as needed.
For information about these tables, see the Administration Guide.
If you have installed the z/OS® System (MVS) component of the IBM Z Performance and Capacity
Analytics System Performance Feature, you can use it to collect and report on IMS region activity. This
information can be helpful when you need reports on IMS availability. To obtain this data, you need to
update the MVS_WORKLOAD_TYPE table. For information, see IBM Z Performance and Capacity Analytics
System Performance Feature Reference Volume I.
You can use collect parameters to set the values for the following data table columns:

IMS_SYSTEM_ID - IMS system ID

SYSPLEX_NAME - sysplex name
MVS_SYSTEM_ID - MVS system ID
IMS_CTRL_REGION - IMS control region
IMS_APPLID - VTAM application ID name

However, the specification of only a single value for each of these parameters is allowed, and therefore it
is not advisable to specify these parameters when you are collecting data from multiple IMS systems in a
shared queue environment. In this case you should populate lookup table IMS_SYSTEM_NAMES with your
IMS system IDs.
The following steps describe how IBM Z Performance and Capacity Analytics populates column
IMS_SYSTEM_ID during the collect:
• IBM Z Performance and Capacity Analytics derives IMS_SYSTEM_ID from the IMS records.
• When IMS_SYSTEM_ID cannot be derived from the IMS records, column IMS_SYSTEM_ID is populated
based on the following conditions:
– When IMS_SYSTEM_ID is specified with the SET statement, IMS_SYSTEM_ID is used to populate
column IMS_SYSTEM_ID.
– When IMS_SYSTEM_ID is not specified with the SET statement, column IMS_SYSTEM_ID is set to
'$UNKNOWN'.
The following steps describe how IBM Z Performance and Capacity Analytics populates columns
SYSPLEX_NAME, MVS_SYSTEM_ID, IMS_CTRL_REGION, and IMS_APPLID during the collect:
• IBM Z Performance and Capacity Analytics uses lookup table IMS_SYSTEM_NAMES to obtain values for
these columns.
• When values are not available for these columns in lookup table IMS_SYSTEM_NAMES, these columns
are populated based on the following conditions:
– When values are specified with the SET statement, these values are used to populate these columns.
– When values are not specified with the SET statement, the columns are set to '$UNKNOWN'.
Lookup table IMS_SYSTEM_NAMES contains no default values.
If you are collecting data from multiple systems, IMS_SYSTEM_NAMES must contain information for all of
these systems.
The following sample shows how you can populate the lookup table:

IMS_SYSTEM_ID SYSPLEX_NAME MVS_SYSTEM_ID IMS_CNTL_REGION IMS_APPLID

% ? ? ? ?
IMSA MYPLEX MVS1 REGIONA IMS1
IMSB MYPLEX MVS2 REGIONB IMS2
IMSC MYPLEX MVS3 REGIONC IMS3

Chapter 1. Introducing the IMS Performance Feature 7

Introducing the IMS Performance Feature

For more instructions about using the administration dialog to edit the contents of this lookup table, see
the chapter that explains working with tables and update definitions in the Administration Guide and
Reference.

Using the IMS Performance Feature

About this task

Before using the IMS Performance Feature, verify that IBM Z Performance and Capacity Analytics is
collecting the right data, storing the data correctly, and using the proper data to generate the reports.
Verify also that the IMS_AVAIL_RESOURCE lookup table contains the appropriate definitions. For
information about verifying your installation, see the Administration Guide.
For information about administering IBM Z Performance and Capacity Analytics, see "Part 3.
Administering IBM Z Performance and Capacity Analytics " in the Administration Guide and Reference. For
specific information about running the IMS Performance Feature, see Chapter 3, “Administering the IMS
Performance Feature,” on page 21.

How the IMS Performance Feature uses log and record procedures
This topic explains the use of log and record procedures within the IMS Performance Feature.
The IMS Performance Feature log procedure takes one or more records from a log and creates another
record called a composite record that includes data from the input records. The log procedure defines the
fields taken from each input record and the contents of the output record.
The IMS Performance Feature record procedure processes composite records created by the log
procedure and prepares them for collection using record and update definitions.

The log procedure

The IMS Performance Feature is based on a log processing routine, that is, a log procedure designed to
process selected records from all the IMS systems that are members of the same IMS shared group. The
procedure produces composite records at IMS transaction level (full-function or fast-path), as well as
composite records containing statistical and accounting information. The log procedure copies only the
required IMS log record fields to the composite records.
The information related to a transaction is distributed among numerous log record types. The log
procedure associates information related to a transaction using the IMS shared-queue unit of work (UOW)
key found in many log records. This key is unique through all the IMS logs produced by the systems in an
IMS shared group. The log procedure also uses the IMS log recordrecovery token (RTKN) as a key to
associate related log records.

Unit of work and recovery token keys

Since IMS version 6, log records related to transaction processing have contained a unit of work identifier
(or key) to uniquely identify messages processed on the IMS message queues. The unit of work identifier
is a 34-byte field found in record types X'01', X'03', X'31', X'33', X'35', X'36', X'37', X'38', X'5901',
X'5903', X'5937', and X'5938' with the following format:
Originating system message identifier (bytes 1-16)
The message identifier assigned by the originating IMS system, comprising the IMS system identifier
and a unique token obtained from the system clock. This is referred to by IBM Z Performance and
Capacity Analytics as the originating unit of work (OUOW).
Processing system message identifier (bytes 17-32)
The message identifier assigned by the processing IMS system, comprising the IMS system identifier
and a unique token obtained from the system clock. This is referred to by IBM Z Performance and
Capacity Analytics as the processing unit of work (PUOW).
Flag bytes (bytes 33-34)
Not used by IBM Z Performance and Capacity Analytics.

8 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

The OUOW does not change during the life of the transaction, and is therefore common to all log records
containing a unit of work identifier that are related to a transaction. The PUOW can be different for some
of the records, depending on which IMS system in a shared group processed the message, and the role of
the log record in relation to the transaction.
The IMS log record recovery token represents work performed by a program during a commit interval. The
recovery token is a 16-byte field found in record types X'07', X'08', X'31', X'35', X'56FA', X'5901', X'5903',
X'5937', and X'5938' with the following format:
IMS system identifier
8 bytes.
Schedule count
4 bytes.
Commit count
4 bytes (zero when the program is scheduled and incremented every time the program commits).
The first 12 bytes of the recovery token do not change as long as the program remains scheduled, and are
therefore common to all log records containing a recovery token that are related to the program.

Record grouping
The IMS log records containing transaction-level information are grouped together by the log procedure
using the unit of work and recovery token keys. The groupings are collections of related records that
represent database and data communication activity taking place for the transaction. The groupings are
organized as follows:
OUOW group
This is the primary grouping. All log records containing a UOW identifier related to a primary
transaction will have the same OUOW.
PUOW group
All log records containing a unit of work identifier related to the same message for a transaction will
have the same PUOW. All PUOW groups sharing the same OUOW will be in the same primary grouping.
PSB group
All log records related to program scheduling that contain a recovery token will have the same 12-
byte RTKN. Some log records in PUOW groups (for example the type X'31' of an input PUOW group)
also contain recovery tokens, which are used to relate the PSB group (program) information to the
PUOW group (transaction) information.

Composite records and subtypes

The log procedure produces composite records with a record type controlled by the RECTYPE log
procedure parameter, which has a default value of X'FF'. Each composite record has a subtype that
determines how the record is subsequently processed.
Composite records with subtype X'01' contain transaction-level information and are processed by the
IMS Performance Feature record procedure. The log procedure can write composite records for
incomplete transactions when the TABLEFLUSH or WRITEPENDING log procedure parameters are
specified.
The remaining composite records have subtypes derived from the IMS log record type on which the
composite record is based. These composite records are processed directly by IBM Z Performance and
Capacity Analytics for z/OS using record and update definitions.

Handling of special IMS cases within the IMS CSQ

The IMS Performance Feature can handle the following cases:
Multiple segment input
The log record for the first (or only) segment processed creates input OUOW and PUOW groups. The
log procedure skips log records for subsequent segments.

Chapter 1. Introducing the IMS Performance Feature 9

Introducing the IMS Performance Feature

Single segment input with operator logical Paging output

All the type X'31' log record output segments are linked to the output PUOW group. The log procedure
skips subsequent X'31' segments and does not consider them in the transit time computation.
Multiple segment output
The log record for the first (or only) segment processed creates output PUOW group. The log
procedure skips log records for subsequent segments.
Multiple outputs
Each output creates an output PUOW group. Multiple outputs are valid only when a corresponding
input group is present.
Multiple transactions per schedule of a PSB
The log procedure normally creates the PSB group when a type X'08' log record is processed. The
type X'31' input log record, which contains a PUOW and the RTKN of the PSB group, creates the
connection between an input PUOW group and the corresponding PSB group. The composite record
for a completed transaction is written as soon as all the type X'33' records for the OUOW group are
found, but the PSB group is kept until composite records have been written for all the related
transactions.
Program-to-program switch
Distinctions are made between transactions that started with a type X'01' log record (root
transaction), and those that started with a type X'03' log record that has the MSGQDES flag set to
X'81' (destination is an SMB, indicating a program switch to a child transaction). The log procedure
retains all the OUOW group information for the root transaction until all the information for all the
child transactions is complete, then writes all the transaction information to a single composite
record.
AOI user-exit initiated transactions
The log procedure treats the type X'03' log record corresponding to the AOI user exit like a type X'01'
log record and creates input OUOW and PUOW groups.
Output message reenqueue
When an output message is reenqueued (for example, when IMS finds that the terminal does not
acknowledge successful receipt of a message) IMS will do one of the following:
1. Save the output message (indicated by a type X'36' log record)
2. Reenqueue the same message to the same destination (indicated by a second type X'35' log
record with the reenqueue flag set)
3. Get the unique message from the output queue again (indicated by a second type X'31' log record)
4. Dequeue the message, if the terminal acknowledges successful receipt of the message (indicated
by a second type X'36' log record)
5. Delete the message from the queue (indicated by a type X'33' log record)
The log procedure detects the output message reenqueue condition and captures all the records for
this message. The record procedure extracts the date and time of the first enqueue and the date and
time of the first get unique, allowing the delay to be attributed to the network.
Message-driven BMP programs
These are treated like full-function transactions.
Non-message-driven BMP programs
There are no message queue log records for these programs so they will only be used to populate the
IMS_PSB_ACCOUNT_x tables.
System-generated output (including master terminal operator (MTO) traffic)
The output OUOW and PUOW groups are created when the type X'03' log record is processed.
Terminal message switch
The input OUOW and PUOW groups are created when the type X'01' log record is processed. This
special case may also include MSC and ISC message switching.
Conversational transactions
The type X'11', X'12', and X'13' log records are not supported.

10 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

Fast Path (EMHs)

The EMH OUOW and PUOW groups are created when the type X'5901' log record is processed. The
log procedure matches subsequent type X'5903', X'5936', and X'5937' log records to the EMH entries
using the unit of work as a key. MSGQ output produced by EMH transactions (if any) is linked through
the 16-byte recovery token in the type X'35' Out log record of the output PUOW. Full-function
transactions accessing fast path databases produce type X'5937' log records. These are linked to the
OUOW and PUOW groups for the full function transaction using the recovery token.
Wait-for-input (WFI) programs
If a region is processing wait-for-input (WFI) or pseudo wait-for-input (PWFI) transactions with a high
PROCLIM value the type X'07' program termination log record may not be contained in the current log.
In this situation you could consider reducing the PROCLIM value, or scheduling a new region so that
the type X'07' log records are written more frequently. If the regions and transactions are properly
classed then a 'Quick Reschedule' will occur such that the type X'07' and X'08' log records are created
but the program is never truly terminated and rescheduled. The only alternative is to include all logs
up to the termination of the relevant regions.
Quick reschedule
Quick reschedule is always enabled for transactions with a PROCLIM value greater than zero. It allows
application programs to process more than the PROCLIM number of messages per schedule. Quick
reschedule eliminates the processing overhead of rescheduling and reloading application programs.
IMS uses the process limit count to ensure that no transaction type can monopolize a message region
if other transactions are waiting and are eligible for processing in that region. The process limit count
(PROCLIM or PLC) of a transaction specifies how many waiting messages can be processed after the
program has been scheduled and before IMS assesses whether it should be allowed to continue
(quick reschedule). PROCLIM has relevance only when transactions arrive faster than they are
processed (or with PWFI or WFI), so that a queue of waiting messages builds up. If PROCLIM=0, one
and only one message is processed per program scheduling (no quick reschedule and no PWFI). If
PROCLIM=65565, the number of messages that can be processed per scheduling is unlimited. When
a quick reschedule occurs IMS may not write the type X'08' log record. In this case the log procedure
creates the PSB entry when the type X'07' log record is processed. The type X'08' log record, if it
follows, is matched to the already-created PSB entry.
ISC, MSC, and front-end switching (FES)
The IMS records written in this case are the same as those written for the terminal message switch
case. Therefore, the log procedure treats this case exactly like a terminal message switch. For more
information, see the discussion of terminal message switch on page “Terminal message switch” on
page 10.
CPI-C Saa Driven Application Programs
The CPI-C program termination log record type X'0A07' is processed by the log collector to populate
the IMS_PSB_ACCOUNT_x tables reporting resource consumption. The rows in the
IMS_PSB_ACCOUNT_x tables resulting from The CPI-C Saa driven application program rows in these
tables have the value ‘CPI REGION’ in the PROGRAM_TYPE column. A CPI-driven application program
can send messages to other terminals (either LU 6.2 or non-LU 6.2) or other IMS transactions (either
local or remote) by inserting an alternative PCB, after allocating the appropriate PSB. CPI-C
application programs that run transactions in an MPP region are reported in the IMS_TRAN_x tables
with the value ‘M---S-C-’ in the TRANS_TYPE key column (refer to TRANS_TYPE explanation in the
“TRANS_TYPE key column” on page 81).
Synchronous APPC(OTMA) Conversations
In this case there are no type X'35' Out log records because the output does not get queued for a
synchronous conversation. Instead, at application syncpoint time, IMS APPC code is called under the
dependent region and it performs a GU and SEND for the output message. The log record flow is
X'03'-X'31'-X'33'. The type X'31' log record is a special bypass enqueue GU record, as designated by
the QLGU1NOE bit.

Release dependency
There is a different version of the log procedure and the record procedure for each version of IMS
supported by the IMS Performance Feature. See the IMSVER parameter in “Log procedure parameter
descriptions” on page 26 for more information.

Chapter 1. Introducing the IMS Performance Feature 11

Introducing the IMS Performance Feature

The log procedure interprets log record layouts to determine logic flow. Because these layouts can
change from one IMS release to another, the log procedure uses DSECTs from the relevant IMS release.
Thus, the log procedure is largely release-independent, because the release dependency is not in the
code but in the data definition. The main procedure module invokes the relevant version of a module for
the IMS release specified at runtime. However, if there are changes in an IMS release that alter the log
record DSECTs, an update or new release of the IMS Performance Feature with new code versions of the
modules will be needed to run with the new release of IMS.

Log procedure DRLOUT reports

During normal processing, the log procedure produces informational, warning, and error messages, and
some reports. For information about messages and codes issued by the log procedure, see the topic "IMS
Feature Messages" in the Messages and Problem Determination manual.
The log procedure parameter report (Figure 3 on page 12) shows the parameters in effect for this log
collector run, indicating the parameters specified from the input parameter file DRLIPARM and those that
used the default value.

DRL2071I Parameters used in this run:

-------------------------------------------------------------
Default FLUSH TYPE = NONE
Default FLUSH MESSAGES = NO
Default TABLE FLUSH = NONE
Default MAX FREE POINTERS = 800
Default START = 0000000F 0000000F
Default STOP = 2050365F 2359599F
Default RECTYPE = FF
DRLIPARM WRITEPENDING = YES
Default SQ NLOGS = 1
Default OTMA TRAN CODE = NO
Default SECONDARY TRANSACTIONS = YES
Default IMS 07 RECORD BUFFER SIZE = 200000
Default STATISTIC = YES
Default ACCOUNT = YES
-------------------------------------------------------------

Figure 3. Example of IMS Performance Feature log procedure parameter report

The log procedure node statistics report (Figure 4 on page 13) appears after the log procedure has
completed and provides information about the allocation and usage of "nodes" in internal tables by type.
Nodes are the internal representation of data before it is grouped into composite records.
When the collect process encounters an X'4001' record indicating an IMS COLD start, any incomplete
table entries need to be processed before continuing with new IMS records because there will be no more
information about existing inflight transactions. If WRITEPENDING=YES was specified in DRLIPARM, the
pending nodes are written as incomplete composite records. If WRITEPENDING=NO is specified or
defaulted, the pending data is discarded.
When the collect process finishes, some nodes may be left pending because the information required to
complete the composite record is not present in the log. The log procedure writes the pending nodes to
the checkpoint file allocated to DRLICHKO (if present), which can be used when processing the next log
for the same IMS system. If DRLICHKO is not allocated and WRITEPENDING=YES was specified in
DRLIPARM, then the pending nodes are written as incomplete composite records. If DRLICHKO is not
allocated and WRITEPENDING=NO is specified or defaulted, then the pending data is discarded.
The node statistics report also includes the highest "max free pointer used". If this value is close to the
specified or defaulted DRLIPARM MAXFREE value for the collect, then you should consider using a higher
value. The collect step terminates if the MAXFREE value is reached and more internal table storage needs
to be allocated.

12 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

+--------------------------------------------------------------+
DRL2072I Statistics for NODEs created this run:
--------------------------------------------------------------
NODE NODE Initially Total Total NODEs
Type length allocated allocated allocated pending
(NODEs) (NODEs) (bytes)
--------------------------------------------------------------
OUOW 80 4000 4000 320000 41
PUOW 80 16000 16000 1280000 116
PSB 48 4000 4000 192000 18
IMS 80 80000 80000 6400000 351
SET07BUF 124 20000 20000 2480000 0
--------------------------------------------------------------
TOTALS 124000 124000 10672000 526
--------------------------------------------------------------
Highest Max Free Pointer used = 5
+--------------------------------------------------------------+

Figure 4. Example of IMS Performance Feature log procedure node statistics report

The log procedure SET07BUF buffer usage statistics report Figure 5 on page 13 follows the node
statistics report and provides information about the buffer used to hold IMS type X'07' program
termination log records. This report can be used in conjunction with the SET07BUF figures in the node
statistics report to help choose an appropriate value for the SET07BUF DRLIPARM parameter.

DRL2087I Buffer usage statistics for IMS 07 record buffer Set07Buf.

Maximum number of Set07Buf entries searched = 1
Minimum number of Set07Buf entries searched = 1
Average number of Set07Buf entries searched = 1
Number of times Set07Buf Searched = 68
Number of times 07 record not found = 0
Number of Set07Buf entries used = 94832
Number of Set07Buf hash table slots = 32707
Number of records written to DRLTMP07 = 0

Figure 5. Example of buffer usage statistics for the IMS 07 record buffer

Note: The DRLTMP07 record count only appears if the DRLTMP07 DDname is allocated in the collect JCL.
• If you are not using a DRLTMP07 data set to buffer IMS type X'07' log records:
The most important statistic in the report is the average number of Set07Buf entries searched. If this
value increases and an associated increase in collect CPU time for the same size log is observed, then
you should consider increasing the value for SET07BUF, using the total allocated nodes figure from the
node statistics report as a guide. To be effective in improving performance an increase in the SET07BUF
parameter should be associated with an increase in the reported number of Set07Buf hash table slots.
If the number of times the 07 record is not found is large then the transaction data stored by IBM Z
Performance and Capacity Analytics may be missing program-specific information for an unacceptable
number of transactions. If you are processing your logs in small quantities then you may be able to
reduce this number by processing more logs in each collect process.
• If you are using a DRLTMP07 data set to buffer IMS type X'07' log records:
The same considerations apply to the average number of Set07Buf entries searched figure as for when
DRLTMP07 is not used.
If the number of times the 07 record is not found is large then the transaction data stored by IBM Z
Performance and Capacity Analytics may be missing program-specific information for an unacceptable
number of transactions. In this case you should consider increasing the SET07BUF setting to allow
more type X'07' log record data to be held in memory.
The number of records written to DRLTMP07 can be used to help size the allocation of the DRLTMP07
data set. Note that the entire IMS type X'07' log record is written to DRLTMP07, so you must use the
appropriate record length for the IMS system producing the log when allocating DRLTMP07.
Refer to the description of SET07BUF in Chapter 3, “Administering the IMS Performance Feature,” on
page 21 for more information.

Chapter 1. Introducing the IMS Performance Feature 13

Introducing the IMS Performance Feature

The log procedure table flush processing statistics report (Figure 6 on page 14) follows the SET07BUF
statistics report if table flush processing was requested. It provides information about how often table
flush processing was invoked, and about the data that was flushed from the log procedure internal tables.

DRL2084I TABLEFLUSH processing statistics:

Number of times TABLEFLUSH invoked = 305
Number of times TABLEFLUSH invoked for SOS = 0
Total number of OUOWs flushed = 389
Total number of PUOWs flushed = 230243
Total number of input PUOWs flushed = 230078
Total number of completed input PUOWs flushed = 229812
Average storage freed per table flush = 0.3 Mb

Figure 6. Example of TABLEFLUSH processing statistics

If table flush processing was invoked one or more times for a short-on-storage condition (SOS) you
should consider increasing the amount of virtual storage available to the collect process, increasing the
frequency of table flush processing, or reducing the "age" at which table entries are flushed.
You can use the ratio of "completed input PUOWs flushed" to "input PUOWs flushed" as an indication of
how much data is being affected by the flush process. Table flush processing can produce duplicate
transactions and incomplete transaction records, but if the proportion of completed input PUOWs is high
the impact on the aggregated data stored in the product database may be small enough to be justified by
the associated improvement in collect performance.
When considering the impact of table flush processing, be aware that the table flush statistics only relate
to records written during the flush process. This means that the impact is generally much less than
implied by the ratio between completed and total input PUOWs flushed.
Note: Aggressive table flush settings may result in null values in key columns when the log procedures
are used to collect data into the IBM Z Performance and Capacity Analytics database. Message
DRL0336W is issued for each affected table.
Refer to the description of FLUSHTYPE and TABLEFLUSH in Chapter 3, “Administering the IMS
Performance Feature,” on page 21 for more information.
Log procedure write pending processing messages Figure 7 on page 14 follow the SET07BUF statistics
or the table flush statistics (if present). These messages provide information about the data that was
written from the log procedure internal tables.

DRL2085I WRITEPENDING processing started after 3054628 records.

DRL2086I WRITEPENDING complete: 19 OUOWs written as composite records.
6985 input PUOWs flushed (6981 completed).

Figure 7. Example of WRITEPENDING processing messages

You can use the ratio of "completed" input PUOWs flushed to "input PUOWs flushed" as an indication of
how much of the pending data is incomplete. If the proportion of completed input PUOWs is high the
impact on the aggregated data stored in the IBM Z Performance and Capacity Analytics database may be
small enough to be justify including all the log data by using the write pending process. Alternatively you
can allocate a DRLICHKO data set and preserve the pending data for processing in a subsequent collect
step.
Refer to the description of WRITEPENDING in Chapter 3, “Administering the IMS Performance Feature,”
on page 21 for more information about the write pending process. Refer to the description of the
DRLICHKO and DRLICHKI data sets in Chapter 3, “Administering the IMS Performance Feature,” on page
21 for more information about using the checkpoint facility.
The log procedure composite record statistics report Figure 8 on page 15 follows the SET07BUF
statistics report, the table flush statistics report, or the write pending processing messages, depending on
the combination of DRLIPARM parameters specified. It provides information about the composite records
written by the log procedure.

14 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Introducing the IMS Performance Feature

DRL20731 Statistics for composite records created this run:

---------------------------------------------------------------------------------
Min Max Total
Record Count Lth Lth Bytes
---------------------------------------------------------------------------------
01 Transaction data 5267 404 26564 123507100
07 Program termination 94832 367 367 34803344
08 Program schedule 94832 131 131 12422992
40 Checkpoint 3 1509 1509 4527
45 Checkpoint statistics 104 47 3771 55084
47 Checkpoint region active 49 83 1213 54553
---------------------------------------------------------------------------------
195087 47 26564 170847600
----------------------------------------------------------------------------------

Figure 8. Example of composite record statistics

The record procedure

The IMS Performance Feature log procedure produces composite records containing transaction
information extracted from the raw IMS log records. Each composite record contains information for a
primary transaction and any secondary transactions it initiated. The composite records are processed by
a record procedure to produce one record for each transaction. These transaction records are called R2
records, and are processed by IBM Z Performance and Capacity Analytics record and update definitions to
update the IMS Performance Feature transaction tables.

Chapter 1. Introducing the IMS Performance Feature 15

Introducing the IMS Performance Feature

16 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Understanding data flow through the IMS Performance Feature

Chapter 2. Understanding data flow through the IMS

Performance Feature

This topic describes the flow of data through the IMS Performance Feature. The topic includes:
• “Log collector data flow” on page 17
• “DRLSLOGP data flow” on page 18
• “IMS CSQ light feature data flow” on page 19
• “IMS Performance Feature object definitions” on page 20

Log collector data flow

Figure 9 on page 17 shows the flow of data from the SLDS to the Db2 tables when you use the IMS CSQ
log collector. For more information about running the log collector, see “Running the log collector” on
page 23.
Log collector
Original records

Db2
IMS DRLLOG (1 ... 9) Composite records tables
logs
DRLIMS07 (7)
(1)
R2 records

Parameter
checker Log procedure Record procedure
(3) DRLSInnL DRLSInn2
(2)(5) (6)

DRLICHKO
DRLICHKI
Input Output
checkpoint checkpoint
records records

Checkpoint files
(4)

Figure 9. IMS CSQ data flow through the log collector

The log collector performs the following steps to update the database tables:
1. The IMS logs contain the original data as written by IMS, or the log that was created by merging the
SLDSs written by IMS systems that are members in an IMS shared group. The DRLLOG (1 ... 9)
DDNAMEs point to the data sets you want to process. You can optionally preprocess the IMS log data
sets and extract all the type X'07' log records. These records can then be input to the log procedure via
the DRLIMS07 DDNAME.
2. The log collector calls the log procedure (DRLSInnL, where nn is the level of IMS you are using; for
example, DRLSID1L for IMS version 13, release 1).
3. The log procedure calls a parameter-checking procedure to verify the processing options you selected.
If you do not specify any parameters, the log procedure uses the default parameters. The DDNAME is
DRLIPARM, which points to the parameter data set or contains in-stream parameters.
4. The DRLICHKI and DRLICHKO DDNAMEs are used to hold pending IMS activity. At key commit times
while the log collector is running, the log procedure writes to the checkpoint file referred to by the
DRLICHKO DDNAME. You can use this file to restart in the event of a failure or when processing the
next log. After the log procedure finishes processing, it writes the remaining unmatched transaction
records to DRLICHKO. Allocate the file output to DRLICHKO to DRLICHKI the next time you use the log
procedure.

© Copyright IBM Corp. 1994, 2017 17

Understanding data flow through the IMS Performance Feature

At the start of processing, the log procedure checks for the file referred to by the DRLICHKI DDNAME.
If DRLICHKI is present, it contains unmatched transaction records from the last time the log collector
was run. The log procedure can complete the composite records for these transactions with the new
IMS input log.
5. If the DRLIMS07 file is not used, the log procedure performs a "first pass" over the log data to extract
and buffer the type X'07' log records. If DRLIMS07 is used, the type X'07' log records are read directly
from the file to the buffer. If the DRLTMP07 file is used, the buffer will be loaded with X'07' records up
to the limit set by the SET07BUF parameter, and any additional X'07' records will be written to the
DRLTMP07 file and moved from the file to the buffer at a later stage (when storage is freed). If the
DRLTMP07 file is not allocated, then all the X'07' records are placed in the buffer irrespective of the
SET07BUF setting. In situations where a log contains too many X'07' records to be held in memory,
the COLLECT may fail with a storage related error. When the buffer has been loaded, the log procedure
performs a "second pass" over the log data to extract the information stored by IBM Z Performance
and Capacity Analytics. The log procedure does all of the matching and processing to create the
composite records. It sends the composite records and, optionally, the original records back to the log
collector.
6. The log collector sends the composite records to the record procedure (DRLSInn2, where nn is the
level of IMS you are using; for example, DRLSID12 for IMS version 13, release 1) where the record
procedure simplifies the records for easier collection and reporting.
7. The log collector uses the R2 records from the record procedure and the composite records from the
log procedure to update Db2 tables.

DRLSLOGP data flow

Figure 10 on page 18 shows the flow of data from the SLDS to the output records when you use
DRLSLOGP. For information about running DRLSLOGP, see “Using DRLSLOGP” on page 29.
DRLSLOGP
(8)
Composite records DRLICOMP Composite
IMS records
DRLLOG (1 ... 9)
logs
(1) DRLIRPT2 R2
DRLIMS07
records

Parameter
checker
Log procedure Record procedure
(2)
DRLSInnL DRLSInn2
(3)(6) (7)

DRLICHKO
DRLICHKI
Input Output
checkpoint checkpoint
records records

Checkpoint files
(5)

Figure 10. Data flow through DRLSLOGP

DRLSLOGP creates composite records or R2 records by following this process:

1. The IMS logs contain the original data as written by IMS, or the log that was created by merging the
SLDSs written by IMS systems that are members in an IMS shared group. The DRLLOG (1 ... 9)
DDNAMEs point to the data sets you want to process. You can optionally preprocess the IMS log data
sets and extract all the type X'07' log records. These records can then be input to the log procedure via
the DRLIMS07 DDNAME.

18 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Understanding data flow through the IMS Performance Feature

2. DRLSLOGP calls a parameter-checking procedure to check the processing options you have selected.
If you do not specify any parameters, DRLSLOGP uses the defaults. The DDNAME is DRLIPARM, which
points to the parameter data set or contains in-stream parameters.
3. DRLSLOGP calls the log procedure (DRLSInnL, where nn is the level of IMS you are using; for example,
DRLSID1L for IMS version 13, release 1).
4. The log procedure calls a parameter-checking procedure to verify the processing options you selected.
If you do not specify any parameters, the log procedure uses the default parameters. The DDNAME is
DRLIPARM, which points to the parameter data set or contains in-stream parameters.
5. The DRLICHKI and DRLICHKO DDNAMEs are used to hold pending IMS activity. After the log
procedure finishes processing, it writes the remaining unmatched transaction records to DRLICHKO.
Allocate DRLICHKO to DRLICHKI the next time you use the log procedure.
The log procedure also checks for the existence of checkpoint file DRLICHKI. If DRLICHKI is present, it
contains unmatched transaction records from the last time the log collector was run. The log
procedure can complete the composite records for these transactions with the new IMS input log.
6. If the DRLIMS07 file is not used, the log procedure performs a "first pass" over the log data to extract
and buffer the type X'07' log records. If DRLIMS07 is used, the type X'07' log records are read directly
from the file to the buffer. If the DRLTMP07 file is used, the buffer will be loaded with X'07' records up
to the limit set by the SET07BUF parameter, and any additional X'07' records will be written to the
DRLTMP07 file and moved from the file to the buffer at a later stage (when storage is freed). If the
DRLTMP07 file is not allocated, then all the X'07' records are placed in the buffer irrespective of the
SET07BUF setting. When the buffer has been loaded, the log procedure performs a "second pass" over
the log data to extract the information stored by IBM Z Performance and Capacity Analytics. The log
procedure does all of the matching and processing to create the composite records, and sends the
composite records back to DRLSLOGP.
7. DRLSLOGP sends the composite records to the record procedure (DRLSInn2, where nn is the level of
IMS you are using; for example, DRLSID12 for IMS version 13, release 1) which can simplify the
records for easier collection and reporting. The record procedure writes the output to DDNAME
DRLIRPT2. There could be one or more R2 records produced by the record procedure for the same
composite record. You can specify this output file as disk, tape, or dummy output.
8. DRLSLOGP sends the composite records to the DRLICOMP file, which you can specify as disk, tape, or
dummy output.
9. Depending on the REPORTS options requested, the data sets DRLIRPT2 and DRLICOMP may also
contain Transaction Level Statistics (56FA) records. Additionally, there may be the optional data set
DRLIRPT0, which contains only Transaction Level Statistics (56FA) records.

IMS CSQ light feature data flow

If you use the IMS light feature, the initial data flow at the remote site is the same as for DRLSLOGP (see
“DRLSLOGP data flow” on page 18), from the IMS logs to DRLIRPT2. When you collect the R2 records at
the central IBM Z Performance and Capacity Analytics database site, the data flow is as shown in Figure
11 on page 19.
Log collector

R2 DRLLOG Db2
records (2) tables
(1) (3)

Figure 11. Data flow through IMS CSQ light feature

The light log collector does the following to update the database tables:
1. The R2 input contains the R2 extract records created at the remote site by running DRLSLOGP. The
DRLLOG DDNAME points to the data set you want to process.
2. The log collector uses the IMS CSQ Light log definition to process the R2 records.

Chapter 2. Understanding data flow through the IMS Performance Feature 19

Understanding data flow through the IMS Performance Feature

3. Db2 tables are updated with data from the R2 records.

IMS Performance Feature object definitions

Figure 12 on page 20 and Figure 13 on page 20 show the flow of data through the IMS Performance
Feature objects when you use the log collector. They show the data flow from the original IMS log data to
the IMS CSQ feature reports.
The characters nn in the figure indicate the IMS release, and can be B (IMS version 11), C (IMS version
12), or D (IMS version 13); the characters nnn also indicate the IMS release, and can be B (IMS version
11), C (IMS version 12), or D (IMS version 13).

Log data IMS SLDS, OLDS, Extract,or Merge

Collect component

Vnnn_COLLECT
Log
definition

DRLSInnL
Log
procedure

Vnnn_ST07 Vnnn_ST0A07 Vnnn_ST4001

Record Vnnn_ST06
definitions Vnnn_ST47

Tables and Reports

IMS_AVAIL_RESOURCE
Lookup
tables

Tables IMS_PSB_ACCOUNT_X IMS_AVAILABILITY_X

CSQA01 - CSQA03
Reports

Figure 12. Data flow: Account and Availability subcomponent

Log data IMS SLDS, OLDS, extract, or merge

Collect component
Log Vnnn_COLLECT
definition

Log DRLSInnL
procedure

Record Vnnn_COMP
definitions

DRLSInn2
Record
procedure

Vnnn_R2
Record
definitions

Tables

IMS_TRAN_QUEUE_H IMS_SYSTEM_TRAN_H IMS_TRAN_H

Hourly
tables

IMS_TRAN_QUEUE_D IMS_SYSTEM_TRAN_D IMS_TRAN_D

Daily
tables

Weekly IMS_TRAN_W
tables

Figure 13. Data flow: Transaction Transit Time subcomponent

20 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

Chapter 3. Administering the IMS Performance

Feature

This topic explains how to use the IMS CSQ performance feature to process and collect IMS data. You can
use the log collector program when collecting data into Db2 tables or you can use DRLSLOGP in batch
mode when generating composite records. For more information about these, see Chapter 2,
“Understanding data flow through the IMS Performance Feature,” on page 17.

IMS logs in a shared-queue environment

In a SYSPLEX IMS environment, each IMS system has its own system log data set (SLDS) where
transaction records are logged.IBM Z Performance and Capacity Analytics reads the log records from the
SLDS file, recreates the transaction flow, and evaluates the IMS statistics, saving the data into the Db2
tables of the IMS Performance Feature.
When working in a shared-queue environment, each IMS continues to write its own log so that the records
for a 'unique' transaction can be spread across several SLDS files. IBM Z Performance and Capacity
Analytics needs to use all the SLDS logs to re-create the complete transaction across the IMS shared-
queue environment.
Within an IMS shared-queue environment, IBM Z Performance and Capacity Analytics provides the
following options:
1. Merge all logs using DFSLTMG0, the merge utility provided with IMS. This processes the IMS logs
before the collect is run. The maximum number of logs merged by the IMS utility is nine. For more
information about this, see “DFSLTMG0 log merge utility” on page 141.
2. Utilize the product's internal merge option by specifying the SQNLOGS parameter. This opens all the
logs from the IMS systems defined in the collect JCL and then reads, in turn, the first record of each
log and decides which one is the next record to be processed. The maximum number of logs that can
be merged internally like this is nine. For more information, see “Running the log collector” on page
23.
In all cases IBM Z Performance and Capacity Analytics can only process the logs together if all the IMS
systems are the same version. The log data should all refer to the same time period.

Using the log collector

The log collector runs the IMS Performance Feature log procedure in a IBM Z Performance and Capacity
Analytics environment. It uses a log procedure and record procedure, invokes Db2 and updates Db2
tables.

Input and output data sets

The log collector has the following inputs and outputs (listed here by DDNAME):
DRLLOG - input IMS logs
The input IMS log data. The input log is usually the SLDS or an appropriate extract, but you can use
the OLDS after IMS has closed it. If your IMS systems are running in a shared-queue environment,
merge all the logs produced by each IMS system that is a member of a shared-queue environment, to
build a unique log to use for the collect. The input log can be a data set built by merging the IMS
SLDSs from all the IMS systems that are members of the same IMS shared group. This can be
obtained by using the merge utility DFSLTMG0 provided by IMS or by using the online merging-option
(SQNLOGS parameter). You can also process each log separately but some data, such as timing
information, might become inaccurate. If the online merge option is used, specify DRLLOG1…n.

© Copyright IBM Corp. 1994, 2017 21

Administering the IMS Performance Feature

DRLOUT - output messages

The IMS CSQ performancefeature writes messages to this DDNAME. You can allocate this DDNAME to
SYSOUT, a physical data set, or dummy.
DRLDUMP - output error information
The IMS Performance Feature writes error information to this DDNAME. You can allocate this
DDNAME to SYSOUT, a physical data set, or dummy.
DRLICHKI - input log procedure checkpoint file (optional)
The input data set that contains the status of all pending IMS activities written when the log
procedure completed processing on a previous run. This data set ensures that the IMS CSQ support
can process IMS log data in discrete data set level parts without loss of data. The layout is internal.
You can allocate this DDNAME to DUMMY or to a previously created output log procedure checkpoint
file. Alternatively, you can omit it.
DRLICHKO - output log procedure checkpoint file (optional)
The output data set that records the status of all pending IMS activities when the log procedure
completes processing the current log data. This data set can be processed later by the log procedure,
if it is allocated to the DRLICHKI DDNAME. This data set can be large, for a large IMS system with
many secondary transactions. You can allocate this DDNAME to DUMMY, allocate it to a data set with
record length of 32 756 bytes, or omit it.
The following DDNAMEs are used by the IMS Performance Feature log procedure:
DRLIPARM - input log procedure parameters
This DDNAME points to the parameter data set or contains the in-stream parameters. See “Specifying
log procedure parameters” on page 25 for more information.
DRLIMS07 - extracted X'07' records
IMS shared-queue processing buffers type X'07' records to make them available for processing when
required. IMS type X'07' records are normally read from the IMS logs in the first pass of a two-pass
process. However, an external facility, such as IBM DFSORT, can be used to replace the first pass by
extracting the IMS X'07' records before running the collect job. This can improve performance by
reducing the amount of time to read the logs.
If the IMS X'07' records have been extracted in this way, specify the data set holding these records in
the DRLIMS07 DDNAME of the collect job; the collect then takes the IMS type X'07' records from this
data set.
If you use the DRLIMS07 DDNAME for input you must ensure that it contains only type X'07' records,
and that the records are in the same sequence as they appear in the related IMS logs. If you are
processing multiple IMS logs from different systems, then merge the logs using DFSLTMG0 before
extracting the type X'07' records.
DRLTMP07 - temporary X'07' records
IMS shared-queue processing buffers the type X'07' records to make them available for processing
when required. IMS type X'07' records are normally read from the IMS logs in the first pass of a two-
pass process. In situations where a log contains too many X'07' records to be held in memory, the
COLLECT may fail with a storage related error. To resolve this, the DRLTMP07 file can be used to
contain the 'excess' records. When the DRLTMP07 file is defined as a DDNAME in the JCL, the
SET07BUF parameter will be used to identify the maximum number of records to be stored in
memory. Additional X'07' records over and above the SET07BUF parameter setting will be written to
the DRLTMP07 dataset. During the second stage once an X'07' record is freed from the SET07BUF it
will be replaced by the X'07' from the DRLTMP07 dataset.
To use this DDNAME, allocate it in the JCL with the following attributes:

//* If using Tape include DCB BLKSIZE parameter with a size suitable
//* for your installation.
//DRLTMP07 DD DCB=(LRECL=456,RECFM=VB),
// DSN=&DRLTMP07,
// SPACE=(456,(1000,200),RLSE),AVGREC=K,
// VOL=(,,,3),
// DISP=(NEW,DELETE)

22 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

This sample allocates DRLTMP07 as a temporary data set that is deleted at the end of the collect
step, with up to three volumes requested for allocation.
The logical record length specified (LRECL) must be at least four more than the length of the type
X'07' records in your IMS log. This value is dependent on the IMS version and maintenance level. You
can set the minimum required value to minimize virtual storage use. Alternatively you can specify a
large value like 32756 to cater for all versions of IMS and allow for future changes, at a small cost in
virtual storage use.
The SPACE parameter reflects the number of type X'07' records you expect to process in a typical log.
Note that the logical record length here is only an estimate and does not have to match the DCB
LRECL parameter or exactly match the true record length. The sample allocation anticipates 1 million
IMS V10 type X'07' log records with considerable margin for error in the secondary space parameter.
Given that a typical log comprises between two and five per cent type X'07' records, this allocation is
sufficient for a log containing in the order of 20 million records.

Running the log collector

About this task

You can run an online collect from the administration dialog, or in batch. To run a collect, refer to the
Administration Guide and Reference. .
Figure 14 on page 23 shows an example of how to run an IMS Performance Feature collect in batch.

//USERIDA JOB (ACC000,001),'IMS1 COLLECT',

// NOTIFY=USERIDA,MSGCLASS=X,CLASS=A,REGION=0M
//COLLECT EXEC PGM=DRLPLC,
// PARM=('SYSTEM=DB21',
// 'SYSPREFIX=DRLSYS',
// '&PREFIX=DRL',
// 'SHOWSQL=NO',
// 'SHOWINPUT=NO')
//STEPLIB DD DISP=SHR,DSN= IBM Z Performance and Capacity Analytics load library
// DD DISP=SHR,DSN=Db2 load library
//DRLLOG DD DISP=SHR,DSN=(IMS SLDS or IMSes shared merged log or DRLICOMP)
//*
//DRLICHKI DD DUMMY -- or previously created checkpoint data set
//DRLICHKO DD DUMMY -- or LRECL=32756 output checkpoint data set
//*
//DRLOUT DD SYSOUT=*,DCB=(LRECL=80)
//DRLDUMP DD SYSOUT=*,DCB=(LRECL=32756)
//DRLIPARM DD *
ACCOUNT=YES
MAXFREE=3000
/*
//DRLIN DD DISP=SHR,DSN=USERIDA.IMS.DEFS.V61(DRL$CVAR)
// DD *
COLLECT CSQ_VB10_COLLECT -- for IMS V11
-- or CSQ_VC10_COLLECT for IMS V12
-- or CSQ_VD10_COLLECT for IMS V13
-- or CSQ_VE10_COLLECT for IMS V14
-- or CSQ_VF10_COLLECT for IMS V15
BUFFER SIZE 50 M; -- Appropriate collect buffer size
/*
//

Figure 14. Sample job for running the log collector within IMS CSQ

To run the log collector for logs from different versions of IMS, change the log name specified on the
collect statement.
The DRLLOG DD statement specifies the input IMS log data. If you are using logs from different IMS
systems and you need to open and merge them online (in the same shared-queue group), then use
DRLLOGx by specifying one DDNAME statement for each different IMS system working in the shared-
queue environment.
For example, if you have four IMS working in a shared-queue environment, and you use the internal log
merge, you must specify four different DRLLOGx DD statements, one for each IMS, to describe its logs:

Chapter 3. Administering the IMS Performance Feature 23

Administering the IMS Performance Feature

DRLLOG1 DD DISP=SHR,DSN=…
DRLLOG2 DD DISP=SHR,DSN=…
DRLLOG3 DD DISP=SHR,DSN=…
DRLLOG4 DD DISP=SHR,DSN=…

and set SQNLOGS=4 in the DRLIPARM.

Note: If you are using only one log, you must specify DRLLOG DD and not DRLLOG1 DD.

Specifying the log collector parameters

About this task

When running the log collector, you can specify log collector parameters and COLLECT statements by
using in-stream JCL or a data set allocated to DDNAME DRLIN. The format for specifying these
parameters is as follows:

SET parameter name='value';

For example, the IMS_SYSTEM_ID of IMS1 is specified as follows:

SET IMS_SYSTEM_ID = 'IMS1';

Refer to the Language Guide and Reference for a description of the COLLECT and SET statements.
The following parameters are applicable to the IMS collect:

IMS_SYSTEM_ID - IMS system ID

SYSPLEX_NAME - sysplex name
MVS_SYSTEM_ID - MVS system ID
IMS_CTRL_REGION - IMS control region
IMS_APPLID - VTAM application ID name

You can use IBM Z Performance and Capacity Analytics to specify only a single value for each of these
parameters, and therefore it is not advisable to use these parameters when you are collecting data from
multiple IMS systems in a shared queue environment. In this case you will need to populate lookup table
IMS_SYSTEM_NAMES with your IMS system IDs.
The following steps describe how IBM Z Performance and Capacity Analytics populates column
IMS_SYSTEM_ID during the collect:
• IBM Z Performance and Capacity Analytics derives IMS_SYSTEM_ID from the IMS records.
• When IMS_SYSTEM_ID cannot be derived from the IMS records, column IMS_SYSTEM_ID is populated
based on the following conditions:
– When IMS_SYSTEM_ID is specified with the SET statement, IMS_SYSTEM_ID is used to populate
column IMS_SYSTEM_ID.
– When IMS_SYSTEM_ID is not specified with the SET statement, column IMS_SYSTEM_ID is set to
'$UNKNOWN'.
The following steps describe how IBM Z Performance and Capacity Analytics populates columns
SYSPLEX_NAME, MVS_SYSTEM_ID, IMS_CTRL_REGION, and IMS_APPLID during the collect:
• IBM Z Performance and Capacity Analytics uses lookup table IMS_SYSTEM_NAMES to obtain values for
these columns.
• When values are not available for these columns in lookup table IMS_SYSTEM_NAMES, these columns
are populated based on the following conditions:
– When values are specified with the SET statement, these values are used to populate these columns.
– When values are not specified with the SET statement, the columns are set to '$UNKNOWN'.
Note: To use the IMS_SYSTEM_NAMES lookup table for augmenting IBM Z Performance and Capacity
Analytics data tables, you must collect raw IMS log records, and not the DRLIMS07 data set to input the
preprocessed X'07' records.

24 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

Specifying log procedure parameters

About this task

You can specify parameters to control the operation of the log procedure. Table 2 on page 25 lists these
parameters with a brief description, the default value, and impact on collect performance. For a full
description of the parameters, see “Log procedure parameter descriptions” on page 26.
You can specify the parameters in a data set or in the in-stream JCL for DDNAME DRLIPARM. All
parameters must start in column 1. There must be an equal sign (=) between the parameter and the
value, with no spaces between. Quotation marks, ending colons, or semicolons are not allowed. For
example, ACCOUNT=YES.
Table 2. Parameter summary for the IMS shared-queue log procedure

Parameter name Possible values Default Description Performance impact

ACCOUNT YES YES Account records written to the Low

NO composite record stream.

FLUSHMSGS YES NO Controls whether the log None

NO procedure issues messages
each time table flush
processing is invoked.

FLUSHTYPE SOS, CHKPT CHKPT if Controls when the log High

or COUNT=n, where n TABLEFLUSH procedure invokes table flush
is a whole number specified, otherwise processing.
greater than or none
equal to one.

MAXFREE Whole numbers between 4 800 Number of internal buffer High

and maximum system pointers.
capacity

MSGTEXT start - whole numbers from None Collect data from the Low
1 to 255 MSGXSTXT field of the IMS
X'01' and X'03' records.
length - whole numbers
from 1 to 60

OTMATRANCODE YES NO Source of OTMA transaction Low

NO name.

PASSLOGREC YES YES IMS log records passed as Low

NO output from the log procedure.

PHASE2ENDTIME YES NO Use data from the IMS 5612 Low

NO record.

RECTYPE Valid hex numbers from 00 FF Record type of composite None

to FF records.

SECONDARY YES YES Process child transactions. Low - High

NO (depending on the
number of children
for each transaction)

SET07BUF Positive integer 200000 Number of X'07' records. None

SQNLOGS Integer numbers between 1 Number of logs opened as Medium - High

1 and 9 input. (according to the
number of logs and
record sequences)

START Valid values for yyyy-mm- Start of log IMS log date and time to start None
dd-hh.mm.ss.t processing.

Chapter 3. Administering the IMS Performance Feature 25

Administering the IMS Performance Feature

Table 2. Parameter summary for the IMS shared-queue log procedure (continued)

Parameter name Possible values Default Description Performance impact

STATISTIC YES YES Records written in the Low

NO DRLICOMP data set as
CSQ_Vnn_STxx records.

STOP Valid values for yyyy-mm- End of log IMS log date and time to stop None
dd-hh.mm.ss.t processing.

TABLEFLUSH Non-negative integers 1 if FLUSHTYPE Age limit (in seconds) of table Low - High
specified, otherwise entries. (depending on the log
none contents and
FLUSHTYPE setting)

VERIFY NONE NONE Controls log procedure data None

WARN verification and determines
FAIL whether the log procedure
continues when an issue is
identified.

WRITEPENDING NO NO Pending table entries written as Low

YES output.

Log procedure parameter descriptions

The parameters for the log procedure within an IMS shared-queue environment are:
ACCOUNT=xxx
Specifies that X'06', X'07', X'08', and X'0A07' records will be written to the composite record stream
as CSQ_Vnn_STxx records. If you do not install the Account and Availability subcomponent, set
ACCOUNT=NO. The default is YES.
FLUSHMSGS=xxx
Specifies whether the log procedure issues messages each time table flush processing is invoked. If
table flush processing is requested by the FLUSHTYPE or TABLEFLUSH parameters and
FLUSHMSGS=YES is specified, then the log procedure issues messages at the beginning and end of
each invocation of table flush processing. These messages provide information about the frequency
and effectiveness of table flush processing.
In most cases the information provided in the table flush processing statistics report (message
DRL2084I) is sufficient. The additional level of detail provided by specifying FLUSHMSGS=YES may be
useful for IBM service personnel diagnosing log collection problems, but is not required in normal
circumstances. It can significantly increase the size of DRLOUT if regular flushing is requested with a
small interval using the FLUSHTYPE=COUNT=nnnnnnnn DRLIPARM parameter.
Possible values are YES and NO. The default is NO.
Note: Specifying FLUSHMSGS=NO does not suppress the table flush processing statistics report
(message DRL2084I) produced when the log procedure completes, nor does it suppress the flush
messages produced during write pending processing. Specifying FLUSHMSGS=YES does not cause
table flush processing to be invoked unless FLUSHTYPE or TABLEFLUSH is also specified.
FLUSHTYPE=xxxxx or FLUSHTYPE=COUNT=nnnnnnnn
Specifies the circumstances in which the log procedure invokes table flush processing.
Here are the possible values. The default is CHKPT if TABLEFLUSH is specified. There is no default if
TABLEFLUSH is not specified.

Table 3. FLUSHTYPE values

Value Description
SOS Only invoke table flush processing when a short-on-storage condition occurs. Use this
value to prevent the collect process from terminating when insufficient virtual storage
is available to continue processing the log.

26 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

Table 3. FLUSHTYPE values (continued)

Value Description
CHKPT Invoke table flush processing whenever an IMS checkpoint log record (type x'4001) is
processed, or when a short-on-storage condition occurs. Use this value to minimise
storage overhead and improve performance by invoking table flush processing on a
regular basis determined by the frequency with which system checkpoints are taken in
the IMS system that produced the log being processed.
Count=nnnnnnnn Invoke table flush processing for every nnnnnnnn input log records processed, or
when a short-on-storage condition occurs. Use this value to minimise storage
overhead and improve performance by invoking table flush processing on a regular
basis independent of the frequency with which system checkpoints are taken in the
IMS system that produced the log being processed.

Refer to the description of the TABLEFLUSH parameter for more information about table flush
processing.
MAXFREE=nnnn
Specifies a limit on the number of internal pointers that will be used to address buffers for holding
incomplete transactions. If message DRL2021W is issued indicating a buffer array pointer shortage,
increase the value of MAXFREE in 50% increments until the problem is resolved. Possible values are
from 4 to the maximum allowed by available virtual storage. The default is 800. Specifying a small
value for MAXFREE is not recommended because only buffer pointer storage is allocated based on
this parameter. The transaction data buffer storage addressed by the pointers is not allocated until it
is required.
MSGTEXT=start, length
Optional. Enables up to 60 bytes of contiguous data to be collected from the MSGXSTXT field of the
IMS type X'01' and X'03' input log records. The default is to not collect data from the MSGXSTXT field.
start is the starting position of the data to be extracted from the MSGXSTXT field. The maximum value
for start is 255 and the minimum is 1.
length is the length of the date to be extracted from the MSGXSTXT field. The maximum value for
length is 60 and the minimum is 1.
To store the extracted data into Db2 you will need to customize your IMS feature tables. For an
example see “Tailoring example using MSGTEXT for IMS_TRAN_x tables” on page 143.
OTMATRANCODE=xxx
Specifies whether the transaction name for OTMA transactions should be taken from OTMA point of
view (LUY_TRANCODE of APPC SECTION) or from IMS point of view (MSGODSTN). The default value is
NO, meaning that the transaction name is equal to MSGODSTN.
PASSLOGREC=xxx
Specifies whether the original IMS log records are passed as output from the log procedure. Possible
values are YES and NO. The default is YES.
Specify NO if you:
• Do not collect logs that contain composite records.
• Do not want to use your own update definitions based on IMS log records.
PHASE2ENDTIME=xxx
Optional. Specifies whether data from the IMS X'5612' log records is collected into the composite and
R2 records. Possible values are YES and NO. The default is NO. Specify YES if you want to collect data
from the IMS X'5612' records.
This parameter causes IBM Z Performance and Capacity Analytics to collect data from the IMS
X'5612' records (External Subsystem - End of Phase 2 Syncpoint (DBCTL)) and store it in the
composite record for the unit of work. If available, the timestamp from these records is used when

Chapter 3. Administering the IMS Performance Feature 27

Administering the IMS Performance Feature

calculating the value for DRLMPROCE in the R2 record. This field is used to set the PROCESS_SEC
column in the IMS_TRAN_H, IMS_SYSTEM_TRAN_H and IMS_SYSTEM_TRAN2_H tables.
Note: If DRLSLOGP is run with PHASE2ENDTIME=YES and the composite or R2 records output from
DRLSLOGP are subsequently input to DRLPLC, specifying PHASE2ENDTIME=NO on the DRLPLC job,
will not prevent the IMS X'5612' data collected by DRLSLOGP from being used in the calculation of
the process_sec value.
RECTYPE=nn
Specifies the record type of composite records. For example, if you want the composite records to
have type X'FA', specify RECTYPE=FA. The default is FF.
SECONDARY=xxx
Specifies whether or not secondary (program switch) transactions are to be processed during the
collect. The default is YES. When SECONDARY is set to NO, the following applies:
• Secondary transactions are not processed by the log procedure. Therefore, they are not included in
the composite record and the transaction tables are not populated by secondary transaction data.
• Any outputs from secondary transactions that cannot be distinguished from the outputs of primary
transactions are processed.
The IMS_PSB_ACCOUNT tables, which are populated by X'08' and X'07' records, are not affected by
any exclusion. Hence, they continue to be populated by PSB cumulative data for primary and
secondary transactions, even if SECONDARY=NO is specified.
SET07BUF=nnnnnn
Specifies the number of type X'07' records to be buffered in the first pass over the IMS log data set. In
the ideal case, this number should be large enough for the buffer to hold all type X'07' records found
in the input log data set. An appropriate value is typically in the range from two to five per cent of the
number of records in the input log data set. The default is 200000. When the DRLTMP07 file is
allocated in the COLLECT JCL, the SET07BUF setting specifies the maximum number of X'07' records
stored in memory, and any X'07' records over and above this setting are stored temporarily in the
DRLTMP07 file.
If the DRLTMP07 file is not allocated in the COLLECT JCL, the SET07BUF is used as an indicator of the
expected number of X'07' records and is used to improve the performance of memory allocation. In
this case the SET07BUF setting does not limit the number of X'07' records in the buffer. Thus, if a log
has more X'07' records than is specified in the SET07BUF parameter, the log procedure will attempt
to allocate a buffer large enough to hold all the X'07' records.
SQNLOGS=n
Specifies the number of IMS logs to be opened for input. This number should match the number of
DRLLOG(n) DD statements specified in the collect or DRLSLOGP job. Valid values are between 1 and 9.
The default is 1.
START=yyyy-mm-dd-hh.mm.ss.t
The transaction date and time starting point for processing, where yyyy-mm-dd-hh.mm.ss.t is the
year, month, day, hour minute, second, and tenth of a second timestamp. The default is the beginning
of the log.
STATISTIC=xxx
Specifies that X'45' records will be written into the composite record stream as CSQ_Vnn_STxx
records. If you do not install the Statistics subcomponent, set STATISTIC=NO. The default is YES.
Note: Record X'4001' is not filterable and X'47' is filterable only by specifying both ACCOUNT=NO and
STATISTIC=NO.
STOP=yyyy-mm-dd-hh.mm.ss.t
The transaction date and time stopping point for processing, where yyyy-mm-dd-hh.mm.ss.t is the
year, month, day, hour, minute, second, and tenth of a second timestamp. The default is the end of
the log.

28 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

TABLEFLUSH=nnnn
Specifies how many seconds an entry can remain in log procedure internal tables. The log procedure
removes entries older than this limit from the tables when the specified or defaulted FLUSHTYPE
conditions are met.
For example, if you specify TABLEFLUSH=5 and flush processing is invoked, transaction entries that
started more than 5 seconds before the date, and time found in the last processed relevant record,
are removed from the tables.
If you do not specify FLUSHTYPE or a TABLEFLUSH value, table entries accumulate until transaction
sets are complete, and the tables are not flushed. This can cause excessive virtual storage use and/or
poor performance if transaction sets become large, or collect process termination if a short-on-
storage condition occurs.
Possible values are non-negative integers. The default is 1 if FLUSHTYPE is specified. There is no
default if FLUSHTYPE is not specified.
VERIFY=xxxx
Specifies whether the log collector should stop or continue if a mismatch occurs between the log
procedure and the IMS log. The default is NONE. When VERIFY is set to WARN or FAIL, the following
is checked:
• Attempt to ensure all logs input to the IMS log collector in a single run are for the same release of
IMS.
• Attempt to confirm that the version of IMS that a log is produced by is the same as the version
specified or defaulted for the collect job.
• Identify all the IMS originating and processing systems participating in the shared queue, and
produce a message if the log data set shows characteristics of not being merged properly.
If the VERIFY parameter is set to WARN, messages are produced and processing continues. If the
VERIFY parameter is set to FAIL, messages are produced and processing terminates.
If the VERIFY parameter is set (or defaults) to NONE, no verification is performed.
WRITEPENDING=xxx
Specifies whether the log procedure writes incomplete table entries to an output record. This can
occur when an X'4001' record found on the input log file indicates an IMS COLD start or at end-of-
job. If WRITEPENDING=NO is specified or defaulted, any pending data during COLD start processing is
discarded. If the DRLICHKO DDNAME is present, the log procedure ignores the WRITEPENDING
parameter at end-of-job to avoid the possibility of duplication.
Possible values are NO and YES. The default is NO.

Using DRLSLOGP
DRLSLOGP is a stand-alone batch program that you can use to run the IMS Performance Feature log
procedure in a non-IBM Z Performance and Capacity Analytics environment. DRLSLOGP uses the same
log procedure and record procedure as the log collector, except that Db2 is never invoked and Db2 tables
are not updated.
You can use DRLSLOGP to extract data from IMS SLDSs in your non-IBM Z Performance and Capacity
Analytics environment and then send the output to the main center where Db2 and IBM Z Performance
and Capacity Analytics are running. You can then populate the tables using either the standard collect (for
collecting composite records) or the IMS light feature (for collecting R2 records).
For information on how to build and distribute DRLSLOGP, see “Using the IMS light feature” on page 32.

Input and output data sets

DRLSLOGP uses the same data sets as the log collector described in “Input and output data sets” on page
21. You should also specify one of the following additional output data sets to produce summarized data
for subsequent collection into a IBM Z Performance and Capacity Analytics Db2 database.

Chapter 3. Administering the IMS Performance Feature 29

Administering the IMS Performance Feature

Note: As the DRLIRPT2 output is effectively a subset of DRLICOMP, you would normally allocate only one
of these DDNAMEs to an output data set.
DRLICOMP — composite record output
The IMS Performance Feature writes composite records containing all the information required to
update the IBM Z Performance and Capacity Analytics Db2 database to the DRLICOMP DDNAME.
Allocate DRLICOMP to a variable record length data set with maximum record length 32756 if you
want to save the composite records, or allocate it to DD DUMMY if you want to discard them.
Optionally by specifying DRLIPARM parameter REPORTS=R0(56) you can also collect the transaction
level statistics (56FA) records into DRLICOMP.
DRLIRPT2 — R2 (transaction) record output
The IMS Performance Feature writes R2 records containing only transaction data required to update
the IBM Z Performance and Capacity Analytics Db2 database using the light collect to the DRLIRPT2
DDNAME. Allocate DRLIRPT2 to a fixed record length data set with record length 246 if you want to
save the R2 records, or allocate it to DD DUMMY if you want to discard them. To write the R2 records
to DRLIRPT2 you must also specify the REPORTS=R2(nn) DRLIPARM statement.
Optionally, by specifying REPORTS=R2(56,FF), you can also collect the transaction level statistics
(56FA) records. If you do collect the 56FA data, the record length of the DRLIRPT2 data set must be
changed to a minimum of 536.
DRLIRPT0
This is an optional data set with a minimum record length of 536. This data set is used in conjunction
with the REPORTS=R0(56) DRLIPARM parameter. Using this parameter will cause the transaction
level statistics (56FA) records to be collected to the data sets DRLIRPT0 and DRLICOMP. If the DD for
DRLIRPT0 is set to DUMMY, the transaction level statistics (56FA) records will be collected to the
DRLICOMP data set only. The composite records will continue to be collected to the DRLICOMP data
set.

Running DRLSLOGP

About this task

DRLSLOGP reads log records from the input IMS log and invokes the IMS log procedure for each of them.
It replicates the log collector functions and maintains the same interface with the log procedure.
DRLSLOGP output is written to DRLICOMP and DRLIRPT2.
Two sample jobs that run DRLSLOGP follow. The job in Figure 15 on page 30 collects a merged or non-
shared queue log from DDNAME DRLLOG. The job in Figure 16 on page 31 collects unmerged shared
queue logs from DDNAME DRLLOGn.

//USERIDA JOB (ACC000,001),'DRLSLOGP',

// NOTIFY=USERIDA,MSGCLASS=X,CLASS=A,REGION=0M
//*
//DRLSLOGP EXEC PGM=DRLSLOGP
//STEPLIB DD DISP=SHR,DSN=IBM Z Performance and Capacity Analytics load library
//*
//DRLLOG DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS
//*
//DRLICOMP DD DUMMY -- or LRECL=32756 output composite data set
//DRLICHKI DD DUMMY -- or previously created checkpoint data set
//DRLICHKO DD DUMMY -- or LRECL=32756 output checkpoint data set
//*
//DRLOUT DD SYSOUT=*,DCB=(LRECL=80)
//DRLDUMP DD SYSOUT=*,DCB=(LRECL=32756)
//DRLIRPT2 DD SYSOUT=*,DCB=(LRECL=536)
//DRLIPARM DD *
IMSVER=B1 -- or IMSVER=C1 (IMS version 12.1)
-- or IMSVER=D1 (IMS version 13.1)
-- or IMSVER=E1 (IMS version 14.1)
-- or IMSVER=F1 (IMS version 15.1)
REPORTS=R2(FF)
//

Figure 15. Sample job 1 for running DRLSLOGP

30 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

//DRLSLOGP JOB (ACC000,001),'DRLSLOGP',

// NOTIFY=USERIDA,MSGCLASS=X,CLASS=A,REGION=0M
//DRLSLOGP EXEC PGM=DRLSLOGP
//STEPLIB DD DISP=SHR,DSN=IBM Z Performance and Capacity Analytics load library
//DRLLOG1 DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS1
//DRLLOG2 DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS2
//DRLLOG3 DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS3
//DRLLOG4 DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS4
//DRLLOG5 DD DISP=SHR,DSN=IMS SLDS/OLDS extract log from IMS5
//DRLICOMP DD DUMMY -- or LRECL=32756 output composite data set
//DRLICHKO DD DUMMY -- or previously created checkpoint data set
//DRLICHKI DD DUMMY -- or LRECL=32756 output checkpoint data set
//*
//DRLOUT DD SYSOUT=*,DCB=(RECFM=F,LRECL=80)
//DRLDUMP DD SYSOUT=*,DCB=(RECFM=VB,LRECL=32756)
//DRLIRPT2 DD SYSOUT=*,DCB=(RECFM=VB,LRECL=536)
//DRLIPARM DD *
SQNLOGS=5
REPORTS=R2(FF)
IMSVER=B1 -- or IMSVER=C1 (IMS version 12.1)
-- or IMSVER=D1 (IMS version 13.1)
-- or IMSVER=E1 (IMS version 14.1)
-- or IMSVER=F1 (IMS version 15.1)
//

Figure 16. Sample job 2 for running DRLSLOGP

The DRLLOG DD statement specifies the input IMS log data. If there is a single IMS log, specify the
SQNLOGS parameter as 1 or let it default. If you are using logs from different IMS systems as input and
you want to merge them internally, then specify DRLLOG1 DD, DRLLOG2 DD, DRLLOG3 DD, …DRLLOGn
DD, specifying one statement for each log.
For example, if you have four IMS systems in the shared-queue environment, and you use the internal log
merge, you need to specify four different DRLLOGx DD statements, one for each IMS system, to define its
logs:

DRLLOG1 DD DISP=SHR,DSN=…
DRLLOG2 DD DISP=SHR,DSN=…
DRLLOG3 DD DISP=SHR,DSN=…
DRLLOG4 DD DISP=SHR,DSN=…

and set SQNLOGS=4 in the DRLIPARM.

Note: If you are using only one log, you must specify DRLLOG DD and you cannot specify DRLLOG1 DD.

Specifying DRLSLOGP parameters

You can specify parameters to control the operation of DRLSLOGP. Table 4 on page 31 lists these
parameters with a brief description, the default value, and impact on collect performance.
You can specify the parameters in a data set or in the in-stream JCL for DDNAME DRLIPARM. All
parameters must start in column 1. There must be an equal sign (=) between the parameter and the
value, with no spaces between. No quotation marks, ending colons, or semicolons are allowed. For
example, IMSVER for IMS version13, release 1 is specified as IMSVER=D1.
Table 4. Parameter summary for DRLSLOGP

Parameter name Possible values Default Description Performance impact

IMSVER B1, C1, D1, E1, F1 D1 IMS version and release number None
(B1 indicates IMS version 11.1,
C1 indicates IMS Version 12.1,
D1 indicates IMS Version 13.1,
E1 indicates IMS Version 14.1,
and F1 indicates IMS Version
15.1).

REPORTS R2(FF) None (do not produce R2 Requests record procedure None
records) processing and specifies
composite record type.

Chapter 3. Administering the IMS Performance Feature 31

Administering the IMS Performance Feature

Table 4. Parameter summary for DRLSLOGP (continued)

Parameter name Possible values Default Description Performance impact

R2(56,FF) None (do not produce R2 Output both R2 and 56FA records None
and 56FA records) to DRLIRPT2.
R0(56) None (do not produce Output 56FA records to None
56FA records) DRLIRPT0 and DRLICOMP.

Note: All parameters in Table 2 on page 25 also apply here because DRLSLOGP runs the log procedure.

DRLSLOGP parameter descriptions

The parameters for DRLSLOGP are:
IMSVER=nn
Specifies which release of IMS log data you are using as input, where nn represents the version and
release number. For example, if you are using IMS version 13, release 1 data, specify IMSVER=D1.
There is no default value. If you do not specify this parameter, message DRL204E will be issued and
the program terminated.
REPORTS=R2(FF)
Specifies whether the record procedures should be run to produce R2 records. R2 identifies the
record procedure and FF identifies the composite record type. The R2 records are written to the
ddname DRLIRPT2.
REPORTS=R2(56,FF) or REPORTS=R0(56)
The 56 parameter specifies that the '56FA'x log records are collected to DRLIRPT2 , DRLIRPT0, or
DRLICOMP data sets. These records are used for the KMP IMS collect component..

Using the IMS light feature

The IMS light feature collects transaction information only.
By using the DRLSLOGP utility as a stand-alone procedure in the remote location, the R2 records are
produced from the complete IMS log. The records (reduced by around 90% compared to the original log
size) can then be transferred to the central location where the final collection process occurs, instead of
transferring the complete IMS log. This process applies only to the transaction transit time component.
To use this process, read the following sections:
1. “Setting up a load library for the IMS light feature” on page 32
2. “Setting up the IMS light feature collect” on page 34

Setting up a load library for the IMS light feature

About this task

You must build and redistribute the load modules library in order to run job DRLSLOGP stand-alone in the
remote locations where IBM Z Performance and Capacity Analytics is not installed. To build this library at
the central location where the product is installed, you can customize and use the sample job DRLJCSQL
in the product SDRLCNTL library (see the following JCL).

//DRLJCSQL JOB (ACCT£), ’IMS SQ LIGHT LIB’

//******************************************************************
//* NAME: DRLJCSQL *
//* *
//* FUNCTION: *
//* PREPARE LOAD LIBRARY TO RUN STANDALONE THE DRLSLOGP IMS LOG*
//* PROCEDURE. *
//* *
//* NOTES: *
//* BEFORE YOU SUBMIT THE JOB: *
//* - CHECK ALL DATA SET NAMES IN LOWER CASE (INLIB, OUTLIB DD*
//* STATEMENTS). *
//* - CHECK ALL THE PARMATERS CAREFULLY (DISP, VOL NAMES IN *
//* INLIB AND OUTLIB DD STATEMENTS). *

32 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

//* INLIB CONTAINS THE IZPCA SMP INSTALLED LIBRARY. *

//* OUTLIB CONTAINS THE OUTPUT COPY LOAD LIBRARY *
//* - DECIDE WHETHER SMP DISTRIBUTION ON TARGET LIBRARIES ARE *
//* SUITABLE TO BE USED AS SOURCE AND UPDATE INLIB DSN *
//* ACCORDINGLY (SDRLLOAD OR ADRLLOAD) *
//* - DELETE THE SELECT STATEMENT ROW CORRESPONDING TO THE *
//* IMS RELEASE YOU ARE NOT INTERESTED IN, IF ANY: *
//* DRLSIB1L,... FOR IMS v11r1m0 *
//* DRLSIC1L,... FOR IMS v12r1m0 *
//* DRLSID1L,... FOR IMS v13r1m0 *
//* DRLSID1L,... FOR IMS V14r1m0 *
//* DRLSIF1L,... FOR IMS v15r1m0 *
//* - DOUBLE CHECK IN THE JOB OUTPUT THAT ALL THE NEEDED *
//* LOAD MODULES HAVE BEEN CORRECTLY COPIED. *
//*******************************************************************
//STEP1 EXEC PGM=IEBCOPY
//SYSPRINT DD SYSOUT=A
//INLIB DD DISP=(SHR,KEEP) ,UNIT=3390, VOL=SER=yyyyy,
// DSN=drl190.SDRLLOAD
//OUTLIB DD DSN=drl190.CSQLIGHT.LOAD,DISP=(NEW,CATLG),
// SPACE=(CYL,(1,1,5)),UNIT=3390,VOL=SER=zzzzzz,
// DCB=(RECFM=U,BLKSIZE=32760,LRECL=0)
//SYSIN DD *
COPY I=INLIB,O=OUTLIB
SELECT MEMBER=(DRLSLOGP, DRL2LOGJ, DRL2CSQJ, DRLPIO24, DRLPMSG)
SELECT MEMBER=(DRLSIB1L, DRLSIB12, DRLSIB1C)
SELECT MEMBER=(DRLSIC1L, DRLSIC12, DRLSIC1C)
SELECT MEMBER=(DRLSID1L, DRLSID12, DRLSID1C)
SELECT MEMBER=(DRLSIE1L, DRLSIE12, DRLSIE1C)
SELECT MEMBER=(DRLSIF1L, DRLSIF12, DRLSIF1C)
/*

INLIB
The input data set points to the SMP/E installation target load library (SDRLLOAD) from which the
required load modules are copied.
OUTLIB
The output data set points to the copy output LOADLIB. It is used in the remote locations as STEPLIB
in the DRLSLOGP job.
If you are interested only in a specific version of IMS, the SELECT MEMBER statements related to the
other IMS versions (DRLSIxxL, DRLSIxx2 and DRLSIxxC) can be deleted from the job.
After you have built the CSQLIGHT.LOAD library, distribute it to the remote centers where it is required.
Ensure that the STEPLIB in the DRLSLOGP job used in the remote centers is updated accordingly. For
example:

//DRLILOGP JOB (ACCT#),'DRLSLOGP IMS'

//******************************************************************
//* NAME: DRLILOGP *
//* *
//* FUNCTION: *
//* RUN THE IMS CSQ LOG PROCEDURE STANDALONE *
//* *
//* NOTES: *
//* BEFORE YOU SUBMIT THE JOB: *
//* - CHECK ALL DATA SET NAMES. *
//* - CHECK ALL THE PARAMETERS CAREFULLY. *
//******************************************************************
//DRLSLOGP EXEC PGM=DRLSLOGP,PARM=('SYSTEM=DSN SYSPREFIX=DRLSYS')
//STEPLIB DD DISP=SHR,DSN=drlxxx.CSQLIGHT.LOAD
//DRLLOG DD DISP=SHR,DSN=... INPUT IMS SLDS LOG
//DRLICOMP DD DISP=SHR,DSN=... OUTPUT COMPOSITE RECORDS OR DUMMY
//DRLICHKI DD DISP=SHR,DSN=... INPUT IMS CHECKPOINT FILE OR DUMMY
//DRLICHKO DD DISP=SHR,DSN=... OUTPUT IMS CHECKPOINT FILE OR DUMMY
//DRLOUT DD SYSOUT=*,DCB=(RECFM=F,LRECL=80)
//DRLDUMP DD SYSOUT=*,DCB=(RECFM=VB,LRECL=32756)
//DRLIRPT2 DD DISP=SHR,DSN=... OUTPUT RECORDS FROM R2 REPORT/PROCEDURE
//DRLIPARM DD *
IMSVER=D -- THIS IS THE DEFAULT IMS RELEASE TO BE PROCESSED
REPORTS=R2(FF) -- CALL THE R2 REPORT/PROCEDURE
/*

Chapter 3. Administering the IMS Performance Feature 33

Administering the IMS Performance Feature

Setting up the IMS light feature collect

About this task

To run the IMS light feature in the central location where IBM Z Performance and Capacity Analytics is
installed, you must have installed the required IMS Performance Feature subcomponents. You can then
collect the R2 records produced by job DRLSLOGP by running a collect for one of the following logs:
• COLLECT CSQ_VB11_COLLECL
• COLLECT CSQ_VC11_COLLECL
• COLLECT CSQ_VD11_COLLECL
• COLLECT CSQ_VE10_COLLECL
• COLLECT CSQ_VF10_COLLECL

Using MAXFREE to allow more pending data to be held

You can increase the parameter MAXFREE to acquire more virtual storage for storing pending data.
However, there are some circumstances in which the quantity of pending data can increase indefinitely.
For example, Fast Path output messages defined with the system definition TERMINAL macro specifying
OPTION=FPACK may never produce a type X'5936' terminal output dequeue log record. In this case the
associated transaction nodes will accumulate for the duration of the collect. In this scenario you can use
the TABLEFLUSH parameter to periodically flush incomplete transaction information based on its age.
If you do not use the ddname DRLICHKO to write checkpoint records at the end of the collect, you can
use the WRITEPENDING parameter to flush the incomplete transactions at the end of the collect.
WRITEPENDING is the only method of writing out incomplete transactions when a COLD start is
encountered in the log file.

Using DRLSLOGP on remote systems

You can use the stand-alone utility DRLSLOGP to process IMS log data sets on remote systems and
transfer the DRLICOMP or DRLIRPT2 output to your central location for collect processing. The
DRLICOMP data set can be collected using the normal IMS log collect, but the DRLIRPT2 data set must be
collected using the IMS light feature log collect. Using either method the amount of data to be transferred
can be significantly smaller than the original logs.
You can further reduce the amount of data to be transferred by using TERSE on the remote system, and
UNTERSE on the central system. Note that only the transaction transit time subcomponent can be
updated using DRLICOMP or DRLIRPT2 input. For more information about using DRLSLOGP and collecting
DRLICOMP and DRLIRPT2, see “Using DRLSLOGP” on page 29.

IEC130I issued for SQNLOGS+1

If SQNLOGS=n, where n is between 1 and 9, you will find an informational message in the collect job
output:

IEC130I DRLLOGx DD STATEMENT MISSING

where x = n + 1.

34 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

Using the light feature with IMS Application Tracing Facility (ATF) summary
data

About this task

The IMS light feature does not update the KPM_IMS_ATF_T and KPM_IMS_ATF_EVT_T data tables.
Therefore, if you are using the light feature and you want to collect IMS ATF summary data, follow these
steps:
1. From the IMS logs that you are processing, extract IMS type nn, subtype 01 records (the IMS ATF
record type is user defined). This could be performed using DFSORT for example.
2. Transfer the R2 data and the extracted IMS ATF summary records to the z/OS system where IBM Z
Performance and Capacity Analytics is installed.
3. Run a collect on IMS ATF summary records to populate the KPM_IMS_ATF_T and
KPM_IMS_ATF_EVT_T data tables.
4. Run a collect by using the IMS light feature on the R2 records to populate the other IMS tables.

Using the light feature with HALDB data

About this task

The IMS light feature does not update the IMS_HALDB_OLR_x tables. Therefore, if you are using this
feature and you want to collect High Availability Large Database Reorganization data, follow these steps:
1. From the IMS logs that you are processing, extract IMS type 29, subtype 50 records. This could be
performed using DFSORT for example.
2. Transfer the R2 data and the IMS type 29 subtype 50 to the z/OS system where IBM Z Performance
and Capacity Analytics is installed.
3. Run a collect on HALDB OLR records to populate the IMS HALDB OLR tables.
4. Run a collect by using the IMS light feature on the R2 records to populate the other IMS tables.

Using TABLEFLUSH to prevent storage shortage failures

About this task

You can use the TABLEFLUSH and FLUSHTYPE parameters to prevent collect failures due to short-on-
storage conditions. To do this specify FLUSHTYPE=SOS and TABLEFLUSH=0. You can use a larger value
for TABLEFLUSH but this means the short-on-storage condition is more likely to recur sooner than if you
use a low value
If a short-on-storage condition occurs during a collect you should consider increasing the region size of
the collect step, or modifying your TABLEFLUSH and FLUSHTYPE parameters to minimize virtual storage
use as described below. The number of times table flush processing was invoked for a short-on-storage
condition is shown in the table flush processing statistics report in DRLOUT.

Using TABLEFLUSH to prevent performance degradation

About this task

IMS transactions can sometimes produce log record sequences that are not expected by the log
procedure. This can happen when a transaction loops, when I/O errors occur, or when the application has
an unusual design. Sometimes this causes the log procedure to store large amounts of transaction data in
the internal tables, which can result in poor performance.
To help handle this situation you can set an upper limit for the collect step REGION size based on past
normal collects. Allow the log procedure to flush the problem entries by requesting short-on-storage

Chapter 3. Administering the IMS Performance Feature 35

Administering the IMS Performance Feature

table flush processing by specifying FLUSHTYPE=SOS and TABLEFLUSH=0. Be aware that setting the
REGION size too low may result in unnecessary table flush processing, while setting it too high may delay
or prevent the invocation of short-on-storage table flush processing.
If you use this technique you should expect most collect steps to complete without short-on-storage
table flush processing being invoked. Successful use of this technique is indicated by short-on-storage
table flush processing being invoked a small number of times when a problem log is processed. It should
be possible to relate this back to an application problem occurring in the IMS system at the time the log
was generated.
The number of times table flush processing was invoked for a short-on-storage condition is shown in the
table flush processing statistics report in DRLOUT.

Using TABLEFLUSH to improve performance

About this task

You can use the TABLEFLUSH and FLUSHTYPE parameters to improve the performance of the collect
process. There is a trade-off between performance gains and data lost by the flush process, so you need
to decide what balance is desirable and monitor the table flush statistics to determine if you need to
adjust your table flush settings.
This technique is only effective if the poor performance occurs because the log being processed produces
long chains of internal table entries. For example if your IMS applications make extensive use of program
switching and you specify or default the DRLIPARM SECONDARY parameter to YES then large sets of
table entries are required to hold information from the related log records being processed.
You can specify either FLUSHTYPE=CHKPT or FLUSHTYPE=COUNT=nnnnnnnn to have table flush
processing occur at regular intervals. Use the TABLEFLUSH parameter to control how much data is
flushed each time the processing is performed. Review the CPU use information for the collect step in
system message IEF374I and the average storage freed per table flush in the table flush processing
statistics reported in message DRL2084I to assess the effectiveness of your settings.
For some value of FLUSHTYPE specific to a log, reducing that value no longer significantly reduces CPU
use, but continues to decrease the proportion of completed input PUOWs (transactions) being flushed,
and may start to increase the amount of CPU used by the collect step. Similar considerations apply to
varying the TABLEFLUSH value. The table flush processing statistics reported in message DRL2084I can
be used to check the proportion of completed input PUOWs being flushed. Setting
FLUSHTYPE=COUNT=10000 with TABLEFLUSH=1 is a good starting point for a balance between CPU use
and data loss.

Using TABLEFLUSH to minimize virtual storage use

About this task

You can use the TABLEFLUSH and FLUSHTYPE parameters to minimize virtual storage use. There is a
trade-off between virtual storage reduction and data lost by the flush process, so you need to decide
what balance is desirable and monitor the table flush statistics to determine if you need to adjust your
table flush settings.
This technique is only effective if the log records being processed require large amounts of virtual storage
for processing. For example, if your IMS applications make extensive use of program switching and you
specify or default the DRLIPARM SECONDARY parameter to YES then a large amount of virtual storage
can be used to hold information from the related log records being processed.
You also need to consider that if you process very large logs the amount of virtual storage used to buffer
type X'07' log records can reduce the relative impact of this technique. You can see how much virtual
storage has been used for buffering type X'07' log records by reviewing the SET07BUF row of the Total
Allocated (bytes) column in the statistics for NODEs created reported in message DRL2072I.

36 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Administering the IMS Performance Feature

You can specify either FLUSHTYPE=CHKPT or FLUSHTYPE=COUNT=nnnnnnnn to have table flush

processing occur at regular intervals. Use the TABLEFLUSH parameter to control how much data is
flushed each time the processing is performed.
Review the virtual storage use information for the collect step in system message IEF374I and the
average storage freed per table flush in the table flush processing statistics reported in message
DRL2084I to assess the effectiveness of your settings.
For some values of FLUSHTYPE specific to a log, reducing that value no longer significantly reduces virtual
storage use, but continues to decrease the proportion of completed input PUOWs (transactions) being
flushed, and may start to increase the amount of CPU used by the collect step. Similar considerations
apply to varying the TABLEFLUSH value. The table flush processing statistics reported in message
DRL2084I can be used to check the proportion of completed input PUOWs being flushed. Setting
FLUSHTYPE=COUNT=10000 with TABLEFLUSH=1 is a good starting point for a balance between virtual
storage use and data loss.

Recovering from abends during collect

The IMS Performance Feature log procedure saves information from the input records and combines this
information to produce composite records. However,the IBM Z Performance and Capacity Analytics
restart procedure skips input records that were processed before a Db2 commit. To protect against loss
of IMS input records when a failure occurs during log procedure processing, the IMS CSQ feature log
procedure has a checkpoint facility.

Recovery using the log procedure checkpoint facility

If the output checkpoint data set, DRLICHKO, was available and filled during a collect abend, and if that
data set is used as DRLICHKI in the restart, use the recovery procedure described in the Administration
Guide.
If you are using the IMS Performance Feature log procedure checkpoint facility, take care if an abend or
failure occurs during a collect commit. If such a failure occurs, backout and recovery become more
complex. Always allocate a sufficiently large (for example, 250 tracks of 3380 or 3390) checkpoint data
set to the DRLICHKO DDNAME to avoid space problems.

Recovery without the checkpoint facility

Recovery when the DRLICHKO data set is not available for use as DRLICHKI can be carried out as follows.

• If the IMS Performance Feature Db2 tables were not updated before the failure, rerun the collect job
after correcting the cause of the error.
• If the IMS Performance Feature Db2 tables were updated before the failure, restore the tables to the
status before the collect job failed. This can be done from Db2 backup copies of the IMS CSQ
performance feature tables. Run the Db2 RECOVER utility, using standard Db2 procedures for point-in-
time recovery.
After restoring the tables and correcting the cause of the error, rerun the collect job.
For information about Db2 recovery, see the Db2 Administration Guide: Volume 2 and Volume 3.

Additional capabilities
The following information is not stored in the Db2 tables provided with the IMS Performance Feature, but
is available for processing into user-defined tables. See the source for the IMS_Vnnn_R2 record definition
for complete information:
• Transaction abend codes and completion codes from X'07' and X'5938' log records
• Program-to-program switch root information about the transaction and terminal that started a
sequence of program to program transactions

Chapter 3. Administering the IMS Performance Feature 37

Administering the IMS Performance Feature

• MSC and ISC root information about the transaction or message switch that started the sequence of
multi-system transactions

38 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Chapter 4. Data table reference

This section describes the data tables, lookup tables, and control tables of the IMS Performance Feature
of IBM Z Performance and Capacity Analytics.

Log and record definitions

This section describes:
• IMS CSQ feature log definitions.
• IMS Performance Feature record definitions corresponding to IMS record types.
• Composite and R2 record definitions as created by the IMS Performance Feature log and record
procedures.
For more information about log and record definitions, refer to the Language Guide and Reference manual.

Log definitions
Log definitions reside in the product system tables. They define each log to the IMS Performance Feature.
You must define a log to IBM Z Performance and Capacity Analytics before any data can be collected. You
specify the log definition that you want to use in the COLLECT statement. See the topic "Collect" in the
Language Guide and Reference for more information about the COLLECT statement.
Depending on which components you install, one or more of these log definitions are installed, where nnn
refers to the related IMS version:
CSQ_Vnnn_COLLECT
This log definition is part of the collect components. It works together with the IMS Performance
Feature log procedures and record procedures. Use it for standard product data collection by issuing
the COLLECT CSQ_Vnnn_COLLECT statement.
CSQ_Vnnn_COLLECL
This log definition is part of the collect components. It works without the IMS Performance Feature
log and record procedures to process R2 records created by running DRLSLOGP. Use it for light
product data collection by issuing the COLLECT CSQ_Vnnn_COLLECL statement.
IMS_Vnnn_SLDS
This log definition is part of the log records components. It does not make use of the IMS
Performance Feature log procedures and record procedures. You can use it for special applications,
such as troubleshooting, debugging, and detailed IMS log analysis. Use it to collect data for these
special applications by issuing the COLLECT IMS_Vnnn_SLDS statement.

Record definitions
Each record in a log belongs to some record type. Record definitions describe each record type to the log
collector.

Descriptions of record definitions

Record definition names for IMS follow this general format:
IMS
Common prefix that indicates an IMS record definition
Vnnn
Indicates the release number (for example VD signifies IMS version 13, release 1)
record type
Corresponding IMS record type

© Copyright IBM Corp. 1994, 2017 39

IMS Performance Feature Reference

Table 5 on page 40 lists IMS record types with the corresponding IMS CSQ feature record definition
name and description. It also indicates support for the IMS record type in the last two columns with these
abbreviations:
X
The record type is supported for this release of IMS.
-
The record type is not supported for this release of IMS.
NA
The record type is not applicable for this release of IMS.

Table 5. IMS record types and log records component record definitions
Record Record definition Description
type
X'01' IMS_Vnnn_01 Message queue record (message received from a CNT).
This record represents the message and its text and control information as it appears
on the IMS message queues.
The DRRN indicates the message queue type:
X'00......' indicates QBLKS
X'04......' indicates short message queue
X'08......' indicates long message queue
The record contains the indicator of the origin of the message, its destination,
whether it uses MSC, and so on.
Note that IMS writes this record when a terminal or another network attached system
receives a message from a CNT (such as MSC/ISC and FES). It may be input to a
program (if it is enqueued to an SMB) or it may switch to another CNT (the latter is
known as a message switch).
The only occasion when the date and time fields represent an approximation of when
the event occurred is for the originating message. Be careful when using these fields
for program-to-program switches and MSC/ISC/FES activity.

X'02' IMS_Vnnn_02 Command log record.

This record represents a condensed version of the command entered by the IMS user
or program. If the command string requires a record longer than the logical record
length of the log, the record is segmented into several type X'02' records.

40 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'03' IMS_Vnnn_03 Message queue record (message received from a PSB or IMS).
This record represents the message, and its text and control information as it appears
on the IMS message queues.
The DRRN indicates the message queue type:
• X'00......' to indicate QBLKS
• X'04......' to indicate short message queue
• X'08......' to indicate long message queue
The record contains the indicator of the origin of the message, its destination,
whether it uses MSC, and so on.
Note that IMS writes this record when a message is received from IMS or a PSB (such
as the output from a program or a system-generated message) as sent to the master
terminal operator (MTO). This record can be input to an SMB (for a program-to-
program switch) or sent to a CNT.
The INode, sequence number, and date and time fields (MSGINODE, MSGTISEQ,
MSGEDATE and MSGETIME) pass to subsequent messages, which associates
subsequent messages with the originating message. However, MSGTISEQ is not
propagated for MSC.
The date and time fields represent an approximation of when the event occurred only
for the originating message. Be careful when using these fields for anything but IMS
system-generated output.
If a program-to-program switch occurs during a conversation, then the conversational
message is not placed in the SPA but rather is present in the type X'03' SMB-
generated message (a X'13' is not generated).

X'06' IMS_Vnnn_06 IMS event accounting record.

IMS writes this record when major IMS system events occur. The Accounting
Identifier field (ACIDENT) lists the events that cause this record to be written.

Chapter 4. Data table reference 41

IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'07' IMS_Vnnn_07 Program termination accounting record.
IMS accounts for all programs scheduled and terminated under its control with this
record (termination) and the type X'08' schedule record.
Type X'08' and type X'07' records are related to each other by the first 12 bytes of the
recovery token, which are unique for the duration of the IMS session.
This termination record contains the date and time of program termination and the
resources it consumed during its scheduling.
Several messages can be processed during the time this record is scheduled (see
field DLRMCNT) and several commits (see field DLRTOKNS) can occur. Therefore, the
precise amount of dependent region CPU and the number of DL/I calls cannot be
calculated for each message or commit.
The only way that the amount of dependent region CPU and number of DL/I calls can
be apportioned to the message or commit truly responsible is by:
• Calculating the mean
• Apportioning according to the proportion of processing time for each message in
relation to the total program schedule time
• Using regression analysis to find the best fit
Despite these restrictions, this record does represent an accurate account of TCB
time consumed by the programs, as scheduled in a region, when compared to the
time captured by SMF and recorded in SMF type 30.
Type X'08' and type X'07' records are written for all region types including MPP, BMP,
IFP, and WFI.

X'08' IMS_Vnnn_08 Program schedule record.

IMS accounts for all programs scheduled and terminated under its control with this
record (schedule) and by a type X'07' termination record.
The type X'08' and type X'07' records are related to each other through the first 12
bytes of the recovery token. These bytes are represented by fields LINTOKNN and
LINTOKNQ, which are unique for the duration of the IMS session.
This schedule record contains the date and time of program schedule and the region
and schedule initiation type.
Type X'08' and type X'07' records are written for all region types including MPP, BMP,
IFP, and WFI.

X'09' Sequential buffering statistics.

X'0A07' IMS_Vnnn_0A07 CPI-C program termination.
X'0A08' IMS_Vnnn_0A08 CPI-C program initialization.
X'0F' LE DECB record.
X'10' IMS_Vnnn_10 Security violation record.
This record indicates that IMS detected a security violation, identifies the precise
nature of the violation and specifies whether it is terminal or program-related.

X'11' IMS_Vnnn_11 Start of conversation record.

This record represents the start of a transaction's conversation session. The
allocation of a CCB and an SPA maintains the continuity between transactions in this
conversation.

42 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'12' IMS_Vnnn_12 End of conversation record.
This record represents the termination of the conversation that was started and
logged by the type X'11' record. This record is linked to the X'11' Start record of
conversation record through the node name.

X'13' IMS_Vnnn_13 SPA insert record.

This record represents the control information and text inserted to the scratch pad
area by a previously started conversation. It is linked to the active transaction
through the recovery token field.
When written to the log, the DC routine packs this data by removing and flagging
removed strings of blanks and zeros.
If a program-to-program switch occurs during a conversation, the conversational
message is not placed in the SPA, but instead is present in the type X'03' SMB-
generated message. A X'13' is not generated in such a case.

X'14' Switched-line disconnect.

X'15' Switched-line connect.
X'16' IMS_Vnnn_16 Sign-on/off record.
This record logs the security maintenance utility (SMU) or Resource Access Control
Facility (RACF®) user sign-on and sign-off.

X'18' IMS_Vnnn_18 Extended checkpoint record.

This record logs the details and checkpoint data for an extended checkpoint.

X'20' IMS_Vnnn_20 Database open record.

This record indicates that a DL/I database was opened and describes various key
characteristics of the database and its files.

X'21' IMS_Vnnn_21 Database close record.

This record indicates that a DL/I database was closed and provides several key
details about the database and its files. A database open X'20' record was created
earlier.

X'24' IMS_Vnnn_24 Database error record.

This record indicates that a DL/I database had an error. It details the program and
transaction accessing the database at the time of the error, the time the error
occurred, and the relative byte address (RBA) and cylinder cylinder head head record
(CCHHR) details of the error.

X'25' EEQE record.

X'26' I/O toleration buffer.
X'27' Database extension.
X'28' Phase 1 DC record.
X'30' IMS_Vnnn_30 Message queue prefix changed record.
This record logs changes made to the message queue record prefix. It links directly to
a previously logged message.

Chapter 4. Data table reference 43

IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'31' IMS_Vnnn_31 Message queue GU record.
This record logs the details of a message that is GU'd from the message queue to be
sent to its destination. The destination may be an SMB or CNT.
The record is present for incoming messages that are processed by a program
scheduled in a message processing region, or for outgoing messages that are sent to
a network destination. In addition, the record is present for message switches.
The timestamp in this record essentially represents the time that the message ceased
waiting on the message queue. If the message is sent to a CNT, a type X'36' record
follows, ultimately being followed by a type X'33' Free record, regardless of
destination.

X'32' IMS_Vnnn_32 Message queue reject record.

This record is produced when the MSGQ rejects a message because an error
occurred, presumably causing an application program abend.

X'33' IMS_Vnnn_33 Message queue DRRN free record.

This record indicates that DRRNs were freed from the message queues, the message
was deleted, and the DRRNs are available for reuse.
This record always indicates that the message is no longer needed by IMS .

X'34' IMS_Vnnn_34 Message queue cancel record.

This record indicates that the message was canceled from the queue and that a
subsequent X'35' Enqueue record was not produced.

X'35' IMS_Vnnn_35 Message queue enqueue record.

This record indicates that the message in the message queue (logged as a type X'01'
or X'03' record) has been placed on the queue for processing. If the destination is an
SMB, it is usually waiting on the input queue for the PSB to issue the Get Unique. If
the destination is a CNT, it is either a message switch or an outbound message and is
waiting in the output queue.
This record follows the logging of the message as a X'01' or X'03' record, and
precedes the X'31' record that indicates the message has been retrieved from the
queue for processing or transmission.
The timestamp can be carried from the preceding X'01' or X'03'. Because the
timestamp from this record may not reflect the real time the message arrived in the
system, exercise caution when using it.

X'36' IMS_Vnnn_36 Message queue dequeue record.

This record indicates that the destination CNT received the message, and the
message has been dequeued or deleted. This action is prompted by receiving an SNA
definite response or exception response.
Under certain circumstances, when the message does not request a definite
response, the timestamp in the record reflects the date and time of the next input
message, and therefore, the estimated network transit time includes user think time.

X'37' IMS_Vnnn_37 Message queue syncpoint transfer record.

This record indicates that the message transferred to the permanent destination and
reflects that a successful commit occurred.

X'38' IMS_Vnnn_38 Message queue syncpoint fail record.

This record indicates that a syncpoint failure occurred and the message transfer will
not occur.

44 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'39' Cleanup outqueue release.
X'3A' DFSQFIX0 free.
X'3B' DFSQFIX0 invalid message.
X'3C' DFSQFIX0 validity check.
X'3D' DFSQFIX0 QBLK altered.
X'3E' Message chain update.
X'4001' IMS_Vnnn_4001 Checkpoint begin record.
This record contains system-wide information about IMS, and represents the
beginning of an IMS system checkpoint.
This record follows the logged buffer and pool statistics record, but represents the
notification of the start of the IMS checkpoint process .

X'4002' IMS_Vnnn_4002 Checkpoint message queue record.

This record contains a checkpoint of all the allocated queue blocks, short message
and long message records at the time of the IMS checkpoint.

X'4003' IMS_Vnnn_4003 Checkpoint CNT record.

This record contains a checkpoint of all the CNTs defined to the IMS system, and their
status at the time of the checkpoint.

X'4004' IMS_Vnnn_4004 Checkpoint SMB record.

This record contains a checkpoint of all the SMBs defined to the IMS system and their
status at the time of the IMS checkpoint.

X'4005' IMS_Vnnn_4005 Checkpoint CTB record.

This record contains a checkpoint of all the CTBs defined to IMS and their status at
the time of the IMS checkpoint.

X'4006' IMS_Vnnn_4006 Checkpoint DMB record.

This record contains a checkpoint of the database manager blocks (DMBs) defined to
the IMS system and their status at the time of the IMS checkpoint.

X'4007' IMS_Vnnn_4007 Checkpoint PSB record.

This record contains a checkpoint of all the PSBs defined to the IMS system at the
time of the checkpoint.

X'4008' IMS_Vnnn_4008 Checkpoint CLB record.

This record contains a checkpoint of all the communications line blocks (CLBs)
defined to the IMS system at the time of the IMS checkpoint.

X'4009' Checkpoint CPT.

X'400A' Checkpoint CPM.
X'400B' Checkpoint CTM.
X'400C' Checkpoint CVB.
X'400D' IMS_Vnnn_400D Checkpoint CCB record.
This record contains a checkpoint of the CCBs defined in the IMS system and their
status at the time of the IMS checkpoint.

Chapter 4. Data table reference 45

IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'400E' IMS_Vnnn_400E Checkpoint SPA record.
This record contains a checkpoint of the currently allocated SPAs for active
conversations.

X'400F' Checkpoint LCB.

X'4010' Checkpoint CRB.
X'4011' Checkpoint TCM.
X'4014' IMS_Vnnn_4014 Checkpoint SPA QB record.
This record contains a checkpoint of all the SPA queue blocks (QBs) defined in the
IMS system and their status at the time of the IMS checkpoint.

X'4015' IMS_Vnnn_4015 Checkpoint EQE record.

This record contains a checkpoint of all the generated error queue elements (EQEs) at
the time of the IMS checkpoint.

X'4020' IMS_Vnnn_4020 Checkpoint CIB record.

This record contains a checkpoint of all the communications interface blocks (CIBs)
defined to the IMS system at the time of the IMS checkpoint.

X'4021' IMS_Vnnn_4021 Checkpoint VTCB record.

This record contains a checkpoint of all the VTAM terminal control blocks (VTCBs)
defined to the IMS system and their status at the time of the IMS checkpoint.

X'4025' Checkpoint EEQE.

X'4026' Checkpoint IEEQE/virtual I/O buffer.
X'4027' In-doubt extended error queue elements (IEEQE).
X'4028' Error queue elements (EQEL) for RIS.
X'4030' Checkpoint SID.
X'4031' Checkpoint RRE.
X'4070' IMS_Vnnn_4070 Checkpoint MSDB begin record.
This record contains system-wide information about IMS MSDBs, such as at the
beginning of the IMS checkpoint.

X'4071' IMS_Vnnn_4071 Checkpoint MSDB ECNT record.

This record contains main storage database (MSDB) ECNT data for the IMS
checkpoint.

X'4072' IMS_Vnnn_4072 Checkpoint MSDB header record.

This record contains the checkpointed MSDB headers as defined to the IMS system
and their contents at the time of the IMS checkpoint.

X'4073' IMS_Vnnn_4073 Checkpoint MSDB pagefixed record.

This record contains the checkpointed pagefixed MSDBs at the time of the IMS
checkpoint.

X'4074', IMS_Vnnn_4074 Checkpoint MSDB pageable record.

This record contains the checkpointed pageable MSDBs at the time of the IMS
checkpoint.

46 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'4079' IMS_Vnnn_4079 Checkpoint MSDB end record.
This record indicates that the IMS MSDB checkpoint process is now complete for this
IMS checkpoint.

X'4080' IMS_Vnnn_4080 Checkpoint Fast Path begin record.

This record contains system-wide information about IMS Fast Path, such as at the
beginning of the IMS checkpoint.

X'4081' IMS_Vnnn_4081 Checkpoint Fast Path ECNT record.

This record contains the checkpointed Fast Path ECNT data at the time of the IMS
checkpoint.

X'4082' IMS_Vnnn_4082 Checkpoint Fast Path EMHB record.

This record contains a checkpoint of all the allocated expedited message handler
blocks at the time of the IMS checkpoint.

X'4083' IMS_Vnnn_4083 Checkpoint Fast Path RCTE record.

This record contains a checkpoint of all the routing code table entries defined to IMS
and their status at the time of the IMS checkpoint.

X'4084' IMS_Vnnn_4084 Checkpoint FP DMCB/DMAC.

X'4085' IMS_Vnnn_4085 Checkpoint Fast Path MTO buffer record
X'4086' IMS_Vnnn_4086 Checkpoint Fast Path DMHR/DEDB record.
X'4087' IMS_Vnnn_4087 Checkpoint Fast Path ADSC record.
X'4088' IMS_Vnnn_4088 Checkpoint Fast Path IEEQE record.
X'4089' IMS_Vnnn_4089 Checkpoint Fast Path end record.
This record indicates that the IMS Fast Path checkpoint process is now complete for
this IMS checkpoint.

X'4098' IMS_Vnnn_4098 Checkpoint end blocks record.

This record indicates that the IMS simple checkpoint is now complete.

X'4099' IMS_Vnnn_4099 Checkpoint end queues record.

This record indicates that the IMS dumpq checkpoint is now complete.

X'41' IMS_Vnnn_41 Checkpoint batch record.

This record indicates that a batch program has issued a checkpoint.

X'42' IMS_Vnnn_42 Log buffer control record.

This record indicates the status of IMS at log buffer end of volume and switch times.

X'43' IMS_Vnnn_43 Log data set control record.

This record indicates the status of the IMS OLDS data sets.

X'4502' IMS_Vnnn_4502 Queue pool statistics record.

This record contains statistics about the use of the message queue pool at the time of
the IMS checkpoint.

X'4503' IMS_Vnnn_4503 Format buffer pool statistics record.

This record contains statistics about the usage of the format buffer pool at the time of
the IMS checkpoint.

Chapter 4. Data table reference 47

IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'4504' IMS_Vnnn_4504 Database buffer pool statistics record.
This record contains statistics about the usage of the database buffer pool at the time
of the IMS checkpoint.

X'4505' IMS_Vnnn_4505 Main pools statistics record.

This record contains statistics about the usage of the principal pools at the time of the
IMS checkpoint.

X'4506' IMS_Vnnn_4506 Scheduling statistics record.

This record contains statistics about scheduling conflicts in IMS at the time of the
checkpoint.

X'4507' IMS_Vnnn_4507 Logger statistics record.

This record contains statistics about the logical logger function of IMS at the time of
the IMS checkpoint.

X'4508' IMS_Vnnn_4508 VSAM subpool statistics record.

This record contains statistics about the VSAM subpools at the time of the IMS
checkpoint.

X'4509' IMS_Vnnn_4509 Program isolation statistics record.

This record contains statistics about IMS program isolation and enqueue/dequeue at
the time of the IMS checkpoint.

X'450A' IMS_Vnnn_450A Latch statistics record.

This record indicates the status of IMS latches at checkpoint time.

X'450B' IMS_Vnnn_450B Dispatch storage statistics record.

This record indicates the selective dispatching storage pool status at the time of the
checkpoint.

X'450C' IMS_Vnnn_450C DFSCBT00 storage statistics record.

This record indicates the status of miscellaneous IMS storage pools at the time of the
IMS checkpoint.

X'450D' IMS_Vnnn_450D RECANY (receive any) buffer statistics.

This record contains statistics about the VTAM receive any (RECANY) buffer usage at
the time of the IMS checkpoint.

X'450E' IMS_Vnnn_450E Storage manager statistics record.

X'450F' IMS_Vnnn_450F Dispatch statistics record.
X'45FF' IMS_Vnnn_45FF Statistics.
X'47' IMS_Vnnn_47 Active region statistics record.
This record contains information about all active regions, including BMP programs, at
the time of the IMS checkpoint.

X'48' IMS_Vnnn_48 OLDS padding record.

This record contains padding and control information for the IMS OLDS.

48 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'4C' IMS_Vnnn_4C Program/database start/stop record. This record indicates the starting and stopping
of program scheduler blocks (PSBs) and database manager blocks (DMBs). It does
not carry a timestamp, but given some locality of reference in relation to other
records containing reliable timestamps, an approximation of PSB and DMB availability
can be made using this record as the start/stop flag.
X'5050' IMS_Vnnn_5050 Full function database update undo/redo successful record.
This record indicates that the logging of undo and or redo data for a full function
database is complete for a database update.

X'5051' IMS_Vnnn_5051 Full function database update unsuccessful record.

This record indicates that the update action indicated by the previous X'50' record
was unsuccessful.

X'5052' IMS_Vnnn_5052 Full function database update undo KSDS insert record.
This record contains the undo data for a KSDS insert. The presence of a subsequent
X'5050' or X'5051' indicates that the action was successful.

X'53' CI/CA and space manager.

X'55' IMS_Vnnn_ 55FE0001 External subsystem Db2 snap in-doubt record.
This record indicates that a Db2 external subsystem had to resolve in-doubt
structures for a database.

X'56' IMS_Vnnn_56 External subsystem record.

This record indicates the status of external subsystem connection and commit
processing.

X'56FA' CSQ_Vnnn_56FA Transaction Level Statistics

This record contains transaction level statistics when the IMS system has been set up
to collect these records.

X'5901' IMS_Vnnn_5901 Fast Path input message.

This record indicates the receipt of an input message to the expedited message
handler buffer for Fast Path processing.

X'5903' IMS_Vnnn_5903 Fast Path output message.

This record indicates the placing of an output message into the expedited message
handler buffer, after completion of Fast Path processing.

X'5920' IMS_Vnnn_5920 Fast Path MSDB change record.

This record indicates the changing of an MSDB.

X'5921' IMS_Vnnn_5921 Fast Path DEDB area data set open record.
This record indicates the opening of a Fast Path DEDB area data set.

X'5922' IMS_Vnnn_5922 Fast Path DEDB area data set close record.
This record indicates the closing of a Fast Path DEDB area data set.

X'5923' IMS_Vnnn_5923 Fast Path DEDB area data set status record.
This record indicates the status of a Fast Path DEDB area data set.

X'5924' IMS_Vnnn_5924 Fast Path DEDB area data set EQE creation record.
This record indicates the creation of an error queue element for a Fast Path DEDB
area data set.

Chapter 4. Data table reference 49

IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'5936' IMS_Vnnn_5936 Fast Path dequeue message record.
This record indicates that an expedited message handler message has been sent and
successfully received by its destination node.

X'5937' IMS_Vnnn_5937 EMH Fast Path syncpoint record.

This record indicates that a successful Fast Path syncpoint occurred, indicating that
any messages can be transmitted.

X'5938' IMS_Vnnn_5938 EMH Fast Path syncpoint failure record.

This record indicates that a Fast Path syncpoint failed and that message transmission
may not occur.

X'5942' Fast Path DMHR dequeue.

X'5947' Fast Path MSSP image copy.
X'5950' IMS_Vnnn_5950 Fast Path DEDB database update record.
This record indicates that a Fast Path database online update occurred.

X'5951' IMS_Vnnn_5951 Fast Path DEDB database update record.

This record indicates an update made in a non-recoverable AREA/DEDB.

X'5953' IMS_Vnnn_5953 Fast Path DEDB database update (utilities) record.

This record indicates that Fast Path database utilities update occurred.

X'5954' IMS_Vnnn_5954 Fast Path DEDB database open record.

This record indicates the opening of a Fast Path DEDB database.

X'5955' IMS_Vnnn_5955 Fast Path sequential dependent syncpoint record.

This record indicates that a new sequential dependent buffer was obtained during
syncpoint processing.

X'5957' IMS_Vnnn_5957 Fast Path database DMAC record.

X'5970' IMS_Vnnn_5970 Fast Path hot standby MSDB relocation record.
This record indicates that an MSDB has been relocated to the XRF hot standby system
during takeover.

X'5B' Buffered Fast Path.

X'5E' Image capture of SB handler.
X'5F' DL/I call trace.
X'62' OSAM error.
X'63' Allocate/deallocate.
X'64' Message discarded by MSC.
X'65' IRSS and SNA restart.
X'66' 3600 standard record.
X'67' IMS_Vnnn_67 Subtypes: 00, 01, 03, 06, E0, ED, EE, EF, FB, FC, FD, FF.
Communications trace, DMHR on I/O error, and snap trace records. These records
contain internal trace information as requested by the systems trace.

50 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 5. IMS record types and log records component record definitions (continued)
Record Record definition Description
type
X'67FA' IMS_Vnnn_67FA Trace table log record.
This record contains the IMS trace table data.

X'69' 3275 switched unauthorized ID.

X'6C' MSC link connect/disconnect.
X'6D' XRF hot standby surveillance.
X'6E' XRF session miscellaneous.
X'70' Online change.
X'71' TCF record.
X'7201' IMS_Vnnn_7201 ETO user structure dynamically created.
X'7202' IMS_Vnnn_7202 ETO user structure dynamically created.
X'7203' IMS_Vnnn_7203 ETO user structure modified.
X'7204' IMS_Vnnn_7204 CNT added to an ETO user structure.

Comparison of performance programs

Table 6 on page 51 compares the IMS records used by the IMS CSQ feature with those used by similar
performance products and programs. The products and programs are abbreviated in the table:
DFSILTA0
IMS Log Transaction Analysis utility
DFSULTA0
IMS Fast Path Log Analysis utility
CSQ
z/OS IMS Performance Feature

Table 6. Comparison of the IMS Performance Feature with other products for IMS record types
Record Description DFSILTA0 DFSULTA0 CSQ
X'01' Message received from a CNT X X
X'03' Message received from DL/I X X
X'07' Program termination X X
X'08' Program initiation X X
X'11' Start of conversation X
X'12' End of conversation X
X'13' SPA record X
X'31' Message queue GU X X
X'32' Message queue reject X X
X'33' Message queue free X X
X'34' Message cancel X X
X'35' Message queue enqueue X X
X'36' Message queue dequeue X X
X'37' Syncpoint record X

Chapter 4. Data table reference 51

IMS Performance Feature Reference

Table 6. Comparison of the IMS Performance Feature with other products for IMS record types (continued)
Record Description DFSILTA0 DFSULTA0 CSQ
X'38' Message after abend X
X'4001' IMS checkpoint begin X
X'4004' Checkpoint SMB X X
X'4098' IMS checkpoint end X X
X'42' Log buffer control X X
X'45' Statistics records X
X'47' Active region X
X'56' External subsystem X
X'56FA' Transaction Level Statistics X
X'5901' Fast Path input X X
X'5903' Fast Path output X X
X'5936' Fast Path dequeue X X
X'5937' Fast Path syncpoint X X
X'5938' Fast Path abend X X

Composite and R2 record definitions

Information about the IMS CSQ composite record definitions is useful if you want to use these records
outside of the IMS Performance Feature. For example, you can write your own application to use these
records. The composite record definitions are as follows:
CSQ_Vnnn_COMP
Composite transaction records containing data for a group of related transactions. This record is
created by the log procedure.
CSQ_Vnnn_STxxxx
Account, Availability and Statistics record, where xxxx is the record type: 4001, 4502, 4503, 4504,
4505, 4506, 4507, 4508, 4509, 450A, 450B, 450C, 450D, 450E, 47, 07, 08, 06, 0A07.
These records are created by the log procedure.
CSQ_Vnnn_R2
R2 record containing data for one transaction. This record is created by the record procedure which
extracts the information from the composite record.
CSQ_Vnnn_R2_LIGHT
R2 record containing data for one transaction. This is used by the light feature.

Composite record sections

IMS CSQ composite transaction records are composed of all fields from the source IMS records that are
used to populate the transaction tables. Each section of the composite record has a different name and
prefix. Table 7 on page 53 explains the sections of the composite record. Each section is identified by a
two-character prefix.

52 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 7. Composite record sections

Short description Name Prefix Section contents and explanation
Originating Unit of Work OUOW OW Each composite record contains one OUOW
section. This section contains common
information for all transactions in the composite
record.
This information is extracted from these IMS
record types:
X'01'
Message queue insert
X'03'
M.Q.I. destination an SMB
X'03'
Message queue enqueue
X'5901'
Aoi exit

Processing Unit of Work PUOW PW Each transaction in the composite record is

represented by a PUOW section. The PUOW
information is extracted from these IMS record
types:
X'01'/X'03'
Message queue insert
X'35'
Message queue enqueue
X'31'
Message queue GU
X'36'
Message queue dequeue
X'37'
Message commit/transfer
X'38'
Message failed
X'5901'
EMH input
X'5903'
EMH output
X'5936'
EMH dequeue
X'5937'
Fast Path commit
X'5938'
Fast Path failed

Chapter 4. Data table reference 53

IMS Performance Feature Reference

Table 7. Composite record sections (continued)

Short description Name Prefix Section contents and explanation
Program Schedule Block PSB PB Each program in the composite record is
represented by a PSB section. Each PSB section
is related to one or more PUOW sections in the
composite record.
The PSB information is extracted from these IMS
record types:
X'08'
Program scheduled
X'07'
Program terminated
Note: One of these records may be absent when
an IMS log is processed. This situation is
especially likely in the case of WFI regions and
IFP regions.

For more information about log and record definitions, see the Language Guide and Reference manual.

Data tables and lookup tables

The product database is a collection of Db2 tables. Each table contains a fixed number of columns. The
number of rows in each table varies with time, because of rows added by the collect function and because
of database maintenance.
The process of entering data into the Db2 tables consists of several stages. IBM Z Performance and
Capacity Analytics first summarizes the data from the log in one table. It then summarizes the contents of
that table into another table, and so on. An update definition specifies how data from one source (a record
type or table) enters into one target (always a table).
This section describes the data and lookup tables used by the IMS CSQ feature. For an explanation of the
naming standard used, see “Naming standard for tables” on page 54. For information about the
relationships between tables and between records and tables, see “IMS Performance Feature object
definitions” on page 20.

Naming standard for tables

Names of the IMS CSQ feature tables use this format:
IMS_content_suffix
Where:
• content is a description (for example, IMS_TRAN for transactions by transaction name and user ID).
• suffix indicates the summarization level of the data in the table (for example, IMS_TRAN_H for IMS
transactions by transaction name and user ID summarized by hour).
A table name can have these summarization-level suffixes:
T
The table holds nonsummarized data (timestamped data).
Q
The table holds data summarized by quarter hour.
H
The table holds data summarized by hour.

54 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

D
The table holds data summarized by day.
W
The table holds data summarized by week.
M
The table holds data summarized by month.
Lookup tables do not have a suffix.

Table descriptions
Each of the data and lookup table descriptions includes information about the table, a description of each
of the key columns, and a description of each of the data columns.
Key columns are marked with a K.
Data columns come after the last key column and are sorted in alphabetic order, with any underscores
ignored.
Note: Tables with similar contents (that is, tables with the same name but different suffixes) are
described under one heading. For example, “IMS_TRAN_H, _D, _W” on page 77 contains information
about three similar tables:

KPM_IMS_TRAN_H
KPM_IMS_TRAN_D
KPM_IMS_TRAN_W

Except for the DATE column, the contents of these tables are identical. Differences in the contents of
similar tables are explained in the column descriptions.
The DATE and TIME information is stored in the standard Structured Query Language (SQL) format and
displayed in the local format. The DATE column contains the first day of the week for weekly (_W) tables,
and the first day of the month for monthly (_M) tables (if any).
Hexadecimal codes in log records are stored as character data in Db2 tables. For example, a 2-byte field
X'FFFF' is stored as a 4-byte character string FFFF.

Control tables
The IMS Performance Feature uses the control tables DAY_OF_WEEK and PERIOD_PLAN, which are used
by many IBM Z Performance and Capacity Analytics features. For complete descriptions of these control
tables, refer to the Administration Guide.

IMS Performance Feature data tables

The data tables for the IMS CSQ collect components are grouped by the following subcomponents and
apply to the following versions:

Table 8. IMS CSQ subcomponent data tables

Subcomponent Data tables
Accounting “IMS_PSB_ACCOUNT_H, _D, _W” on page 60

Availability “IMS_AVAILABILITY_D, _W” on page 56

Extended Accounting “IMS_PSB_ACCOUNT2_H, _D, _W, _M” on page 65

Ext System Tran Transit Time “IMS_SYSTEM_TRAN2_H,_D, _M” on page 73

Chapter 4. Data table reference 55

IMS Performance Feature Reference

Table 8. IMS CSQ subcomponent data tables (continued)

Subcomponent Data tables
HALDB OLR “IMS_HALDB_OLR_T, _H, _D, _W” on page 59

Statistics “IMS_CHKPT_IOSAM_T” on page 58

“IMS_CHKPT_POOLS_T” on page 58
“IMS_CHKPT_REGION_T” on page 59
“IMS_CHKPT_STATS_T” on page 59
“IMS_CHKPT_VSAM_T” on page 59

System Tran Transit Time “IMS_SYSTEM_TRAN_H,_D” on page 70

Transaction Level Statistics “IMS_TRAN_H, _D, _W” on page 77

Transaction Transit Time “IMS_TRAN_H, _D, _W” on page 77

“IMS_TRAN_QUEUE_Q, _D” on page 82
“IMS_TRAN_QUEUE_QV, _DV” on page 84

Key Performance Metrics “KPM_IMS_TRAN_H,_D, _W” on page 85

“KPM_IMS_ATF_T” on page 89

“KPM_IMS_ATF_EVT_T” on page 90

The IMS Performance Feature maintains separate counts of:

• Full function and Fast Path transactions
• Non message-driven BMP programs
• Response times, including their component parts
The IMS Performance Feature also classifies transactions within response-time boundaries that you can
customize. So, you can maintain counts of full function and Fast Path transactions that fall into each
boundary.
The tables for the transaction and system subcomponents contain information that is:
• Taken from the records produced by log procedure DRLSInnL and record procedure DRLSInn2 (where
nn identifies the version of IMS).
• Supported by record definitions CSQ_Vnnn_COMP and CSQ_Vnnn_R2 (where nn identifies the version of
IMS).
Consequently, each table contains:
• Summaries of resources consumed (CPU and DL/I, DEDB and MSDB calls)
• Response-time statistics

IMS_AVAILABILITY_D, _W
These tables provide daily and weekly statistics on the availability of IMS subsystems and regions. They
contain consolidated data from the IMS_AVAILABILITY_T table.
For more information about the Availability subcomponent, also see page “The collect components” on
page 5.
The default retention period for these tables are:
IMS_AVAILABILITY_D
45 days

56 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

IMS_AVAILABILITY_W
365 days

Column name Key Data type Description

DATE K DATE Date the availability data applies to. For the _W table, this is the
date of the first day of the week.
SYSPLEX_NAME K Char(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
MVS_SYSTEM_ID K Char(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn)
by the systems programmer. This column is derived from the run
time parameter :MVS_SYSTEM_ID or from the
IMS_SYSTEM_NAMES lookup table because IMS log records do
not contain this field.
IMS_SYSTEM_ID K Char(8) The IMS subsystem ID.
RESOURCE_NAME K Char(18) Resource Name.
RESOURCE_TYPE K Char(8) Resource Type. Possible values are:
IMSSYS
IMS system
IMSREG
IMS region

AVAIL_OBJ_PCT DECIMAL (4,1) Availability objective for the resource, in percent. This is from the
column AVAIL_OBJ_PCT in the IMS_AVAIL_RESOURCE lookup
table. This value should be compared with the actual availability,
which is calculated as: 100*UP_IN_SCHEDULE/
SCHEDULE_HOURS
MEASURED_HOURS FLOAT Number of hours measured.
SCHEDULE_HOURS FLOAT Number of hours the resource is scheduled to be up.
STARTS SMALLINT Number of times the resource was started.
STARTS_IN_SCHEDULE SMALLINT Number of times the resource was started within the schedule.
STOPS SMALLINT Number of times the resource was stopped.
STOPS_IN_SCHEDULE SMALLINT Number of times the resource was stopped within the schedule.
UP_HOURS FLOAT Number of hours the resource was up.
UP_IN_SCHEDULE FLOAT Number of hours the resource was up within the schedule.

IMS_AVAILABILITY_T
This table provides detailed availability data about the IMS subsystem and regions. The data comes from
different IMS records. It is updated by the IMS_AVAIL_RESOURCE lookup table.
For more information about the Availability subcomponent, see section on “The collect components” on
page 5.
The default retention period for this table is 10 days.

Column name Key Data type Description

SYSPLEX_NAME K Char(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.

Chapter 4. Data table reference 57

IMS Performance Feature Reference

Column name Key Data type Description

MVS_SYSTEM_ID K Char(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn)
by the systems programmer. This column is derived from the run
time parameter :MVS_SYSTEM_ID or from the
IMS_SYSTEM_NAMES lookup table because IMS log records do
not contain this field.
IMS_SYSTEM_ID K Char(8) The IMS subsystem ID. This is from the IMS_SYSTEM_ID collect
parameter or from the specific field for the different record type
collected.
RESOURCE_NAME K Char(18) Resource Name. This is from the RESOURCE_TARGET_NM or
from RESOURCE_SOURCE_NM into IMS_AVAIL_RESOURCE
lookup table.
RESOURCE_TYPE K Char(8) Resource Type. Possible values are:
IMSSYS
IMS system
IMSREG
IMS region

INTERVAL_TYPE K Char(3) Interval type. Possible values are: ===, |==, ==|, |=|, XXX, |XX,
XX|, |X|, and blank, where:
=
Indicates that the resource is up (available)
X
Indicates that the resource is down
|
Indicates an interval start or end
blank
Means that the status is unknown

START_TIME K TIMESTAMP Start time of the interval.

END_TIME TIMESTAMP End time of the interval.
QUIET_INTERVAL_SEC INTEGER Number of seconds after the interval end that the resource is
expected to remain in the same status. If another interval with a
start time within this range appears, the two interval are merged.

IMS_CHKPT_IOSAM_T
This table contains an unsummarized record of the accumulated counts of ISAM and OSAM buffer pool
activity at each IMS system checkpoint.
It relates to the Statistics subcomponent. For more information about this subcomponent, see
“Statistics” on page 5.
The default retention period for the table is 7 days.

IMS_CHKPT_POOLS_T
This table contains an unsummarized record of the accumulated and nonaccumulated system pool usage
for the I/OP, CWAP, and HIOP pools at each IMS system checkpoint.
It relates to the Statistics subcomponent. For more information about this subcomponent, see page
“Statistics” on page 5.
The default retention period for the table is 7 days.

58 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

IMS_CHKPT_REGION_T
This table contains an unsummarized record of the dependent regions active at each IMS system
checkpoint, and the transactions and programs active at that time, if any.
This table relates to the Statistics subcomponent. For more information about this subcomponent, see
page “Statistics” on page 5.
The default retention period for this table is 7 days.

IMS_CHKPT_STATS_T
This table contains an unsummarized record of the accumulated and nonaccumulated IMS system-wide
statistics, MSGQ counts, and format buffer pool counts at each IMS system checkpoint.
This table relates to the Statistics subcomponent. For more information about this subcomponent, see
“The collect components” on page 5.
The default retention period for the table is 7 days.

IMS_CHKPT_VSAM_T
This table contains an unsummarized record of the accumulated counts of VSAM buffer pool activity at
each IMS system checkpoint.
It relates to the Statistics subcomponent. For more information about this subcomponent, see page
“Statistics” on page 5.
The default retention period for the table is 7 days.

IMS_HALDB_OLR_T, _H, _D, _W

The IMS_HALDB_OLR_T table provides data on IMS Versions 9 and 10 High Availability Large Database
Online Reorganization activity at Unit of Reorganization (UOR) level.
The IMS_HALDB_OLR_H, _D, _W tables provide statistics on IMS Version 9 High Availability Large
Database Online Reorganization activity at database partition level.
For IMS Versions 9 and 10 only, these tables relate to the HALDB OLR subcomponent. For more
information about this subcomponent, see page “HALDB OLR” on page 5.
All of these tables collect data from IMS type 29, subtype 50 records. The default retention periods are:
IMS_HALDB_OLR_T
7 days
IMS_HALDB_OLR_H
10 days
IMS_HALDB_OLR_D
45 days
IMS_HALDB_OLR_W
365 days

Column name Key Data type Description

DATE K DATE The date the activity occurred. From DATE.
TIME K TIME The time when the activity started, in the format HH.00.00. This applies
only to the _H and _T tables. From TIME.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 weekdays. This column is derived
using the MVS_SYSTEM_ID, DATE, and TIME columns as parameters in
the PERIOD function.

Chapter 4. Data table reference 59

IMS Performance Feature Reference

Column name Key Data type Description

PST_ID K CHAR(2) The IMS assigned number for the partition specification table (PST)
that contains the management and control information for the
dependent region that processed the transaction. From PSTNUM.
IMS_ID K CHAR(8) IMS system ID. From IMSSYSID.
DBD_NAME K CHAR(8) Database description name. From DBDNAME.
PARTITION_NAME K CHAR(7) Partition name. It is the database partition where the online
reorganization is being done. From PARTNAME.
EXEC_TIME FLOAT Execution time, in 0.00 seconds. From HORMTIME.
INPUT_DS CHAR(5) Input data sets. Possible value are: A-J, X (X for PHIDAM format, only)
and M-V, Y (Y for PHIDAM format, only). From HORMFLG1.
LOCK_COUNT INTEGER Count of locks. From HORMLOCK.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by the
systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES lookup
table because IMS log records do not contain this field.
PART_TYPE CHAR(7) Partition type. Possible value are PHDAM and PHIDAM. From
HORMFLG1.
ROOTS_MOVED FLOAT The number of moved roots. From HORMORSZ.
SEG_MOVED FLOAT The number of moved segments. From HORMUORS.
SIZE_MOVED FLOAT Size moved, in bytes. From HORMUORZ.
SYSPLEX_NAME CHAR(8) This column is derived from the run time parameter :SYSPLEX_NAME
or from the IMS_SYSTEM_NAMES lookup table because IMS log
records do not contain this field.
UOR_COUNT INTEGER Number of Units of Reorganization (UOR) occurred. This applies only to
the _H, _D, _W tables.
WAIT_TIME FLOAT Wait time, in 0.00 seconds. From HORMWAIT.

IMS_PSB_ACCOUNT_H, _D, _W
These tables contain hourly, daily, and weekly statistics on counts of transactions and resources used by
transaction name. They contain information that includes data for transactions scheduling a PSB. The PSB
Account tables give statistics for CPU time and elapsed time during a specified period for regions,
transactions, and programs (PSB).
These tables can help you determine such things as who is using too much CPU time or, conversely, what
programs or transactions, in which regions, are in a wait state too long. From the PGM_CPU_SEC column,
you can monitor the actual CPU time required for each transaction. For a given program, the CPU times
should be approximately the same across regions and from day to day. However, these mean times
should be interpreted based on the number of transactions per scheduling, which is also reported. If the
time begins to increase, the most likely reason is increased database activity. This could be a sign that
databases need to be reorganized. The other columns in the tables are related to database operations.
The exclusion of child transactions by setting SECONDARY=NO does not affect these tables. They are
populated by PSB cumulative data for root and child transactions, even when SECONDARY=NO is
specified in the collect job.
You can use these tables to identify transaction utilization and resource consumption on the IMS system.
The default retention periods for these tables are:
IMS_PSB_ACCOUNT_H
10 days

60 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

IMS_PSB_ACCOUNT_D
45 days
IMS_PSB_ACCOUNT_W
365 days

Column name Key Data type Description

DATE K DATE The date the activities occurred. For the _W table, this is the date of
the first day of the week.
TIME K TIME The time when the activity started, in the format HH.00.00. This
applies only to the _H table.
IMS_SUBSYSTEM_NAME K CHAR(8) The IMS subsystem name. From DLRTOKN.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested. From
DLRTRNCD.
PSB_NAME K CHAR(8) The name of the IMS program used to process the
transaction. This column contains the program specification
block (PSB).
From DLRNPSB.

PST_ID K CHAR(2) The IMS-assigned number for the partition specification table
(PST)that contains the management and control information for the
dependent region that processed the transaction. The PST can be
reused by IMS after a dependent region terminates, so region
occupancy and processing analysis are less meaningful if only the
region PST ID is used. So, you must also use the region job name
(REGION_JOB_NAME) to identify the dependent region.
REGION_JOB_NAME K CHAR(8) The MVS- and JES-identified job name for the IMS dependent
region. This column uniquely identifies the transaction processing
activity for each region, because the region identifier or PST ID can
be reused by IMS. From DRLNJOB.
APPLICATION_NAME CHAR(18) Application name. This is from the RESOURCE_TARGET_NM in
IMS_AVAIL_RESOURCE lookup table. If nothing is found,
$UNKNOWN is used as default.
DLI_APSB_CALLS REAL Number of DL/I APSB calls, derived from he count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRAPSB.
DLI_CHKP_CALLS REAL Number of DL/I CHKP calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRCHKP.
DLI_CMD_CALLS REAL The total number of DL/I CMD calls derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRCMD.
DLI_DPSB_CALLS REAL Number of DL/I DPSB calls, derived from he count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRDPSB.
DLI_EXCL_DEQUEUES REAL The total number of DL/I exclusive dequeue calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLREXCDQ.
DLI_EXCL_ENQUEUES REAL The total number of DL/I exclusive enqueue calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLREXCNQ.

Chapter 4. Data table reference 61

IMS Performance Feature Reference

Column name Key Data type Description

DLI_EXCL_ENQWAITS REAL The total number of waits DL/I-exclusive enqueue calls, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLREXCWT.
This column indicates possible impact due to the degree or type of
program isolation activity during this interval, either throughout the
system or resulting from this user or transaction.

DLI_GCMD_CALLS REAL The total number of DL/I GCMD calls, erived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRGCMD.
DLI_GMSG_CALLS REAL Number of DL/I GMSG calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRGMSG.
DLI_ICMD_CALLS REAL Number of DL/I ICMD calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRICMD.
DLI_INIT_CALLS REAL Number of DL/I INIT calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRINIT.
DLI_INQY_CALLS REAL Number of DL/I INQY calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRINQY.
DLI_MSG_AUTH_CALLS REAL Number of DL/I message AUTH calls, derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRAUTH.
DLI_MSG_CHNG_CALLS REAL Number of DL/I message CHNG calls, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRCHNG.
DLI_MSG_SETO_CALLS REAL Number of DL/I message SETO calls, derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRSETO.
DLI_QCMD_DEQUEUES REAL The total number of DL/I queue command dequeue calls, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLRQCODQ.
DLI_QCMD_ENQUEUES REAL The total number of DL/I queue command enqueue calls, derived
from the count stored in the program termination record (record
type X'07').Calculated as Sum of DLRQCONQ.
DLI_QCMD_ENQWAITS REAL The total number of waits for DL/I queue commands and enqueues,
derived from the count stored in the program termination record
(record type X'07'). Calculated as Sum of DLRQCOWT.
DLI_RCMD_CALLS REAL Number of DL/I RCMD calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRRCMD.
DLI_ROLB_CALLS REAL Number of DL/I ROLB calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRROLB.
DLI_ROLS_CALLS REAL Number of DL/I ROLS calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRROLS.
DLI_SETS_CALLS REAL Number of DL/I SETS calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSETS.

62 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

DLI_SETU_CALLS REAL Number of DL/I SETU calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSETU.
DLI_SLOG_CALLS REAL Number of DL/I SLOG calls, derived from he count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSLOG.
DLI_TEST_DEQUEUES REAL The total number of DL/I test dequeues, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRTSTDQ.
DLI_TEST_ENQUEUES REAL The total number of DL/I test enqueues, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRTSTNQ.
DLI_TEST_ENQWAITS REAL The total number of DL/I waits on test enqueues, derived from the
count stored in the program termination record (record type
X'07').Calculated as Sum of DLRTSTWT.
DLI_UPDT_DEQUEUES REAL The total number of DL/I update dequeues, derived from the count
stored in the program termination record (record type
X'07').Calculated as Sum of DLRSUPDQ.
DLI_UPDT_ENQUEUES REAL The total number of DL/I update enqueues, derived from the count
stored in the program termination record (record type
X'07').Calculated as Sum of DLRSUPNQ.
DLI_UPDT_ENQWAITS REAL The total number of DL/I waits on update and enqueues, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLRSUPWT.
DLI_XRST_CALLS REAL
DLIDB_CALLS REAL The total number of DL/I database calls, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRCLCNT.
DLIDB_DEQ_CALLS FLOAT
DLIDB_DLET_CALLS REAL The total number of DL/I database DLET calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRDLET.
DLIDB_GHN_CALLS REAL The total number of DL/I database GHN calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHN.
DLIDB_GHNP_CALLS REAL The total number of DL/I database GHNP calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHNP.
DLIDB_GHU_CALLS REAL The total number of DL/I database GHU calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHU.
DLIDB_GN_CALLS REAL The total number of DL/I database GN calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGN.
DLIDB_GNP_CALLS REAL The total number of DL/I database GNP calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGNP.
DLIDB_GU_CALLS REAL The total number of DL/I database GU calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGU1.

Chapter 4. Data table reference 63

IMS Performance Feature Reference

Column name Key Data type Description

DLIDB_ISRT_CALLS REAL The total number of DL/I database ISRT calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRISRT.
DLIDB_REPL_CALLS REAL The total number of DL/I database REPL calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRREPL.
DLIDBCTL_DBIO_SEC REAL The total elapsed time for DB I/O for DBCTL, in seconds, derived
from the value stored in the program termination record (record
type X'07'). Calculated as Sum of DLRTMEIO/1000000.
DLIDBCTL_DBIOS REAL The total number of DB I/Os for DBCTL, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRIOCNT.
DLIDBCTL_LOCK_SEC REAL The total elapsed time for locking for DBCTL, in seconds, derived
from the value stored in the program termination record (record
type X'07'). Calculated as Sum of DLRTMEPL/1000000.
DLIDC_GN_CALLS REAL The total number of DL/I message queue GN calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLRGUMES.
DLIDC_GU_CALLS REAL The total number of DL/I message queue GU calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLRGUMES.
DLIDC_ISRT_CALLS REAL The total number of DL/I message queue ISRT calls derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRISMES.
DLIDC_PURGE_CALLS FLOAT The total number of DL/I message queue PURGE calls derived from
the count store.
DLIACC_ACCUM_SEC REAL The total transaction time subqueue 6, in seconds, as stored in
record type X'07'. This represents the total time spent waiting in a
wait-for-input or pseudo wait-for-input region with no work to do.
Calculated as Sum of DLRACCQ6.
DLISQ6_ACCUM_SEC REAL The Region Subq 6 Time per Message in seconds. Calculated as
Sum of DLRSQ6TM.
LAST_MSG_USERID CHAR(8) User ID of the last message processed in this dependent region.
From DLRUSID.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
PGM_CPU_SEC REAL The total dependent region CPU seconds from the termination
record (record type X'07'). Calculated as DRLEXTIM/4096E6, or if
that is invalid, as DRLTIME/38400.

PGM_FAILURES REAL The total number of program abends.

64 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

PROGRAM_TYPE Char(10) From DRLTYPE. When:
X'80'
UOR END
X'10'
CPI REGION
X'08'
QUICK RESC
X'04'
DBCTL THRD
X'02'
BMP REGION
X'01'
MPP REGION
X'22'
JBP REGION
X'21'
JMP REGION

SYSPLEX_NAME CHAR(8) This column is derived from the run time

parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANS_PRIOR CHAR(2) Transaction priority. From DLRPRTY.
TRANSACTIONS REAL Number of transactions processed by the PSB. Calculated as Sum
of DLRMCNT. For CPI REGIONs, DBCTL THREADs, and non-
message-driven BMPs, this value is a count of the X'07' records.

IMS_PSB_ACCOUNT2_H, _D, _W, _M

These tables contain hourly, daily, weekly, and monthly statistics on counts of transactions and resources
used by transaction name and MVS system identifier.
They contain information that includes data for transaction scheduling a PSB and give statistics for CPU
time and elapsed time during a specified period for Regions, Transactions, Programs (PSB). These tables
can help you determine such things as who is using too much CPU time or, conversely, what programs or
transactions, in which regions, are in a wait state for too long. From the PGM_CPU_SEC column, you can
monitor the actual CPU time required for each transaction. For a given program, the CPU times should be
approximately the same across regions and from day to day; however, these mean times should be
interpreted based on the number of transactions per scheduling, which is also reported. If the time begins
to increase, the most likely reason is increased database activity. This could be a sign that databases
need to be reorganized. The other columns in the tables are related to DB operations.
For IMS Versions 9 and 10, these tables relate to the Extended Accounting subcomponent. For more
information about this subcomponent, see page “Extended Accounting” on page 5.
The default retention periods for these tables are:
IMS_PSB_ACCOUNT2_H
10 days
IMS_PSB_ACCOUNT2_D
45 days
IMS_PSB_ACCOUNT2_W
365 days
IMS_PSB_ACCOUNT2_M
765 days

Chapter 4. Data table reference 65

IMS Performance Feature Reference

Column name Key Data type Description

DATE K DATE The date the activities occurred. For the _M table, this is the date of
the first day of the month.
TIME K TIME The time when the activity started, in the format HH.00.00. This
applies only to the _H table.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 weekdays. This column is
derived using the MVS_SYSTEM_ID, DATE, and TIME columns as
parameters in the PERIOD function.
MVS_SYSTEM_ID K CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.
PSB_NAME K CHAR(8) The name of the IMS program used to process the transaction. This
column contains the program specification block (PSB).
IMS_SUBSYSTEM_NAME K CHAR(8) The IMS subsystem name. From DLRTOKN.
REGION_JOB_NAME K CHAR(8) The MVS- and JES-identified job name for the IMS-dependent
region. This column uniquely identifies the transaction processing
activity for each region, because the region identifier or PST ID can
be reused by IMS. From DRLNJOB.
PST_ID K CHAR(2) The IMS-assigned number for the partition specification table (PST)
that contains the management and control information for the
dependent region that processed the transaction. The PST can be
reused by IMS after a dependent region terminates, so region
occupancy and processing analysis are less meaningful if only the
region PST ID is used. So, you must also use the region job name
(REGION_JOB_NAME) to identify the dependent region.
APPLICATION_NAME CHAR(18) Application name. This is from the RESOURCE_TARGET_NM in
IMS_AVAIL_RESOURCE lookup table. If nothing is found,
$UNKNOWN is used as default.
DLI_APSB_CALLS REAL Number of DL/I APSB calls, derived from he count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRAPSB.
DLI_CHKP_CALLS REAL Number of DL/I CHKP calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRCHKP.
DLI_CMD_CALLS REAL The total number of DL/I CMD calls derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRCMD.
DLI_DPSB_CALLS REAL Number of DL/I DPSB calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRDPSB.
DLI_EXCL_DEQUEUES REAL The total number of DL/I exclusive dequeue calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLREXCDQ.
DLI_EXCL_ENQUEUES REAL The total number of DL/I exclusive enqueue calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLREXCNQ.

66 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

DLI_EXCL_ENQWAITS REAL The total number of waits DL/I-exclusive enqueue calls, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLREXCWT.
This column indicates possible impact due to the degree or type of
program isolation activity during this interval, either throughout the
system or resulting from this user or transaction.

DLI_GCMD_CALLS REAL The total number of DL/I GCMD calls, derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRGCMD.
DLI_GMSG_CALLS REAL Number of DL/I GMSG calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRGMSG.
DLI_ICMD_CALLS REAL Number of DL/I ICMD calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRICMD.
DLI_INIT_CALLS REAL Number of DL/I INIT calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRINIT.
DLI_INQY_CALLS REAL Number of DL/I INQY calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRINQY.
DLI_MSG_AUTH_CALLS REAL Number of DL/I message AUTH calls, derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRAUTH.
DLI_MSG_CHNG_CALLS REAL Number of DL/I message CHNG calls, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRCHNG.
DLI_MSG_SETO_CALLS REAL Number of DL/I message SETO calls, derived from the count stored
in the program termination record (record type X'07'). Calculated
as Sum of DLRSETO.
DLI_PURGE_CALLS REAL The total number of DL/I message queue PURGE calls derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRPUMES.
DLI_QCMD_DEQUEUES REAL The total number of DL/I queue command dequeue calls, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLRQCODQ.
DLI_QCMD_ENQUEUES REAL The total number of DL/I queue command enqueue calls, derived
from the count stored in the program termination record (record
type X'07').Calculated as Sum of DLRQCONQ.
DLI_QCMD_ENQWAITS REAL The total number of waits for DL/I queue commands and enqueues,
derived from the count stored in the program termination record
(record type X'07'). Calculated as Sum of DLRQCOWT.
DLI_RCMD_CALLS REAL Number of DL/I RCMD calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRRCMD.
DLI_ROLB_CALLS REAL Number of DL/I ROLB calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRROLB.
DLI_ROLS_CALLS REAL Number of DL/I ROLS calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRROLS.

Chapter 4. Data table reference 67

IMS Performance Feature Reference

Column name Key Data type Description

DLI_SETS_CALLS REAL Number of DL/I SETS calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSETS.
DLI_SETU_CALLS REAL Number of DL/I SETU calls, derived from the count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSETU.
DLI_SLOG_CALLS REAL Number of DL/I SLOG calls, derived from he count stored in the
program termination record (record type X'07'). Calculated as Sum
of DLRSLOG.
DLI_TEST_DEQUEUES REAL The total number of DL/I test dequeues, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRTSTDQ.
DLI_TEST_ENQUEUES REAL The total number of DL/I test enqueues, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRTSTNQ.
DLI_TEST_ENQWAITS REAL The total number of DL/I waits on test enqueues, derived from the
count stored in the program termination record (record type
X'07').Calculated as Sum of DLRTSTWT.
DLI_UPDT_DEQUEUES REAL The total number of DL/I update dequeues, derived from the count
stored in the program termination record (record type
X'07').Calculated as Sum of DLRSUPDQ.
DLI_UPDT_ENQUEUES REAL The total number of DL/I update enqueues, derived from the count
stored in the program termination record (record type
X'07').Calculated as Sum of DLRSUPNQ.
DLI_UPDT_ENQWAITS REAL The total number of DL/I waits on update and enqueues, derived
from the count stored in the program termination record (record
type X'07'). Calculated as Sum of DLRSUPWT.
DLI_XRST_CALLS REAL
DLIDB_CALLS REAL The total number of DL/I database calls, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRCLCNT.
DLIDB_DEQ_CALLS FLOAT
DLIDB_DLET_CALLS REAL The total number of DL/I database DLET calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRDLET.
DLIDB_GHN_CALLS REAL The total number of DL/I database GHN calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHN.
DLIDB_GHNP_CALLS REAL The total number of DL/I database GHNP calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHNP.
DLIDB_GHU_CALLS REAL The total number of DL/I database GHU calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGHU.
DLIDB_GN_CALLS REAL The total number of DL/I database GN calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGN.
DLIDB_GNP_CALLS REAL The total number of DL/I database GNP calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGNP.

68 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

DLIDB_GU_CALLS REAL The total number of DL/I database GU calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRGU1.
DLIDB_ISRT_CALLS REAL The total number of DL/I database ISRT calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRISRT.
DLIDB_REPL_CALLS REAL The total number of DL/I database REPL calls issued, derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRREPL.
DLIDBCTL_DBIO_SEC REAL The total elapsed time for DB I/O for DBCTL, in seconds, derived
from the value stored in the program termination record (record
type X'07'). Calculated as Sum of DLRTMEIO/1000000.
DLIDBCTL_DBIOS REAL The total number of DB I/Os for DBCTL, derived from the count
stored in the program termination record (record type X'07').
Calculated as Sum of DLRIOCNT.
DLIDBCTL_LOCK_SEC REAL The total elapsed time for locking for DBCTL, in seconds, derived
from the value stored in the program termination record (record
type X'07'). Calculated as Sum of DLRTMEPL/1000000.
DLIDC_GN_CALLS REAL The total number of DL/I message queue GN calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLRGUMES.
DLIDC_GU_CALLS REAL The total number of DL/I message queue GU calls derived from the
count stored in the program termination record (record type X'07').
Calculated as Sum of DLRGUMES.
DLIDC_ISRT_CALLS REAL The total number of DL/I message queue ISRT calls derived from
the count stored in the program termination record (record type
X'07'). Calculated as Sum of DLRISMES.
DLIACC_ACCUM_SEC REAL The total transaction time subqueue 6, in seconds, as stored in
record type X'07'. This represents the total time spent waiting in a
wait-for-input or pseudo wait-for-input region with no work to do.
Calculated as Sum of DLRACCQ6.
DLISQ6_ACCUM_SEC REAL The Region Subq 6 Time per Message in seconds. Calculated as
Sum of DLRSQ6TM.
LAST_MSG_USERID CHAR(8) User ID of the last message processed in this dependent region.
From DLRUSID.
PGM_CPU_SEC REAL The total dependent region CPU seconds from the termination
record (record type X'07'). Calculated as DRLEXTIM/4096E6, or if
that is invalid, as DRLTIME/38400.
PGM_FAILURES REAL The total number of program abends.

Chapter 4. Data table reference 69

IMS Performance Feature Reference

Column name Key Data type Description

PROGRAM_TYPE Char(10) From DRLTYPE. When:
X'80'
UOR END
X'10'
CPI REGION
X'08'
QUICK RESC
X'04'
DBCTL THRD
X'02'
BMP REGION
X'01'
MPP REGION
X'22'
JBP REGION
X'21'
JMP REGION

SYSPLEX_NAME CHAR(8) This column is derived from the run time

parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANS_PRIOR CHAR(2) Transaction priority. From DLRPRTY.
TRANSACTIONS REAL Number of transactions processed by the PSB. Calculated as Sum
of DLRMCNT. For CPI REGIONs, DBCTL THREADs, and non-
message-driven BMPs, this value is a count of the X'07' records.

IMS_SYSTEM_TRAN_H,_D
These tables contain hourly and daily statistics on counts of transactions and response times summarized
by IMS system. They contain information that includes data for message-queue-driven transactions and
BMPs, EMH-driven Fast Path transactions, and message switches. You can use these tables to view IMS
capacity, to monitor service-level trends by system, and for trend analysis of volumes and response
times.
For IMS Version 8, these tables relate to the Transaction Transit Time subcomponent. For more
information about this subcomponent, see page “Transaction Transit Time” on page 5. For IMS Versions 9
and 10, these tables relate to the System Tran Transit Time subcomponent. For more information about
this subcomponent, see page “System Tran Transit Time” on page 5.
The default retention periods for these tables are:
IMS_SYSTEM_TRAN_H
10 days
IMS_SYSTEM_TRAN_D
45 days

Column name Key Data type Description

DATE K DATE The date the activities occurred.
TIME K TIME The time when the activity started, in the format HH.00.00. This
applies only to the _H table.
ORIGIN_IMS K CHAR(8) The IMS subsystem ID defined in the origin part of the UOW token. It
identifies the activity origin.
PROCESS_IMS K CHAR(8) The IMS subsystem ID defined in the processing part of the UOW
token. It identifies the activity processor.

70 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.
PST_ID K CHAR(2) The IMS-assigned number for the partition specification table
(PST)that contains the management and control information for the
dependent region that processed the transaction. The PST can be
reused by IMS after a dependent region terminates, so region
occupancy and processing analysis are less meaningful if only the
region PST ID is used. So, you must also use the region job name
(REGION_JOB_NAME) to identify the dependent region.
REGION_JOB_NAME K CHAR(8) The MVS- and JES-identified job name for the IMS dependent
region. This column uniquely identifies the transaction processing
activity for each region, because the region identifier or PST ID can
be reused by IMS.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 weekdays. This column is
derived using the MVS_SYSTEM_ID, DATE, and TIME columns as
parameters in the PERIOD function.
FF_ABORTS REAL The total number of Full Function transactions that aborted their
commits.
FF_COMMITS REAL The total number of Full function transactions that completed their
commits.
FP_ABORTS REAL The total number of Fast Path transactions that aborted their
commits.
FP_CI_NHNH_CONT REAL The total number of Fast Path CI contentions between non-HSSP
and non-HSSP EPCBs, derived from the count stored in the FP
syncpoint record (record type X'5937' or X'5938'). This represents
the number of CI contentions between non-HSSP and non-HSSP
EPCBs that the program experienced while the transactions were
active.
FP_COMBINATIONS REAL The total number of Fast Path combinations during logging of type
X'5950' records, derived from the count stored in the FP syncpoint
record (record type X'5937' or X'5938'). This represents the number
of combinations during logging of type X'5950' records that the
program experienced while the transactions were active.
FP_COMMITS REAL The total number of Fast Path transactions that completed their
commits.
FP_DEDB_BFR_WAITS REAL The total number of Fast Path waits for DEDB buffers, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of waits for DEDB buffers that
the program experienced while the transactions were active.
FP_DEDB_CALLS REAL The total number of Fast Path DEDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of waits for DEDB calls that the program
experienced while the transactions were active.
FP_DEDB_NH_PUTGET REAL The total number of Fast Path DEDB PUTs/GETs on area data sets,
derived from the count stored in the FP syncpoint record (record
type X'5937' or X'5938'). This represents the number of DEDB PUTs/
GETs on area data sets that the program experienced while the
transactions were active.
FP_LOGGED_CI REAL The total number of Fast Path whole Is logged, derived from the
count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of whole CIs logged by the
program while the transactions were active.

Chapter 4. Data table reference 71

IMS Performance Feature Reference

Column name Key Data type Description

FP_MSDB_CALLS REAL The total number of Fast Path MSDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of MSDB calls that the program issued
while the transactions were active.
FP_OVERFLOW_BFR REAL The total number of Fast Path overflow buffers used, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of overflow buffers used by
the program while the transactions were active.
INPUT_CSQ REAL The total number of input messages issued by transactions and BMP
programs queued through IMS CSQ.
INPUT_LOCAL REAL The total number of input messages issued by transactions and BMP
programs, not using CSQ.
INPUT_SEC REAL The total time, in seconds, that transactions and BMP programs
spent on the IMS input message queue, including input queue time
for program-to-program switch transactions.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
NETWORK_SEC REAL The total time that responding transactions spent in network
transmission to the ultimate destination, in seconds, as measured
using SNA definite response. This may also include user think time
to the next transaction, if the transaction is so defined in IMS.
OUTPUT_CSQ REAL The total number of output messages issued by transactions and
BMP programs queued through IMS CSQ.
OUTPUT_LOCAL REAL The total number of output messages issued by transactions and
BMP programs, not using CSQ.
OUTPUT_SEC REAL The total time that responding transactions spent on the IMS output
queue waiting for transmission to the originating network
destination, in seconds.
OUTPUT_CSQ_SEC REAL The time between the completed output transaction put on the
queue and the get from the queue for routing the output to the
terminal. It is always blank for APPC/OTMA transactions.
PGM_SWITCHES REAL Number of program-to-program switches, calculated when
secondary transactions are processed (SECONDARY=YES).
PGM_SWITCHES_SEC REAL The time between the X'07' of the root/children transaction and the
X'08' of the children/root transaction. It is calculated when
secondary transactions are processed (SECONDARY=YES).
PROCESS_SEC REAL The total elapsed time that transactions and BMP programs spent
processing in the dependent regions, in seconds.
RESPONSE_SEC REAL The total time, in seconds, that responding transactions spent in
network transmission to the ultimate destination, as measured using
SNA definite response plus host transit time.
RESPONSES REAL The total number of responding transactions and BMP programs that
sent messages to the originating terminal.
SEC_TRAN_EXCL REAL The number of secondary transactions that were not processed
(SECONDARY=NO). For each root transaction, this value is
calculated as the total number of X'03' records that have MSGQDES
flag set to X'81' (destination is an SMB) and are not CSQPUT.

72 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

SQ6_TIME REAL The total transaction time for subqueue 6, in
seconds, as stored in the DL/I GU (record type X'31') and
program termination (record type X'07') records. This
represents the total time spent waiting in a
wait-for- input or pseudo wait-for-input region with
no work to do.

SYSPLEX_NAME CHAR(8) This column is derived from the run time

parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRAN_CNTR_1 REAL The total number of IMS transactions whose transit time was less
than the user-specified boundary 1 (default for boundary is 0.7
seconds).
TRAN_CNTR_2 REAL The total number of IMS transactions whose transit time was less
than the user-specified boundary 2 (default for boundary is 1.5
seconds).
TRAN_CNTR_3 REAL The total number of IMS transactions whose transit time was less
than the user-specified boundary 3 (default for boundary is 3
seconds).
TRAN_CNTR_4 REAL The total number of IMS transactions whose transit time was less
than the user-specified boundary 4 (default for boundary is 10
seconds).
TRANSACTIONS REAL The total number of IMS transactions for the given interval and
unique key combination.
TRANSIT_SEC REAL The total time, in seconds, transactions and BMP programs spent in
the IMS system from first enqueue of the input message to first GU
of the responding output message (or transaction termination),
excluding the network transmission time.
VSO_NORMAL_BA REAL The Normal Buffer Allocation (NBA) value. From IMS record type
X'5937'.
VSO_OVER_BA REAL The Overflow Buffer Allocation (OBA) value. From IMS record type
X'5937'.
VSO_READS_DASD REAL Number of reads from DASD to dataspace. From IMS record type
X'5937' or X'5938'.
VSO_READS_DTSP REAL Number of reads from dataspace. From IMS record type X'5937' or
X'5938'.
VSO_UPDS_DTSP REAL Number of updates to dataspace. From IMS record type X'5937' or
X'5938'.

IMS_SYSTEM_TRAN2_H,_D, _M
Available for IMS Versions 9 and 10, the IMS_SYSTEM_TRAN2_H,_D, _M tables contain hourly, daily, and
monthly statistics on counts of transactions and response times. They contain information that include
data for message-queue-driven transactions and BMPs, EMH-driven Fast Path transactions, and message
switches. Use these tables to identify transaction utilization and subsequent elapsed time, transmission,
and queueing effects on the IMS system. Also use them to identify what users did, how their volumes
differed, and their response-time experiences.
For IMS Versions 9 and 10, these tables relate to the Extended System Tran Transit Time subcomponent.
For more information about this subcomponent, see page “Extended System Transaction Transit Time”
on page 5.
The default retention periods are:

Chapter 4. Data table reference 73

IMS Performance Feature Reference

IMS_SYSTEM_TRAN2_H
10 days
IMS_SYSTEM_TRAN2_D
45 days
IMS_SYSTEM_TRAN2_M
765 days

Column name Key Data type Description

DATE K DATE The date the activities occurred. For the _M table, this is the date of
the first day of the month.
TIME (*) K TIME The time when the activity started, in the format HH.00.00. This
applies only to the _H table.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 weekdays. This column is
derived using the MVS_SYSTEM_ID, DATE, and TIME columns as
parameters in the PERIOD function.
MVS_SYSTEM_ID K CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.
PROGRAM_NAME K CHAR(8) The name of the IMS application program used to process the
transaction. For full function and Fast Path activity, this column
contains the program specification block (PSB) if available. For APPC
activity this column contains the TPI used.
TRANSACTION_CLASS K CHAR(4) The assigned transaction class.
PERFORMANCE_GROUP K CHAR(1) The transaction type based on (DRLMINPUT+ DRLMPROCE) value
ranges. It could be:
F = FAST
(0.0 - 0.7)
G = GOOD
(0.7 - 1.5)
M = MEDIUM
(1.5 - 3.0)
L = LOW
(3.0 - 10.0)
S = SLOW
(above 10)

ORIGIN_IMS K CHAR(8) The IMS subsystem ID defined in the origin part of the UOW token. It
identifies the activity origin.
PROCESS_IMS K CHAR(8) The IMS subsystem ID defined in the processing part of the UOW
token. It identifies the activity processor.
REGION_JOB_NAME K CHAR(8) The MVS- and JES-identified job name for the IMS-dependent
region. This column uniquely identifies the transaction processing
activity for each region, because the region identifier or PST ID can
be reused by IMS.

74 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

PST_ID K CHAR(2) The IMS-assigned number for the partition specification table (PST)
that contains the management and control information for the
dependent region that processed the transaction. The PST can be
reused by IMS after a dependent region terminates, so region
occupancy and processing analysis are less meaningful if only the
region PST ID is used. So, you must also use the region job name
(REGION_JOB_NAME) to identify the dependent region.
APPC_MODE_NAME CHAR(8) Code used by EMH to enable transactions to be routed to programs
within LBL. From X'5901' record.
APPC_NETID CHAR(8) The destination Network ID for the APPC session.
FF_ABORTS REAL The total number of Full Function transactions that aborted their
commits.
FF_COMMITS REAL The total number of Full function transactions that completed their
commits.
FP_ABORTS REAL The total number of Fast Path transactions that aborted their
commits.
FP_CI_NHNH_CONT REAL The total number of Fast Path CI contentions between non-HSSP
and non-HSSP EPCBs, derived from the count stored in the FP
syncpoint record (record type X'5937' or X'5938'). This represents
the number of CI contentions between non-HSSP and non-HSSP
EPCBs that the program experienced while the transactions were
active.
FP_COMBINATIONS REAL The total number of Fast Path combinations during logging of type
X'5950' records, derived from the count stored in the FP syncpoint
record (record type X'5937' or X'5938'). This represents the number
of combinations during logging of type X'5950' records that the
program experienced while the transactions were active.
FP_COMMITS REAL The total number of Fast Path transactions that completed their
commits.
FP_DEDB_BFR_WAITS REAL The total number of Fast Path waits for DEDB buffers, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of waits for DEDB buffers that
the program experienced while the transactions were active.
FP_DEDB_CALLS REAL The total number of Fast Path DEDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of waits for DEDB calls that the program
experienced while the transactions were active.
FP_DEDB_NH_PUTGET REAL The total number of Fast Path DEDB PUTs/GETs on area data sets,
derived from the count stored in the FP syncpoint record (record
type X'5937' or X'5938'). This represents the number of DEDB PUTs/
GETs on area data sets that the program experienced while the
transactions were active.
FP_LOGGED_CI REAL The total number of Fast Path whole Is logged, derived from the
count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of whole CIs logged by the
program while the transactions were active.
FP_MSDB_CALLS REAL The total number of Fast Path MSDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of MSDB calls that the program issued
while the transactions were active.

Chapter 4. Data table reference 75

IMS Performance Feature Reference

Column name Key Data type Description

FP_OVERFLOW_BFR REAL The total number of Fast Path overflow buffers used, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of overflow buffers used by
the program while the transactions were active.
INPUT_CSQ REAL The total number of input messages issued by transactions and BMP
programs queued through IMS CSQ.
INPUT_LOCAL REAL The total number of input messages issued by transactions and BMP
programs, not using CSQ.
INPUT_SEC REAL The total time, in seconds, that transactions and BMP programs
spent on the IMS input message queue, including input queue time
for program-to-program switch transactions.
NETWORK_SEC REAL The total time that responding transactions spent in network
transmission to the ultimate destination, in seconds, as measured
using SNA definite response. This may also include user think time
to the next transaction, if the transaction is so defined in IMS.
OUTPUT_CSQ REAL The total number of output messages issued by transactions and
BMP programs queued through IMS CSQ.
OUTPUT_LOCAL REAL The total number of output messages issued by transactions and
BMP programs, not using CSQ.
OUTPUT_SEC REAL The total time that responding transactions spent on the IMS output
queue waiting for transmission to the originating network
destination, in seconds.
OUTPUT_CSQ_SEC REAL The time between the completed output transaction put on the
queue and the get from the queue for routing the output to the
terminal. It is always blank for APPC/OTMA transactions.
PGM_CPU_APPROX REAL The total dependent region CPU TCB seconds, derived form the
count of CPU timer units stored in the program termination record
(record type X'07') divided by 38400 (the number of time units per
CPU seconds). This column represents the sum of approximate
number of CPU seconds of program execution time while the
transactions were active. This value is not provided for WFI or PWFI
transactions.
PGM_SWITCHES REAL Number of program-to-program switches, calculated when
secondary transactions are processed (SECONDARY=YES).
PGM_SWITCHES_SEC REAL The time between the X'07' of the root/children transaction and the
X'08' of the children/root transaction. It is calculated when
secondary transactions are processed (SECONDARY=YES).
PROCESS_SEC REAL The total elapsed time that transactions and BMP programs spent
processing in the dependent regions, in seconds.
RESPONSE_SEC REAL The total time, in seconds, that responding transactions spent in
network transmission to the ultimate destination, as measured using
SNA definite response plus host transit time.
RESPONSES REAL The total number of responding transactions and BMP programs that
sent messages to the originating terminal.
SEC_TRAN_EXCL REAL The number of secondary transactions that were not processed
(SECONDARY=NO). For each root transaction, this value is
calculated as the total number of X'03' records that have MSGQDES
flag set to X'81' (destination is an SMB) and are not CSQPUT.

76 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

SQ6_TIME REAL The total transaction time for subqueue 6, in seconds, as stored in
the DL/I GU (record type X'31') and program termination (record
type X'07') records. This represents the total time spent waiting in a
wait-for- input or pseudo wait-for-input region with no work to do.
SYSPLEX_NAME CHAR(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANS_PRIOR CHAR(2) Message priority.
TRANSACTIONS REAL The total number of IMS transactions for the given interval and
unique key combination.
TRANSIT_SEC REAL The total time, in seconds, transactions and BMP programs spent in
the IMS system from first enqueue of the input message to first GU
of the responding output message (or transaction termination),
excluding the network transmission time.
VSO_NORMAL_BA REAL The Normal Buffer Allocation (NBA) value. From IMS record type
X'5937'.
VSO_OVER_BA REAL The Overflow Buffer Allocation (OBA) value. From IMS record type
X'5937'.
VSO_READS_DASD REAL Number of reads from DASD to dataspace. From IMS record type
X'5937' or X'5938'.
VSO_READS_DTSP REAL Number of reads from dataspace. From IMS record type X'5937' or
X'5938'.
VSO_UPDS_DTSP REAL Number of updates to dataspace. From IMS record type X'5937' or
X'5938'.

Note: (*) Field present only in the IMS_SYSTEM_TRAN2_H table.

IMS_TRAN_H, _D, _W
The IMS_TRAN_H, _D, _W tables contain hourly, daily, and weekly statistics on counts of transactions and
response times summarized by transaction name and user ID. They contain information that includes
data for message-queue-driven transactions and BMPs, EMH driven Fast Path transactions, and message
switches. Use these tables to identify transaction utilization and subsequent elapsed time, transmission,
and queuing effects on the IMS system. Use these table to also identify what users did, how their volumes
differed, and their response-time experiences.
These tables relate to the Transaction Transit Time subcomponent. For more information about this
subcomponent, see page Table 8 on page 55.
The default retention periods are:
IMS_TRAN_H
10 days
IMS_TRAN_D
45 days
IMS_TRAN_W
365 days

Column name Key Data type Description

DATE K DATE The date the activities occurred.
TIME K TIME The time when the activity started, in the format HH.00.00. This
applies only to the _H table.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.

Chapter 4. Data table reference 77

IMS Performance Feature Reference

Column name Key Data type Description

PROGRAM_NAME K CHAR(8) The name of the IMS application program used to process the
transaction. For full function and Fast Path activity, this column
contains the program specification block (PSB) if available. For APPC
activity this column contains the TPI used.
ROUTING_CODE K CHAR(8) Code used by EMH to enable transactions to be routed to programs
within LBL. From X'5901' record.
PST_ID K CHAR(2) The IMS-assigned number for the partition specification table (PST)
that contains the management and control information for the
dependent region that processed the transaction. The PST can be
reused by IMS after a dependent region terminates, so region
occupancy and processing analysis are less meaningful if only the
region PST ID is used. So, you must also use the region job name
(REGION_JOB_NAME) to identify the dependent region.
REGION_JOB_NAME K CHAR(8) The MVS- and JES-identified job name for the IMS dependent
region. This column uniquely identifies the transaction processing
activity for each region, because the region identifier or PST ID can
be reused by IMS.
TRANS_TYPE K CHAR(8) Activity type as detailed in “TRANS_TYPE key column” on page 81.
PERFORMANCE_GROUP K CHAR(1) The transaction type based on (DRLMINPUT+ DRLMPROCE) value
ranges. It could be:
F = FAST
(0.0 - 0.7)
G = GOOD
(0.7 - 1.5)
M = MEDIUM
(1.5 - 3.0)
L = LOW
(3.0 - 10.0)
S = SLOW
(above 10)

ORIGIN_IMS K CHAR(8) The IMS subsystem ID defined in the origin part of the UOW token. It
identifies the activity origin.
PROCESS_IMS K CHAR(8) The IMS subsystem ID defined in the processing part of the UOW
token. It identifies the activity processor.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred,
forexample, PRIME shift 08:00 to 17:00 weekdays. This column is
derived using the MVS_SYSTEM_ID, DATE, and TIME columns as
parameters in the PERIOD function.
ORIGIN_LTERM K CHAR(8) The IMS-defined logical name for the terminal used to request the
transaction or OTMA Tpipe name.
DESTINATION_LTERM K CHAR(8) The IMS-defined logical name for the terminal used to receive the
transaction output. Missing if APPC OTMA.
USER_ID K CHAR(8) The user identifier used to gain authorized access to IMS resources.
This column contains the logical terminal name if security is not
being managed by the IMS-supported /SIGN ON.
APPC_MODE_NAME CHAR(8) The mode name for the APPC session.
APPC_NETID CHAR(8) The destination Network ID for the APPC session.
FF_ABORTS REAL The total number of Full Function transactions that aborted their
commits.

78 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

FF_COMMITS REAL The total number of Full function transactions that completed their
commits.
FP_ABORTS REAL The total number of Fast Path transactions that aborted their
commits.
FP_CI_NHNH_CONT REAL The total number of Fast Path CI contentions between non-HSSP
and non-HSSP EPCBs, derived from the count stored in the FP
syncpoint record (record type X'5937' or X'5938'). This represents
the number of CI contentions between non-HSSP and non-HSSP
EPCBs that the program experienced while the transactions were
active.
FP_COMBINATIONS REAL The total number of Fast Path combinations during logging of type
X'5950' records, derived from the count stored in the FP syncpoint
record (record type X'5937' or X'5938'). This represents the number
of combinations during logging of type X'5950' records that the
program experienced while the transactions were active.
FP_COMMITS REAL The total number of Fast Path transactions that completed their
commits.
FP_DEDB_BFR_WAITS REAL The total number of Fast Path waits for DEDB buffers, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of waits for DEDB buffers that
the program experienced while the transactions were active.
FP_DEDB_CALLS REAL The total number of Fast Path DEDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of waits for DEDB calls that the program
experienced while the transactions were active.
FP_DEDB_NH_PUTGET REAL The total number of Fast Path DEDB PUTs/GETs on area data sets,
derived from the count stored in the FP syncpoint record (record
type X'5937' or X'5938'). This represents the number of DEDB PUTs/
GETs on area data sets that the program experienced while the
transactions were active.
FP_LOGGED_CI REAL The total number of Fast Path whole Is logged, derived from the
count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of whole CIs logged by the
program while the transactions were active.
FP_MSDB_CALLS REAL The total number of Fast Path MSDB calls, derived from the count
stored in the FP syncpoint record (record type X'5937' or X'5938').
This represents the number of MSDB calls that the program issued
while the transactions were active.
FP_OVERFLOW_BFR REAL The total number of Fast Path overflow buffers used, derived from
the count stored in the FP syncpoint record (record type X'5937' or
X'5938'). This represents the number of overflow buffers used by
the program while the transactions were active.
INPUT_CSQ REAL The total number of input messages issued by transactions and BMP
programs queued through IMS CSQ.
INPUT_LOCAL REAL The total number of input messages issued by transactions and BMP
programs, not using CSQ.
INPUT_SEC REAL The total time, in seconds, that transactions and BMP programs
spent on the IMS input message queue, including input queue time
for program-to-program switch transactions.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.

Chapter 4. Data table reference 79

IMS Performance Feature Reference

Column name Key Data type Description

NETWORK_SEC REAL The total time that responding transactions spent in network
transmission to the ultimate destination, in seconds, as measured
using SNA definite response. This may also include user think time
to the next transaction, if the transaction is so defined in IMS.
OUTPUT_CSQ REAL The total number of output messages issued by transactions and
BMP programs queued through IMS CSQ.
OUTPUT_LOCAL REAL The total number of output messages issued by transactions and
BMP programs, not using CSQ.
OUTPUT_SEC REAL The total time that responding transactions spent on the IMS output
queue waiting for transmission to the originating network
destination, in seconds.
OUTPUT_CSQ_SEC REAL The time between the completed output transaction put on the
queue and the get from the queue for routing the output to the
terminal. It is always blank for APPC/OTMA transactions.
PGM_CPU_APPROX REAL The total dependent region CPU TCB seconds, derived form the
count of CPU timer units stored in the program termination record
(record type X'07') divided by 38400 (the number of time units per
CPU seconds). This column represents the sum of approximate
number of CPU seconds of program execution time while the
transactions were active. This value is not provided for WFI or PWFI
transactions.
PGM_SWITCHES REAL Number of program-to-program switches, calculated when
secondary transactions are processed (SECONDARY=YES).
PGM_SWITCHES_SEC REAL The time between the X'07' of the root/children transaction and the
X'08' of the children/root transaction. It is calculated when
secondary transactions are processed (SECONDARY=YES).
PROCESS_SEC REAL The total elapsed time that transactions and BMP programs spent
processing in the dependent regions, in seconds.
RESPONSE_SEC REAL The total time, in seconds, that responding transactions spent in
network transmission to the ultimate destination, as measured using
SNA definite response plus host transit time.
RESPONSES REAL The total number of responding transactions and BMP programs that
sent messages to the originating terminal.
SEC_TRAN_EXCL REAL The number of secondary transactions that were not processed
(SECONDARY=NO). For each root transaction, this value is
calculated as the total number of X'03' records that have MSGQDES
flag set to X'81' (destination is an SMB) and are not CSQPUT.
SQ6_TIME REAL The total transaction time for subqueue 6, in seconds, as stored in
the DL/I GU (record type X'31') and program termination (record
type X'07') records. This represents the total time spent waiting in a
wait-for- input or pseudo wait-for-input region with no work to do.
SYSPLEX_NAME CHAR(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANS_PRIOR CHAR(2) Message priority.
TRANSACTIONS REAL The total number of IMS transactions for the given interval and
unique key combination.
TRANSIT_SEC REAL The total time, in seconds, transactions and BMP programs spent in
the IMS system from first enqueue of the input message to first GU
of the responding output message (or transaction termination),
excluding the network transmission time.

80 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

VSO_NORMAL_BA REAL The Normal Buffer Allocation (NBA) value. From IMS record type
X'5937'.
VSO_OVER_BA REAL The Overflow Buffer Allocation (OBA) value. From IMS record type
X'5937'.
VSO_READS_DASD REAL Number of reads from DASD to dataspace. From IMS record type
X'5937' or X'5938'.
VSO_READS_DTSP REAL Number of reads from dataspace. From IMS record type X'5937' or
X'5938'.
VSO_UPDS_DTSP REAL Number of updates to dataspace. From IMS record type X'5937' or
X'5938'.

TRANS_TYPE key column

The TRANS_TYPE key column, which is used in the IMS_TRAN_H table, contains transaction
characteristics. It also appears in the composite record produced by the R2 record procedure.
Each character in this 8-byte column has a specific meaning:
The first byte is the region type flag:
M
MPP
F
IFP
B
BMP
-
Not available
The second byte is the queue type flag:
C
Common CSQ
L
Local CSQ
-
Not shared
The third byte is the data communication type flag:
M
MSC
I
ISC
A
APPC
C
APPC and MSC, or OTMA and MSC
O
OTMA
-
Not available
The fourth byte is the thread management type flag:

Chapter 4. Data table reference 81

IMS Performance Feature Reference

W
WFI or PWFI
Q
Quick reschedule
M
Missing PSB usage values
-
Not available
The fifth byte is the program-to-program switch flag:
P
Primary
S
Secondary
-
Not available
The sixth byte is the mixed mode flag:
F
A transaction starting as Fast Path and ending as full function
P
A transaction starting as full function and ending as Fast Path
-
Not available
The seventh byte is the environment type flag:
J
Java™
C
CPI-C
-
Not available
The eighth byte is the transaction completion status flag:
R
Aborted and retried
C
Input canceled
A
Aborted, or program abended
-
Not available

IMS_TRAN_QUEUE_Q, _D
The IMS_TRAN_QUEUE_Q, _D tables provide quarter-hourly and daily statistics on IMS Message Queue
usage by IMS transactions.
These tables relate to the Transaction Transit Time subcomponent. For more information about this
subcomponent, see page “Transaction Transit Time” on page 5.
The default retention periods for these tables are:
IMS_TRAN_QUEUE_Q
10 days

82 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

IMS_TRAN_QUEUE_D
35 days

Column name Key Data type Description

DATE K DATE Date when the activity occurred.
TIME K TIME The time when the activity started, in the format HH.15.00. This
applies only to the _Q table.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 in the weekdays. This
column is derived using MVS_SYSTEM_ID, DATE and TIME
columns as parameters in the PERIOD function.
ORIGIN_IMS K CHAR(8) The IMS subsystem ID defined in the origin part of the UOW
token. It identifies the activity origin.
PROCESS_IMS K CHAR(8) The IMS subsystem ID defined in the processing part of the UOW
token. It identifies the activity processor.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.
PERFORMANCE_GROUP K CHAR(1) The transaction type based on (DRLMINPUT+ DRLMPROCE) value
ranges. It could be:
F = FAST
(0.0 - 0.7)
G = GOOD
(0.7 - 1.5)
M = MEDIUM
(1.5 - 3.0)
L = LOW
(3.0 - 10.0)
S = SLOW
(above 10)

QUEUE_TYPE K CHAR(12) Queue Type. It can be: MSGQ LOCAL, EMHQ LOCAL, MSGQ
SHARED.
INPUT_MSG REAL The total number of input messages processed.
INPUT_MSG_SEC REAL The total time, in seconds, the input messages spent on that
queue.
INP_EMHQ_AVG(*) REAL Average number of the messages on the input queue for fast path
transactions present before processing.
INP_EMHQ_MAX(*) REAL Maximum number of the messages on the input queue for fast
path transactions present before processing.
INP_EMHQ_MIN(*) REAL Minimum number of the messages on the input queue for fast
path transactions present before processing.
INP_MSGQ_SHMSG_AVG(*) REAL Average number of the messages on the input short message
queue for full function transactions present before processing.
INP_MSGQ_SHMSG_MAX(*) REAL Maximum number of the messages on the input short message
queue for full function transactions present before processing.
INP_MSGQ_SHMSG_MIN(*) REAL Minimum number of the messages on the input short message
queue for full function transactions present before processing.
INP_MSGQ_LGMSG_AVG(*) REAL Average number of the messages on the input long message
queue for full function transactions present before processing.
INP_MSGQ_LGMSG_MAX(*) REAL Maximum number of the messages on the input long message
queue for full function transactions present before processing.

Chapter 4. Data table reference 83

IMS Performance Feature Reference

Column name Key Data type Description

INP_MSGQ_LGMSG_MIN(*) REAL Minimum number of the messages on the input long message
queue for full function transactions present before processing.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run
time parameter :MVS_SYSTEM_ID or from the
IMS_SYSTEM_NAMES lookup table because IMS log records do
not contain this field..
OUTPUT_MSG REAL The total number of output messages processed.
OUTPUT_MSG_SEC REAL The total time, in seconds, the output messages spent on that
queue.
OUT_EMHQ_AVG(*) REAL Average number of the messages on the output queue for fast
path transactions present before processing.
OUT_EMHQ_MAX(*) REAL Maximum number of the messages on the output queue for fast
path transactions present before processing.
OUT_EMHQ_MIN(*) REAL Minimum number of the messages on the output queue for fast
path transactions present before processing.
OUT_MSGQ_SHMSG_AVG(*) REAL Average number of the messages on the output short message
queue for full function transactions present before processing.
OUT_MSGQ_SHMSG_MAX(*) REAL Maximum number of the messages on the output short message
queue for full function transactions present before processing.
OUT_MSGQ_SHMSG_MIN(*) REAL Minimum number of the messages on the output short message
queue for full function transactions present before processing.
OUT_MSGQ_LGMSG_AVG(*) REAL Average number of the messages on the output long message
queue for full function transactions present before processing.
OUT_MSGQ_LGMSG_MAX(*) REAL Maximum number of the messages on the output long message
queue for full function transactions present before processing.
OUT_MSGQ_LGMSG_MIN(*) REAL Minimum number of the messages on the output long message
queue for full function transactions present before processing.
SYSPLEX_NAME CHAR(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANSACTIONS REAL The total number of IMS transactions for the given interval and
unique key.

Note: The fields marked with (*) are loaded with the following rules:
• The IMS log collected contains an x'4001' record indicating an IMS cold start. The statistics on the
queue utilization will start after an IMS cold start has cleared these queues.
• The checkpoint file is necessary to maintain the statistics from one collection to the next. It is also
important that the logs do not have gaps between them.
• The rows relative to the Full Function or the Fast Path queue types will load only the relative fields. The
other fields will contain a NULL value.
• In the CSQ environment, these fields contain valid data only if the collection is done from a merged log.

IMS_TRAN_QUEUE_QV, _DV
These views provide quarter-hourly and daily statistics on IMS Message Queue usage by IMS
transactions. They are based on the IMS_TRAN_QUEUE_Q and _D table.
These views relate to the Transaction Transit Time subcomponent. For more information about the this
subcomponent, see page “Transaction Transit Time” on page 5.

84 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

DATE K DATE Date when the activity occurred.
TIME K TIME The time when the activity started, in the format HH.15.00. This
applies only to the _Q table.
PERIOD_NAME K CHAR(8) The name of the period or shift in which the activity occurred, for
example, PRIME shift 08:00 to 17:00 in the weekdays. This column
is derived using MVS_SYSTEM_ID, DATE and TIME columns as
parameters in the PERIOD function.
ORIGIN_IMS K CHAR(8) The IMS subsystem ID defined in the origin part of the UOW token.
It identifies the activity origin.
PROCESS_IMS K CHAR(8) The IMS subsystem ID defined in the processing part of the UOW
token. It identifies the activity processor.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction the user requested.
PERFORMANCE_GROUP K CHAR(1) The transaction type based on (DRLMINPUT+ DRLMPROCE) value
ranges. It could be:
F = FAST
(0.0 - 0.7)
G = GOOD
(0.7 - 1.5)
M = MEDIUM
(1.5 - 3.0)
L = LOW
(3.0 - 10.0)
S = SLOW
(above 10)

QUEUE_TYPE K CHAR(12) Queue Type. It can be: MSGQ LOCAL, EMHQ LOCAL, MSGQ SHARED.
INPUT_MSG_SEC_AVG REAL Average time spent by an input message on that queue. Calculated
as INPUT_MSG_SEC / INPUT_MSG.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
OUTPUT_MSG_SEC_AVG REAL Average time spent by an output message on that queue. Calculated
as OUTPUT_MSG_SEC / OUTPUT_MSG.
SYSPLEX_NAME CHAR(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
TRANS_RATE REAL Average transaction arrival rate over the interval. Calculated as
TRANSACTIONS / 900 (for quarter-hourly) or TRANSACTIONS /
86400 (for daily).

Note: In addition to the key and data columns described here, these views also contain all the data
columns described in "IMS_TRAN_QUEUE_Q/_D" on page “IMS_TRAN_QUEUE_Q, _D” on page 82.

KPM_IMS_TRAN_H,_D, _W
These tables contain hourly, daily, and weekly statistics from the IMS Transaction Level Statistics records
(56FA).
The default retention periods for these tables are:
KPM_IMS_SYSTEM_TRAN_H
10 days

Chapter 4. Data table reference 85

IMS Performance Feature Reference

KPM_IMS_SYSTEM_TRAN_D
60 days
KPM_IMS_SYSTEM_TRAN_W
500 days

Column name Key Data type Description

DATE K DATE The start date of the UOR (GMT). From the UTC field TPSTARTD.
TIME K TIME The minute when the activity started. From the UTC field TPSTARTT.
ORIGIN_IMS K CHAR(8) The IMS IMS system ID. From TPIMSID.
PSB_NAME K CHAR(8) The name of the IMS program used to process the transaction. This
column contains the program specification block (PSB). From
TPCPPSB.
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction user requested. From TPTRA.
REGION TYPE K CHAR(5) The type of IMS Region. From TPTYPE.
PROCESS_IMS K CHAR(8) The IMS IMS process ID. From TPCPOSSN.
PROGRAM_NAME K CHAR(8) Program name. From TPPGMNM.
PST_ID K CHAR(8) The IMS assigned number for partition spec table (PST) that
contains management and control information for dependent region
that processed the transaction. From TPCPSTN.
REGION_JOB_NAME K CHAR(8) The MVS and JES identified job for IMS dependent region. This
column uniquely identifies the tran processing activity for each
region, because the region id or PST ID can be reused by IMS.From
TPJOBN.
LTERM_NAME K CHAR(8) The name of the logical terminal on which the activity occurred.
From TPLTERM.
USER_ID K CHAR(8) User ID of the message processed in this dependent region. From
TPCPUSID. or if TPCPUSID is blank, zero, or xFF the User ID is set
from the value of TPTYPE: x01 = $MPP, X02 = $BMP, x10 = $CPIC-C,
x21 = $JMP, x22 = $JBL.
MVS_SYSTEM_ID CHAR(4) The MVS (SMF) system ID defined SYS1.PARMLIB (SMFPRMnn) by
the systems programmer. This column is derived from the run time
parameter :MVS_SYSTEM_ID or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
SYSPLEX_NAME CHAR(8) This column is derived from the run time
parameter :SYSPLEX_NAME or from the IMS_SYSTEM_NAMES
lookup table because IMS log records do not contain this field.
GROUP_NAME CHAR(8) Group name for non RACF users, otherwise set to unknown. From
TPCPGRPN.
TRANSACTION_CLASS CHAR(4) From TPCLASS.
TRANSACTIONS INTEGER Transaction count.
CPU_EXEC_TOT REAL Sum of CPU execution time for the Unit Of Recovery. From
TPEXTUOR.
CPU_EXEC_MIN REAL Minimum CPU execution time for the Unit Of Recovery. From
TPEXTUOR.
CPU_EXEC_MAX REAL Minimum CPU execution time for the Unit Of Recovery. From
TPEXTUOR.
ELAPSED_TOT REAL TOT of Elapsed Transaction time (microseconds).
ELAPSED_MIN REAL Min of Elapsed Transaction time (microseconds). From difference
between TPSTARTU AND TPENDU.

86 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

ELAPSED_MAX REAL Max of Elapsed Transaction time (microseconds). From difference
between TPSTARTU AND TPENDU.
ELAPSED_DB_IO_TOT REAL Sum of Elapsed Database IO time (microseconds). From TPTDBIO.
ELAPSED_DB_IO_MIN REAL The Minimum Elapsed Database IO time (microseconds). From
TPTDBIO.
ELAPSED_DB_IO_MAX REAL The Maximum Elapsed Database IO time (microseconds). From
TPTDBIO.
ELAPSED_DB_LK_TOT REAL Sum of Elapsed Database locking time (microseconds). From
TPTDBP.
ELAPSED_DB_LK_MIN REAL The Minimum Database locking time (microseconds). From
TPTDBPL.
ELAPSED_DB_LK_MAX REAL The Maximum Database locking time (microseconds). From
TPTDBPL.
DLIDB_GU_CALLS INTEGER The total number of DL/I database GU calls issued. Calculated as
Sum of TPDGU.
DLIDB_GN_CALLS INTEGER The total number of DL/I database GN calls issued. Calculated as
Sum of TPDGN.
DLIDB_GNP_CALLS INTEGER The total number of DL/I database GNP calls issued. Calculated as
Sum of TPDGNP.
DLIDB_GHU_CALLS INTEGER The total number of DL/I database GHU calls issued. Calculated as
Sum of TPDGHU.
DLIDB_GHN_CALLS INTEGER The total number of DL/I database GHN calls issued. Calculated as
Sum of TPDGHN.
DLIDB_GHNP_CALLS INTEGER The total number of DL/I database GHNP calls issued. Calculated as
Sum of TPDGHNP.
DLIDB_ISRT_CALLS INTEGER The total number of DL/I database ISRT calls issued. Calculated as
Sum of TPDISRT.
DLIDB_DLET_CALLS INTEGER The total number of DL/I database DLET calls issued. Calculated as
Sum of TPDDLET.
DLIDB_REPL_CALLS INTEGER The total number of DL/I database REPL calls issued. Calculated as
Sum of TPDREPL.
DLIDB_CALLS INTEGER The total number of DL/I database calls. Calculated as Sum of
TPDCMD.
DLI_MSG_GU INTEGER The total number of DL/I message queue GU calls. Calculated as
Sum of TPMGU.
DLI_MSG_GN INTEGER The total number of DL/I message queue GN calls. Calculated as
Sum of TPMGN.
DLI_MSG_ISRT INTEGER The total number of DL/I message queue insert calls. Calculated as
Sum of TPMISRT.
DLI_MSG_PURG INTEGER The total number of DL/I message queue PURGE calls. Calculated
as Sum of TPMPURG.
DLI_TEST_ENQUEUES INTEGER The total number of DL/I test enqueues. Calculated as Sum of
TPTSTNQ.
DLI_TEST_ENQWAITS INTEGER The total number of DL/I waits on test enqueues. Calculated as Sum
of DPTSTWT.
DLI_TEST_DEQUEUES INTEGER The total number of DL/I test dequeues. Calculated as Sum of
TPTSTDQ.

Chapter 4. Data table reference 87

IMS Performance Feature Reference

Column name Key Data type Description

DLI_Q_ENQUEUES INTEGER The total number of DL/I update enqueues. Calculated as Sum of
TPQCONQ.
DLI_Q_ENQWAITS INTEGER The total number of DL/I waits on update and enqueues. Calculated
as Sum of TPQCOWT.
DLI_Q_DEQUEUES INTEGER The total number of DL/I update dequeues. Calculated as Sum of
TPQCODQ.
DLI_UPDT_ENQUEUES INTEGER The total number of DL/I update enqueues. Calculated as Sum of
TPSUPNQ.
DLI_UPDT_ENQWAITS INTEGER The total number of DL/I waits on update and enqueues. Calculated
as Sum of TPSUPWT.
DLI_UPDT_DEQUEUES INTEGER The total number of DL/I update dequeues. Calculated as Sum of
TPSUPDQ.
DLI_EXCL_ENQUEUES INTEGER The total number of DL/I exclusive enqueue calls. Calculated as
Sum of TPEXCNQ.
DLI_EXCL_ENQWAITS INTEGER The total number of waits DL/I-exclusive enqueue calls. Calculated
as Sum of TPEXCWT.
DLI_EXCL_DEQUEUES INTEGER The total number of DL/I exclusive dequeue calls. Calculated as
Sum of TPEXCDQ.
DLI_CMD_CALLS INTEGER The total number of DL/I CMD calls. Calculated as Sum of TPMCMD.
DLI_GCMD_CALLS INTEGER The total number of DL/I GCMD calls. Calculated as Sum of
TPMGCMD.
DLI_MSG_CHNG_CALLS INTEGER Number of DL/I message CHNG calls. Calculated as Sum of
TPMCHNG.
DLI_MSG_AUTH_CALLS INTEGER Number of DL/I message AUTH calls. Calculated as Sum of
TPMAUTH
DLI_MSG_SETO_CALLS INTEGER Number of DL/I message SETO calls. Calculated as Sum of
TPMSETO.
DLI_APSB_CALLS INTEGER Number of DL/I APSB calls. Calculated as Sum of TPSAPSB.
DLI_DPSB_CALLS INTEGER Number of DL/I DPSB calls. Calculated as Sum of TPSDPSB.
DLI_GMSG_CALLS INTEGER Number of DL/I GMSG calls. Calculated as Sum of TPSGMSG.
DLI_ICMD_CALLS INTEGER Number of DL/I ICMD calls. Calculated as Sum of TPSICMD.
DLI_RCMD_CALLS INTEGER Number of DL/I RCMD calls. Calculated as Sum of TPSRCMD.
DLI_CHKP_CALLS INTEGER Number of DL/I CHKP calls. Calculated as Sum of TPSCHKP.
DLI_XRST_CALLS INTEGER Number of DL/I XRST calls. Calculated as Sum of TPSXRST.
DLI_ROLB_CALLS INTEGER Number of DL/I ROLB calls. Calculated as Sum of TPSROLB.
DLI_ROLS_CALLS INTEGER Number of DL/I ROLS calls. Calculated as Sum of TPSROLS.
DLI_SETS_CALLS INTEGER Number of DL/I SETS calls. Calculated as Sum of TPSSETS.
DLI_SETU_CALLS INTEGER Number of DL/I SETU calls. Calculated as Sum of TPSSETU.
DLI_INIT_CALLS INTEGER Number of DL/I INIT calls.Calculated as Sum of TPSINIT.
DLI_INQY_CALLS INTEGER Number of DL/I INQY calls. Calculated as Sum of TPSINQY.
DLI_SLOG_CALLS INTEGER Number of DL/I SLOG calls. Calculated as Sum of TPSLOG.
DLIDB_DEQ_CALLS INTEGER The total number of DL/I database deque calls. Sum of TPDDEQ.
TLS_VSAM_READS INTEGER Number of vsam read calls. From sum of TPVSAMR.

88 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Column name Key Data type Description

TLS_VSAM_WRITE INTEGER Number of vsam write calls. From sum of TPVSAMW.
TLS_OSAM_READS INTEGER Number of osam read calls. From sum of TPOSAMR.
TLS_OSAM_WRITE INTEGER Number of osam write calls. From sum of TPOSAMW.
TLS_TOT_VSOSIO INTEGER Sum of vsam and osam calls.
TLS_TOT_ESAF FLOAT Number of esaf calls. From sum of TPTOTIO.
TLS_FAST_FLD INTEGER Number of Field calls. From sum of TPFLD.
TLS_FAST_POS INTEGER Number of fastpath position calls. From sum of TPPOS.
TLS_RLSE_CALLS INTEGER Number of release calls. From sum of TPRLSE.
TLS_XQERY_SAVE INTEGER Number of xquery save calls. From sum of TPXSAVE.
TLS_XQERY_RSTR INTEGER Number of xquery restore calls. From sum of TPXRSTR.
TLS_XQERY_COPY INTEGER Number of xquery copy calls. From sum of TPXCOPY.
TLS_ICAL_CALLS INTEGER Number of ICAL calls. From sum of TPSICAL.
TLS_GUR_CALLS INTEGER Number of GUR calls. From sum of TPDGURL (V12+).

KPM_IMS_ATF_T
This table contains IMS Application Tracing Facility information on the transaction level.
The default retention period is 10 days.

Column name Key Data type Description

START_TIMESTAMP K TIMESTAM The start date of the UOR (GMT) from the UTC field ATFCORT.
P
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction user requested. From ATFTRAN.
PSB_NAME K CHAR(8) The name of the IMS program used to process the transaction. This
column contains the program specification block (PSB). From
ATFPSB.
LTERM_NAME K CHAR(8) The name of the logical terminal on which the activity occurred.
From ATFLTERM.
USER_ID K CHAR(8) The user ID. From ATFUSRID.
JOB_NAME K CHAR(8) The job name. From ATFJNAM.
REGION_TYPE CHAR(5) Type of IMS Region. From ATFRTYPE.
TRANSACTION_CLASS CHAR(4) Transaction class. From ATFCLASS.
QUICK_SCHEDULE CHAR(1) Quick scheduled. Possible values: Y or N. From ATFSCHF1.
PRIMED_MESSAGE CHAR(1) Primed message. Possible values: Y or N. From ATFVSFLG.
SPA_SIZE INTEGER The current SPA size. From ATFSPAL.
ELAPSED_TIME FLOAT Total elapsed time, in seconds. Calculated as the sum of ATFELAPS/
1E6.
TOTAL_CPU_TIME FLOAT Total CPU time, in seconds. Calculated as the sum of ATFCPU/1E6.
REGION_CPU_TIME FLOAT The total CPU time, in seconds, in the DEP region. Calculated as the
sum of ATFCPUD/1E6.
DLI_CPU_TIME FLOAT Total DLI CPU time, in seconds. Calculated as the sum of ATFCPUI/
1E6.

Chapter 4. Data table reference 89

IMS Performance Feature Reference

Column name Key Data type Description

DB2_CPU_TIME FLOAT Total Db2 CPU time, in seconds. Calculated as the sum of ATFCPUB/
1E6.
MQ_CPU_TIME FLOAT Total MQ CPU time, in seconds. Calculated as the sum of ATFCPUQ/
1E6.
CREGION_CPU_TIME FLOAT Total CPU time, in seconds, in CTL region. Calculated as the sum of
ATFCPUC/1E6.
DLISAS_CPU_TIME FLOAT Total CPU time, in seconds, in DLISAS region. Calculated as the sum
of ATFCPUS/1E6.
OREGION_CPU_TIME FLOAT Total CPU time, in seconds, in all other regions. Calculated as the
sum of ATFCPUO/1E6.
ELAPSED_DLI_TIME FLOAT Total DLI elapsed time, in seconds. Calculated as the sum of
ATFELTI/1E6.
ELAPSED_DB2_TIME FLOAT Total Db2 elapsed time, in seconds. Calculated as the sum of
ATFELTB/1E6.
ELAPSED_MQ_TIME FLOAT Total MQ elapsed time, in seconds. Calculated as the sum of
ATFELTQ/1E6.

KPM_IMS_ATF_EVT_T
This table contains IMS Application Tracing Facility information on transaction events.

Column name Key Data type Description

START_TIMESTAMP K TIMESTAM The start date of the UOR (GMT) from the UTC field ATFCORT.
P
TRANSACTION_NAME K CHAR(8) The name of the IMS transaction user requested. From ATFTRAN.
PSB_NAME K CHAR(8) The name of the IMS program used to process the transaction. This
column contains the program specification block (PSB). From
ATFPSB.
LTERM_NAME K CHAR(8) The name of the logical terminal on which the activity occurred.
From ATFLTERM.
USER_ID K CHAR(8) The user ID. From ATFUSRID.
JOB_NAME K CHAR(8) The job name. From ATFJNAM.
EVENT_NAME K CHAR(8) Event name. Possible values: DLI DB, DLI TM, Db2 SQL, MQ and
OTHER.
EVENT_TYPE K CHAR(20) Event type. From ATFXS*T.
DATABASE_NAME CHAR(8) Database name. From ATFXSDBN.
EVENT_GROUP CHAR(8) IMS DLI group bucket name. Possible values: READ, INSERT,
UPDATE, DELETE (DB) or GU, GN, ISRT, PURG (TM).
EVENT_COUNT SMALLINT Number of events. From ATFXS*N.
ELAPSED_TIME_TOT FLOAT Total elapsed time, in seconds. Calculated as the sum of ATFXS*E/
1E6.
ELAPSED_TIME_AVG FLOAT Average elapsed time, in seconds. Calculated as the average of
ATFXS*E/1E6.
CPU_TIME_TOT FLOAT Total CPU time, in seconds. Calculated as the sum of ATFXS*C/1E6.
CPU_TIME_AVG FLOAT Average CPU time, in seconds. Calculated as the average of
ATFXS*C/1E6.

90 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Lookup tables
This section describes the lookup table.

IMS_AVAIL_RESOURCE
This lookup table defines the IMS resources that are used for tracking availability. It also contains the
schedule names and availability objectives to use for the different resources.

Column name Key Data type Description

SYSPLEX_NAME K Char(8) Sysplex name that the resource is associated with. This can contain
global search characters.
MVS_SYSTEM_ID K Char(4) MVS system ID that the resource is associated with. This can contain
global search characters.
IMS_SYSTEM_ID K Char(8) IMS subsystem ID that the resource is associated with. This can
contain global search characters.
RESOURCE_SOURCE_NM K Char(8) Resource source name. This specifies the resource you want to
monitor. This can contain global search characters.
RESOURCE_TYPE K Char(8) Resource type:
IMSSYS
IMS system
IMSREG
IMS region
IMSPGM
IMS program
IMSTRAN
IMS transaction
IMSPGM and IMSTRAN will be used for tracking application.

AVAIL_OBJ_PCT Decimal Availability objective for the resource, in percentage.

(4,1)
RESOURCE_CHKPOINT Integer Resource checkpoint interval in seconds.
RESOURCE_TARGET_NM Char(18) Resource target name. This is the final name that will be associated
with the resource. For IMSPGM and IMSTRAN resource types it will
be used to specify the application name. If nothing is present,
$UNKNOWN is used as default.
SCHEDULE_NAME Char(8) Schedule name to use for the resource. If nothing is specified,
STANDARD is used as default.

Example of table contents

MVS IMS RESOURCE RESOURCE AVAIL

SYSPLEX SYSTEM SYSTEM SOURCE RESOURCE TARGET RESOURCE SCHEDULE OBJ
NAME ID ID NM TYPE NM CHKPOINT NAME PCT
-------- -------- -------- ---------- ----------- ------------ --------- --------- -------
% % % IMS71A IMSSYS --- 3600 STANDARD 95.0
% % % DSWM% IMSREG --- 3600 STANDARD 95.0
% % % MPP% IMSREG MPP 3600 STANDARD 95.0
% % % PGM% IMSPGM APPL1 --- STANDARD 95.0
% % % TRAN% IMSTRAN APPL1 --- STANDARD 95.0

IMS_SYSTEM_NAMES
This lookup table defines the IMS system names.

Chapter 4. Data table reference 91

IMS Performance Feature Reference

Table 9. Descriptions of the IMS_SYSTEM_NAMES lookup table

Column name Key Data type Description
IMS_SYSTEM_ID K Char(8) IMS system ID
associated with the
DRLLOGn data set
SYSPLEX_NAME Char(8) Name of the sysplex
MVS_SYSTEM_ID Char(4) The MVS system ID on
which the IMS system
runs
IMS_CNTL_REGION Char(8) The name of the IMS
control region
associated with the IMS
system ID
IMS_APPLID Char(8) The VTAM application ID
name

Lookup table IMS_SYSTEM_NAMES is not provided with any default values.

If you are collecting data from multiple systems, IMS_SYSTEM_NAMES must contain information for all of
these systems.
The following sample shows how you can populate the lookup table::

IMS_SYSTEM_ID SYSPLEX_NAME MVS_SYSTEM_ID IMS_CNTL_REGION IMS_APPLID

% ? ? ? ?
IMSA MYPLEX MVS1 REGIONA IMS1
IMSB MYPLEX MVS2 REGIONB IMS2
IMSC MYPLEX MVS3 REGIONC IMS3

For more instructions about using the administration dialog to edit the contents of this lookup table, see
the chapter that explains working with tables and update definitions in the Administration Guide and
Reference.
The following steps describe how IBM Z Performance and Capacity Analytics populates column
IMS_SYSTEM_ID during the collect:
• IBM Z Performance and Capacity Analytics derives IMS_SYSTEM_ID from the IMS records.
• When IMS_SYSTEM_ID cannot be derived from the IMS records, column IMS_SYSTEM_ID is populated
based on the following conditions:
– When IMS_SYSTEM_ID is specified with the SET statement, IMS_SYSTEM_ID is used to populate
column IMS_SYSTEM_ID.
– When IMS_SYSTEM_ID is not specified with the SET statement, column IMS_SYSTEM_ID is set to
'$UNKNOWN'.
The following steps describe how IBM Z Performance and Capacity Analytics populates columns
SYSPLEX_NAME, MVS_SYSTEM_ID, IMS_CTRL_REGION, and IMS_APPLID during the collect:
• IBM Z Performance and Capacity Analytics uses lookup table IMS_SYSTEM_NAMES to obtain values for
these columns.
• When values are not available for these columns in lookup table IMS_SYSTEM_NAMES, these columns
are populated based on the following conditions:
– When values are specified with the SET statement, these values are used to populate these columns.
– When values are not specified with the SET statement, the columns are set to '$UNKNOWN'.

92 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Note: To use the IMS_SYSTEM_NAMES lookup table for augmenting IBM Z Performance and Capacity
Analytics data tables, you must collect raw IMS log records, and not the DRLIMS07 data set to input the
preprocessed X'07' records.

Mapping between table fields and CSQ records

The following sections map the columns of the tables with the CSQ records from which they are derived.

IMS_TRAN_x (H, D, W)
Table 10. IMS_TRAN_x (H, D, W) Description
Column name Column description CSQ record fields
DESTINATION_LTERM The IMS-defined logical name for the terminal used to DRL_TRANT1, DRL_TERM
receive the transaction output. Missing if APPC OTMA. from CSQ_Vxxx_R2
ROUTING_CODE Code used by EMH to enable transactions to be routed to DRL_TRANT1,
programs within LBL. From X5901 record. DRL_ROUTCD from
CSQ_Vxxx_R2
PGM_SWITCHES Number of program-to-program switches. DRLPGMSWN from
CSQ_Vxxx_R2
PGM_SWITCHES_SEC It is the time from the x07 of the root/children transaction DRL_TRANT1,
and the x08 of the children/root transaction. DRLPGMSWS from
CSQ_Vxxx_R2
SQ6_TIME The total transaction time for subqueue 6, in seconds, as DRLSQ6TME from
storedin the DL/I GU (record type X31) and program CSQ_Vxxx_R2
termination(record type X07) records. This represents the
total time spent waiting in a wait-for-input or pseudo
wait-for-input region with no work to do.
TRANS_PRIOR Message priority. DRL_TXPRI from
CSQ_Vxxx_R2

IMS_TRAN_QUEUE_x (Q, D)
Table 11. IMS_TRAN_QUEUE_x (Q, D) Description
Column name Column description CSQ record fields
QUEUE_TYPE Queue Type. It can be: MSGQ LOCAL, EMHQ LOCAL, DRL_TRANT1
MSGQ SHARED. DRL_TRANT2from
CSQ_Vxxx_R2
INPUT_MSG The total number of input processed. DRL_TXFLOW
DRL_TXFLAGfrom
CSQ_Vxxx_R2
INPUT_MSG_SEC The total time, in seconds, the input messages spent on DRL_TRANT1
that queue DRLMINPUT from
CSQ_Vxxx_R2
INP_EMHQ_MAX Maximum number of the messages on the input queue for DRL_TRANT1
fast path transactions present before processing. DRLIEMHQMfrom
CSQ_Vxxx_R2
INP_EMHQ_MIN Minimum number of the messages on the input queue for DRL_TRANT1
fast path transactions present before processing. DRLIEMHQMfrom
CSQ_Vxxx_R2
OUTPUT_MSG The total number of output processed. DRL_TXFLAGfrom
CSQ_Vxxx_R2

Chapter 4. Data table reference 93

IMS Performance Feature Reference

Table 11. IMS_TRAN_QUEUE_x (Q, D) Description (continued)

Column name Column description CSQ record fields
OUTPUT_MSG_SEC The total time, in seconds, the output messages spent on DRL_TRANT1
that queue. DRLMOUTPUT from
CSQ_Vxxx_R2
OUT_EMHQ_AVG Average number of the messages on the output queue for DRL_TRANT1
fast path transactions present before processing DRLOEMHQM
DRL_TXFLOW from
CSQ_Vxxx_R2
OUT_EMHQ_MAX Maximum number of the messages on the output queue DRL_TRANT1
forfast path transactions present before processing DRLOEMHQM from
CSQ_Vxxx_R2
OUT_EMHQ_MIN Minimum number of the messages on the output queue DRL_TRANT1
for fast path transactions present before processing. DRLOEMHQM from
CSQ_Vxxx_R2

IMS_SYSTEM_TRAN_x (H, D)
Table 12. IMS_SYSTEM_TRAN_x (H, D) Description
Column name Column description CSQ record fields
PGM_SWITCHES Number of program-to-program switches. DRLPGMSWN from
CSQ_Vxxx_R2
PGM_SWITCHES_SEC It is the time from the x07 of the root/children transaction DRL_TRANT1,
and the x08 of the children/root transaction. DRLPGMSWS from
CSQ_Vxxx_R2
SQ6_TIME The total transaction time for subqueue 6, in seconds, as DRLSQ6TME from
storedin the DL/I GU (record type X31) and program CSQ_Vxxx_R2
termination (record type X07) records. This represents
the total time spent waiting in a wait-for- input or pseudo
wait-for-input region with no work to do.

IMS_PSB_ACCOUNT_x (H, D, W)
Table 13. IMS_PSB_ACCOUNT_x (H, D, W) Description
Column name Column description CSQ record fields
DLI_APSB_CALLS Number of DL/I APSB calls, derived from the count stored DLRAPSB from
in the program termination record (record type X07). CSQ_Vxxx_ST07
Calculated as Sum of DLRAPSB.
DLI_CHKP_CALLS Number of DL/I CHKP calls, derived from the count DLRCHKP from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRCHKP.
DLI_DPSB_CALLS Number of DL/I DPSB calls, derived from the count stored DLRDPSB from
in the program termination record (record type X07). CSQ_Vxxx_ST07
Calculated as Sum of DLRDPSB.
DLI_GMSG_CALLS Number of DL/I GMSG calls, derived from the count DLRGMSG from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07).Calculated as Sum of DLRGMSG.
DLI_ICMD_CALLS Number of DL/I ICMD calls, derived from the count DLRICMD from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRICMD.

94 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 IMS Performance Feature Reference

Table 13. IMS_PSB_ACCOUNT_x (H, D, W) Description (continued)

Column name Column description CSQ record fields
DLI_RCMD_CALLS Number of DL/I RCMD calls, derived from the count DLRRCMD from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRRCMD.
DLI_xRST_CALLS Number of DL/I XRST calls, derived from the count DLRXRST from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07).Calculated as Sum of DLRXRST.
DLI_ROLB_CALLS Number of DL/I ROLB calls, derived from the count DLRROLB from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRROLB.
DLI_ROLS_CALLS Number of DL/I ROLS calls, derived from the count DLRROLS from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRROLS.
DLI_SETS_CALLS Number of DL/I SETS calls, derived from the count stored DLRSETS from
in the program termination record (record type X07). CSQ_Vxxx_ST07
Calculated as Sum of DLRSETS.
DLI_SETU_CALLS Number of DL/I SETU calls, derived from the count stored DLRSETU from
in the program termination record (record type X07). CSQ_Vxxx_ST07
Calculated as Sum of DLRSETU.
DLI_INIT_CALLS Number of DL/I INIT calls, derived from the count stored DLRINIT from
inthe program termination record (record type CSQ_Vxxx_ST07
X07).Calculated as Sum of DLRINIT.
DLI_INQY_CALLS Number of DL/I INQY calls, derived from the count DLRINQY from
storedin the program termination record (record type CSQ_Vxxx_ST07
X07). Calculated as Sum of DLRINQY.
DLI_SLOG_CALLS Number of DL/I SLOG calls, derived from the count stored DLRSLOG from
in the program termination record (record type X07). CSQ_Vxxx_ST07
Calculated as Sum of DLRSLOG.
TRANS_PRIOR TRANS_PRIOR CHAR(2) priority. L0APRTY from
CSQ_Vxxx_ST07

Chapter 4. Data table reference 95

IMS Performance Feature Reference

96 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Chapter 5. Report reference

This section describes the reports provided with the collect component of the IMS Performance Feature.
They include management, service level, performance, and problem-related reports known to be useful in
monitoring and analyzing IMS-related activity.
These reports are intended to be a subset of the reports you use to analyze your IMS activity.
IBM Z Performance and Capacity Analytics produces reports based on the data in the IBM Z Performance
and Capacity Analytics database. Reports can show data from tables or from views. You can request
reports using the product reporting dialog or by submitting batch jobs. Typically, you use online reporting
for reports that you use once, and batch reporting for regularly required reports.

Report format and general description

This section describes the elements that are common among IBM Z Performance and Capacity Analytics
feature reports:
• Report title
• Report ID
• Report group
• Source
• Attributes
• Variables
• Report types
• Standard report formats

Report title
Each report has a title. Each report title begins with an abbreviation that identifies the component. IMS
Performance Feature reports begin with IMS. The rest of the title describes the report.

Report ID
Each report has a unique report identifier. The report ID consists of:
• The prefix IMS or CSQ.
• A one-character identifier of the IMS Performance Feature subcomponent that provides the report:
T
Transaction subcomponent report
Y
System subcomponent report
A
Application subcomponent report
S
Statistics subcomponent report
O
HALDB OLR subcomponent report
• Sequential numbers given to the reports in a subcomponent; for example, IMST03.

© Copyright IBM Corp. 1994, 2017 97

Reports

Report group
To make it easier to find reports, IBM Z Performance and Capacity Analytics organizes reports into report
groups, which correspond to feature components. IMS Performance Feature reports belong to the IMS
and CSQ report groups.

Source
Each report contains information adapted from a Db2 table. The table name is listed for each report.

Attributes
Each report has certain attributes associated with it. The attributes enable you to search for reports using
the dialog. These attributes are supplied for each report:
• The area the report belongs to (for example IMS, VM, or NETWORK)
• The tasks that the report supports:
Performance
Performance control task
Service
Service level planning task
Capacity
Capacity planning task
Security
Security control task
Configuration
Configuration management discipline
Operation
Operations management discipline
Change
Change management discipline
Problem
Problem management discipline
These are also specified where appropriate:
• Resource types reported (for example, storage or CPU)
• Performance issue reported (for example, availability or response)
• Presentation forms (for example, trend or overview)
• Time resolution in the report (hourly, daily, weekly, or monthly)

Variables
Each report has several variables associated with it. When you select a report to display, IBM Z
Performance and Capacity Analytics prompts you for the variables listed in the description.

Report types
The IMS Performance Feature produces these types of reports:
Overview
An overview report lists status for all resources of the specified type.
Trend
A trend report gives information about the behavior of a resource over a specified period.

98 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Detail (or technical)

A detail report presents detailed information on a selected resource. Use this type of report to get as
much information as possible in a critical situation.
Worst case
A worst case report lists the resources (usually a maximum of 15) with the worst performance record.
However, a worst case report does not imply that the resources listed have a negative performance
record, only that the performance has been worse for these than for other resources of the same type.

Standard report formats

Reports are presented in tabular or graphic format. All reports have the same basic report layout. Tabular
reports are low-resolution reports that show information in a table format. Graphic reports are high-
resolution graphs that give a pictorial representation of the data.

Tabular reports
Figure 17 on page 99 shows an example of a tabular report.

MVSPM, DASD Activity Report

Date: '2019-06-11' System: 'MVS1'
Period: 'PRIME'

I/O Response Response Queue Connect Disconn Pending Allocs Dev

Period Model LCU Device rate avg avg avg avg avg avg avg busy
name name number number Volser /sec mpl msec msec msec msec msec no %
-------- -------- ------ ------ ------ -------- -------- -------- ------ ------- ------- ------- ------ ------
PRIME - 0F 014E DBSL05 6.9 0.076 11.3 3.3 3.1 0.1 4.8 119.4 5.44
PRIME - 09 0105 COR105 9.1 0.075 8.2 0.4 1.8 0.1 5.8 66.6 7.01
PRIME - 0F 0140 TSOL01 5.4 0.036 6.1 0.3 3.2 0.1 2.5 122.8 3.41
PRIME - 0F 0142 TSOL03 4.7 0.035 6.8 0.4 2.6 0.1 3.7 176.8 3.19
PRIME - 09 0112 SF2US3 1.1 0.033 27.6 1.6 12.5 0.4 13.1 55.6 2.90
PRIME - 0F 0148 TSOL05 2.8 0.031 9.7 2.3 2.4 0.2 4.8 74.6 2.08
PRIME - 0F 0143 TSOL04 8.2 0.029 4.6 0.4 2.2 0.3 1.6 154.4 2.63
PRIME - 09 0106 COR106 1.2 0.025 20.4 0.9 3.6 0.3 15.6 826.6 2.33
PRIME - 0F 0141 TSOL02 3.2 0.023 6.6 0.2 2.7 0.1 3.7 120.2 2.19
PRIME - 0F 014B ICR003 9.7 0.021 2.7 0.1 2.2 0.0 0.3 2.9 1.91

IBM Z Performance and Capacity Analytics Report: MVSPM05

Figure 17. Tabular reports example

Graphic reports
In some cases, the meaning of data is best presented in graphic form. Graphic reports in IBM Z
Performance and Capacity Analytics have both a QMF format and a Graphical Data Display Manager
(GDDM) format. Figure 18 on page 100 illustrates a graphic report.

Chapter 5. Report reference 99

Reports

IMS Application Response Time Trend

Date: '2019-02-01' to '2019-03-30'
System: MVS2 IMS System: IMS2
Appl: 'C2C201 PPSC2A01'
100
100%
Response
80 time
boundary
4
Response
60 time
boundary
3
40 Response
time
boundary
2
20 Response
time
boundary
0 1

-21 2-2
8
3-0
6
3-1
3
3-2
0
3-2
7
-02 -0 -0 -0 -0 -0
19 19 19 19 19 19
20 20 20 20 20 20
Week start date

Figure 18. Graphic reports example

For complete information on QMF, refer to the QMF Learner's Guide and to the QMF Advanced User's
Guide.

Samples of reports across non-CSQ and CSQ tables

If you have existing non-CSQ reports that you want to continue to use with the IMS Performance Feature,
you may have to create new reports to provide equivalent functionality. This is because the IMS
Performance Feature uses tables with different fields and structures compared to the non-CSQ tables.
Also, if you wish to report on data from both non-CSQ and IMS Performance Feature tables, then you will
need to develop customized reports to combine the data.
Here is a sample using the IMS-shipped report IMSY01: "IMS System Response Time Trend Report". The
Query associated with this report is contained in the member DRLQIY01, as shown in Figure 19 on page
101.

100 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

-----------------------------------------------------------------------
-- Name: DRLQIY01
--
-- Function:
-- Define IMS report SQL query.
------------------------------------------------------------------------
SELECT
MVS_SYSTEM_ID
, IMS_SYSTEM_ID
, DATE
, VALUE(
((SUM(EMH_TRAN_CNTR_1) + SUM(MSGQ_TRAN_CNTR_1))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_2) + SUM(MSGQ_TRAN_CNTR_2))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_3) + SUM(MSGQ_TRAN_CNTR_3))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_4) + SUM(MSGQ_TRAN_CNTR_4))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
FROM &PREFIX.IMS_SYSTEM_D
WHERE
MVS_SYSTEM_ID = &MVS_SYSTEM_ID
AND IMS_SYSTEM_ID = &IMS_SYSTEM_ID
AND DATE >= &FROM_DATE
AND DATE <= &TO_DATE
GROUP BY
MVS_SYSTEM_ID
, IMS_SYSTEM_ID
, DATE

Figure 19. IMSY01 Report Query (DRLQIY01 member)

This query uses the following fields from the IMS_SYSTEM_D non-CSQ table:
• MVS_SYSTEM_ID
• IMS_SYSTEM_ID
• EMH_TRAN_CNTR_x 1=< x =< 4
• EMH_TRANSACTIONS
• MSGQ_TRAN_CNTR_x 1=< x =< 4
• MSGQ_TRANSACTIONS
Once you have set up the new CSQ implementation engine in the product, you need to use a different
query, but with a similar structure to that shown in Figure 20 on page 102:

Chapter 5. Report reference 101

Reports

SELECT
MVS_SYSTEM_ID
,ORIGIN_IMS
,DATE
,VALUE(
(SUM(TRAN_CNTR_1))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_2))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_3))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_4))
/(SUM(TRANSACTIONS)), 0)*100
FROM &PREFIX.IMS_SYSTEM_TRAN_D
WHERE
MVS_SYSTEM_ID = &MVS_SYSTEM_ID
AND ORIGIN_IMS = &IMS_SYSTEM_ID
AND DATE >= &NEWSQ_FROM_DATE
GROUP BY
MVS_SYSTEM_ID
, ORIGIN_IMS
, DATE ;

Figure 20. Example of Query from IMS_SYSTEM_TRAN_D CSQ table

This query uses the following fields from the new IMS_SYSTEM_TRAN_D CSQ table:
• MVS_SYSTEM_ID
• ORIGIN_IMS
• TRAN_CNTR_x 1=< x =< 4
• TRANSACTIONS
If you now want to generate a unique report including data from both the non-CSQ table and the new CSQ
table you can build a new query using the UNION SQL keyword.
The query might look like this:

102 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

SELECT
MVS_SYSTEM_ID
, IMS_SYSTEM_ID
, DATE
, VALUE(
((SUM(EMH_TRAN_CNTR_1) + SUM(MSGQ_TRAN_CNTR_1))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_2) + SUM(MSGQ_TRAN_CNTR_2))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_3) + SUM(MSGQ_TRAN_CNTR_3))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
, VALUE(
((SUM(EMH_TRAN_CNTR_4) + SUM(MSGQ_TRAN_CNTR_4))
/(SUM(EMH_TRANSACTIONS) + SUM(MSGQ_TRANSACTIONS))), 0)*100
FROM &PREFIX.IMS_SYSTEM_D - - PRE SQ TABLE
WHERE
MVS_SYSTEM_ID = &MVS_SYSTEM_ID
AND IMS_SYSTEM_ID = &IMS_SYSTEM_ID
AND DATE >= &PRESQ_FROM_DATE
AND DATE <= &PRESQ_TO_DATE
GROUP BY
MVS_SYSTEM_ID
, IMS_SYSTEM_ID
, DATE
UNION
SELECT
MVS_SYSTEM_ID
,ORIGIN_IMS
,DATE
,VALUE(
(SUM(TRAN_CNTR_1))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_2))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_3))
/(SUM(TRANSACTIONS)), 0)*100
,VALUE(
(SUM(TRAN_CNTR_4))
/(SUM(TRANSACTIONS)), 0)*100
FROM &PREFIX.IMS_SYSTEM_TRAN_D - - PRE SQ TABLE
WHERE
MVS_SYSTEM_ID = &MVS_SYSTEM_ID
AND ORIGIN_IMS = &IMS_SYSTEM_ID
AND DATE >= &NEWSQ_FROM_DATE
AND DATE <= &NEWSQ_TO_DATE
GROUP BY
MVS_SYSTEM_ID
, ORIGIN_IMS
, DATE ;

Figure 21. Example of mixed query

In Figure 21 on page 103 you have PRESQ_FROM_DATE and PRESQ_TO_DATE variables used to set the
time range of the data needed in the report, from the pre-CSQ environment tables, and
NEWSQ_FROM_DATE and NEWSQ_TO_DATE variables used to set the time range of the data needed in
the report from the new CSQ environment tables.
For example, assuming that you want to run this report for the whole month of January 2019, but until
January 14th you used the non-CSQ feature, and from January 15th you started using the CSQ feature,
when you are prompted for the input variables in the data selection panel (DRLDRSEL), you can select the
correct date range, as follows:

Variable Value
MVS_SYSTEM_ID ZOS1
IMS_SYSTEM_ID IMS1
PRESQ_FROM_DATE 2019-01-01
PRESQ_TO_DATE 2019-01-14
NEWSQ_FROM_DATE 2019-01-15
NEWSQ_TO_DATE 2019-01-31

Figure 22. Example of query input variables panel DRLDRSEL

Chapter 5. Report reference 103

Reports

In the report output, you will get the complete data for the whole month, for both the two time periods in
which you run the old and the new engine.

IMS Detail reports

IMS Detail reports give details of items for a specified day.

IMS Message Queue Pool Detail by Date report

This report shows the utilization of the message queue pool buffer for the day and time period selected.
Values are shown cumulatively at each IMS checkpoint.
Figure 23 on page 104 shows an example of a report.
This information identifies the report:
Report ID
IMSS01
Report group
IMS reports
Source
IMS_CHKPT_STATS_T
Attributes
IMS, utilization, performance, qpool, daily
Variables
MVS_SYSTEM_ID, IMS_SYSTEM_ID, DATE, FROM_TIME, TO_TIME
IMS Message Queue Pool Detail by Date
Date: 2019-04-13
System: 'MVS1' IMS System: 'IMS1'

High High Enq

IMS High short long Queue deq
check QBLK message message Message Message manager Buffer buffer Ilog Purge
point Time DRRN DRRN DRRN enqueues dequeues calls waits waits waits waits
----- -------- -------- -------- -------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
1 02.16.11 00000295 000001C1 0000000A 420 380 5281 0 0 0 0
2 02.18.22 00000271 000000FF 0000000F 3369 2964 35616 0 0 0 1
3 02.20.17 0000027A 0000010F 0000000E 6207 5296 63143 0 0 0 2
4 02.45.06 0000026D 000000E5 0000000F 8105 7004 82225 0 0 0 2

IBM Z Performance and Capacity Analytics Report: IMSS01

Figure 23. Example of IMS Message Queue Pool Detail by Date tabular report

The report contains this information:

IMS check point
The numeric ID of the checkpoint for the IMS session.
Time
The time when the activity started, in the format HH.MM.SS.
High QBLK DRRN
The DRRN of the highest queue block.
High short message DRRN
The DRRN of the highest short message queue.
High long message DRRN
The DRRN of the highest long message queue.
Message enqueues
The number of calls to enqueue messages.
Message dequeues
The number of calls to dequeue messages.
Queue manager calls
The total number of calls to QMGR.

104 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Buffer waits
The number of waits for a free buffer.
Enq deq buffer waits
The number of waits for conflicting enque-dequeue buffer requests.
Ilog waits
The number of waits for ILOG.
Purge waits
The number of waits for purge completion.

IMS OSAM/ISAM Buffer Pool Detail by Date report

This report shows the utilization of IMS OSAM/ISAM buffers and pools for the day and time period
selected. Values are shown cumulatively at each IMS checkpoint.
Figure 24 on page 105 shows an example of a report.
This information identifies the report:
Report ID
IMSS02
Report group
IMS reports
Source
IMS_CHKPT_IOSAM_T
Attributes
IMS, buffers, utilization, performance, daily, OSAM, ISAM
Variables
MVS_SYSTEM_ID, IMS_SYSTEM_ID, DATE, FROM_TIME, TO_TIME

IMS OSAM/ISAM Buffer Pool Detail by Date

Date: 2019-04-12
System: 'MVS1' IMS System: 'IMS1'

IMS Found Writes Permanent

Buffer check Pool in Read buffer Blocks write
size point Time requests pool I/O's steal written errors
-------- ----- -------- ----------- ----------- ----------- ----------- ----------- -----------
ALL 1 02.16.11 0 0 0 0 0
ALL 2 02.18.22 8953 8550 142 0 108 0
ALL 3 02.20.17 19403 18720 316 0 243 0
ALL 4 02.45.06 24416 23572 426 0 318 0

IBM Z Performance and Capacity Analytics Report: IMSS02

Figure 24. Example of IMS OSAM/ISAM Buffer Pool Detail by Date tabular report

The report contains this information:

Buffer size
The size of the buffers in the buffer pool, or ALL.
IMS checkpoint
The numeric ID of the checkpoint for the IMS system.
Time
The time when the activity started, in the form HH.MM.SS.
Pool requests
The number of requests.
Found in pool
The number of requests satisfied from pool (I/O not required).
Read I/O's
The number of read I/O operations performed.
Writes buffer steal
The number of QSAM writes issued (single block writes because of buffer steal).

Chapter 5. Report reference 105

Reports

Blocks written
The number of blocks written by purge requests.
Permanent write errors
The number of permanent write error buffers currently locked in the pool.

IMS VSAM Buffer Pool Detail by Date report

This report shows the utilization of IMS VSAM buffers and pools for the day and time period selected.
Values are shown cumulatively at each IMS checkpoint.
Figure 25 on page 106 shows an example of a report.
This information identifies the report:
Report ID
IMSS03
Report group
IMS reports
Source
IMS_CHKPT_VSAM_T
Attributes
IMS, buffers, utilization, performance, daily, VSAM
Variables
MVS_SYSTEM_ID, IMS_SYSTEM_ID, DATE, FROM_TIME, TO_TIME

IMS VSAM Buffer Pool Detail by Date

Date: 2019-04-13
System: 'MVS1' IMS System: 'IMS1'

IMS Number VSAM Retrieves Retrieves VSAM VSAM VSAM VSAM

Buffer check of GET VSAM by by user non-user KSDS ESDS
size point Time buffers calls reads RBA key writes writes inserts inserts
------ ----- -------- ------- --------- --------- --------- --------- -------- -------- -------- --------
139264 1 02.16.11 320 30 295 30 0 0 0 0 0
139264 2 02.18.22 320 2558 2439 1880 1319 292 0 61 1
139264 3 02.20.17 320 6196 5069 4997 2870 777 0 160 3
139264 4 02.45.06 320 8145 6399 6760 3658 1045 0 214 5
2048 1 02.16.11 40 0 207 0 0 0 0 0 0
2048 2 02.18.22 40 0 870 0 0 1 0 0 0
2048 3 02.20.17 40 0 1668 0 0 1 0 0 0
2048 4 02.45.06 40 0 2041 0 0 1 0 0 0
4096 1 02.16.11 240 22 72 22 0 0 0 0 0
4096 2 02.18.22 240 2448 1498 1817 1272 291 0 61 1
4096 3 02.20.17 240 5966 3267 4871 2766 776 0 160 3
4096 4 02.45.06 240 7868 4197 6607 3534 1044 0 214 5
8192 1 02.16.11 40 8 16 8 0 0 0 0 0
8192 2 02.18.22 40 110 71 63 47 0 0 0 0
8192 3 02.20.17 40 230 134 126 104 0 0 0 0
8192 4 02.45.06 40 277 161 153 124 0 0 0 0

IBM Z Performance and Capacity Analytics Report: IMSS03

Figure 25. Example of IMS VSAM Buffer Pool Detail by Date tabular report

The report contains this information:

Buffer size
The size of the buffers in the buffer pool.
IMS check point
The numeric ID of the checkpoint for the IMS session.
Time
The time of the checkpoint, in the form HH.MM.SS.
Number of buffers
The number of buffers in the buffer pool for the IMS session.
VSAM GET calls
The number of VSAM GET calls issued for the IMS session.
VSAM reads
The number of VSAM read I/O operations for the IMS session.

106 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Retrieves by RBA
The number of requests to retrieve by RBA for the IMS session.
Retrieves by key
The number of requests to retrieve by key for the IMS session.
VSAM user writes
The number of VSAM user write requests for the IMS session.
VSAM non-user writes
The number of VSAM space write requests for the IMS session.
VSAM KSDS inserts
The number of logical records inserted to KSDS for the IMS session.
VSAM ESDS inserts
The number of logical records inserted to ESDS for the IMS session.

IMS Availability reports

IMS Availability reports show you the availability of IMS subsystem and region over a specific time period.

IMS CSQ Subsystem Availability, Daily Trend report

This report shows availability for one IMS subsystem in a daily trend between the FROM_DATE and
TO_DATE specified.
This information identifies the report:
Report ID
CSQA01
Report group
CSQ reports
Source
IMS_AVAILABILITY_D
Attributes
IMS, Availability, Daily, Trend
Variables
IMS_System_ID, From_Date, To_Date

IMS CSQ Subsystem Availability, Daily Trend

Date: '2019-09-27 to '2019-09-30'
IMS System: 'IMS 71A

Up In Up In
Up Schedule Schedule Objective
Date Hours Hours % %
---------- ---------- ---------- ---------- ----------
2019-09-27 12 9 100.00 95.00
2019-09-28 24 9 100.00 95.00
2019-09-29 24 9 100.00 95.00
2019-09-30 3 3 33.33 95.00

IBM Z Performance and Capacity Analytics Report: CSQA01

Figure 26. Example of an IMS CSQ subsystem Availability, Daily Trend Report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Up Hours
The total time, in hours, when the IMS subsystem was up and running.

Chapter 5. Report reference 107

Reports

Up In Schedule (Hours)
The time within the schedule, in hours, when the IMS subsystem was up and running. The
IMS_AVAIL_RESOURCE is used to specify the schedule name.
Up In Schedule (%)
The time within the schedule, in percent of scheduled hours, when the IMS subsystem was up and
running. The IMS_AVAIL_RESOURCE is used to specify the schedule name.
Objective (%)
Availability objective for this resource in the scheduled hours.

IMS CSQ Region Availability, Daily Overview report

This report gives you a daily overview of the availability of all the IMS regions in an IMS subsystem.
This information identifies the report:
Report ID
CSQA02
Report group
CSQ reports
Source
IMS_AVAILABILITY_D
Attributes
IMS, Availability, Daily, Overview
Variables
IMS_System_ID, Date

IMS CSQ Region Availability, Daily Overview

Date: 2019-09-27
IMS System: 'IMS71A'

Up In Up In
Region Up Schedule Schedule Objective
Name Hours Hours % %
--------- --------- ---------- -------- ---------
REGION1 12 9 100.00 95.00
REGION2 12 9 100.00 95.00

IBM Z Performance and Capacity Analytics Report: CSQA02

Figure 27. Example of an IMS CSQ Region Availability, Daily Overview report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Region Name
The name of the IMS region.
Up Hours
The total time, in hours, when the IMS region was up and running.
Up In Schedule (Hours)
The time within the schedule, in hours, when the IMS region was up and running. The
IMS_AVAIL_RESOURCE is used to specify the schedule name.
Up In Schedule (%)
The time within the schedule, in percent of scheduled hours, when the IMS region was up and
running. The IMS_AVAIL_RESOURCE is used to specify the schedule name.
Objective (%)
Availability objective for this resource in the scheduled hours.

108 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

IMS CSQ Application Usage and Availability report

This report gives you a daily overview of the availability and CPU usage of all the IMS applications in an
IMS subsystem. The availability is based on the IMS subsystem availability.
This information identifies the report:
Report ID
CSQA03
Report group
CSQ reports
Source
IMS_AVAILABILITY_D, IMS_PSB_ACCOUNT_D
Attributes
IMS, Availability, Daily, Overview
Variables
IMS_System_ID, Date

IMS CSQ Application Usage and Availability

Date: 2019-09-24
IMS System: 'IMS71A'

Up In Up In
Application Up CPU Usage Schedule Schedule Objective
Name Hours Hours Hours % %
-------------------- ---------- ---------- ---------- ---------- ----------
Applic_1 12 1 9 100.00 95.00
Applic_2 12 3 9 100.00 95.00

IBM Z Performance and Capacity Analytics Report: CSQA03

Figure 28. Example of an IMS CSQ application Usage and Availability report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Application Name
The name of the IMS application.

IMS CSQ Resource Utilization, Daily Overview report

This report gives you a daily overview of resource utilization of all the IMS applications in an IMS
subsystem.
This information identifies the report:
Report ID
CSQA04
Report group
CSQ reports
Source
IMS_PSB_ACCOUNT_D
Attributes
IMS, Resource Utilization, Daily, Overview
Variables
IMS_System_ID, Date

Chapter 5. Report reference 109

Reports

IMS CSQ Resource Utilization, Daily Overview

IMS SUBSYSTEM NAME:'IMS1 '
DATE: 2019-02-04

Program DC DC DC
Transaction Total CPU calls calls calls
name Transactions time GN GU ISRT
------------ ------------ ------------ ---------- ---------- ----------
$BMP 0.00000 8.33333E-04 0.000 0.000 1.109E+04
DE1A 4.80000E+01 3.04245E-01 4.800E+01 6.200E+01 1.440E+02
DE1B 5.60000E+01 3.22109E-01 5.600E+01 6.900E+01 1.680E+02
DE1C 6.10000E+01 3.40781E-01 6.100E+01 7.400E+01 1.830E+02
DE1D 3.60000E+01 2.64036E-01 3.600E+01 5.000E+01 1.080E+02

DB DB DB DB DB DB
calls calls calls calls calls calls
DLET GHN GHNP GHU GN GNP
---------- ---------- ---------- ---------- ---------- ----------
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 4.800E+01 0.000 0.000
0.000 0.000 0.000 5.600E+01 0.000 0.000
0.000 0.000 0.000 6.100E+01 0.000 0.000
0.000 0.000 0.000 3.600E+01 0.000 0.000

DB DB DB
calls calls calls
GU ISRT REPL
-------- ---------- ----------
0.000 0.000 0.000
0.000 4.800E+01 4.800E+01
0.000 5.600E+01 5.600E+01
0.000 6.100E+01 6.100E+01
0.000 3.600E+01 3.600E+01

IBM Z Performance and Capacity Analytics Report: CSQA04

Figure 29. Example of an IMS CSQ Resource Utilization, Daily Overview report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Transaction name
The name of the transaction.
Total Transactions
The total number of transactions.
Program CPU time
The total dependent region CPU seconds.
DC calls GN
The total number of DL/I message queue GN calls.
DC calls GU
The total number of DL/I message queue GU calls.
DC calls ISRT
The total number of DL/I message queue ISRT calls.
DB calls DLET
The total number of DL/I database DLET calls.
DB calls GHN
The total number of DL/I database GHN calls.
DB calls GHNP
The total number of DL/I database GHNP calls.

110 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

DB calls GHU
The total number of DL/I database GHU calls.
DB calls GN
The total number of DL/I database GN calls.
DB calls GNP
The total number of DL/I database GNP calls.
DB calls GU
The total number of DL/I database GU calls.
DB calls ISRT
The total number of DL/I database ISRT calls.
DB calls REPL
The total number of DL/I database REPL calls.

IMS CSQ Application Usage and Availability report

This report gives you a daily overview of the availability and CPU usage of all the IMS applications in an
IMS subsystem. The availability is based on the IMS subsystem availability.
Note: This report is only available if you have installed the Extended Accounting subcomponent.
This information identifies the report:
Report ID
CSQB01
Report group
CSQ reports
Source
IMS_AVAILABILITY_D, IMS_PSB_ACCOUNT2_D
Attributes
IMS, Availability, Daily, Overview
Variables
IMS_System_ID, Date

IMS CSQ Application Usage and Availability

Date: 2019-12-12
IMS System: 'IMSA '

Up In Up In
Application Up CPU Usage Schedule Schedule Objective
Name Hours Hours Hours % %
------------------ ---------- ---------- ---------- ---------- ----------
$UNKNOWN 1.00 0.18 0.10 1.13 95.00

IBM Z Performance and Capacity Analytics Report: CSQB01

Figure 30. Example of an IMS CSQ application Usage and Availability report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Application name
The name of the IMS application.
Up Hours
The total time, in hours, when the IMS region was up and running.
CPU Usage (Hours)
The processor time for the application, in hours.

Chapter 5. Report reference 111

Reports

Up In Schedule (Hours)
The time within the schedule, in hours, when the IMS subsystem for this application was up and
running. The IMS_AVAIL_RESOURCE is used to specify the schedule name.
Up In Schedule (%)
The time within the schedule, in percent of scheduled hours, when the IMS subsystem for this
application was up and running. The IMS_AVAIL_RESOURCE is used to specify the schedule name.
Objective (%)
Availability objective for the IMS subsystem related to this application in the scheduled hours.

IMS CSQ Resource Utilization, Daily Overview report

This report gives you a daily overview of resource utilization of all the IMS applications in an IMS
subsystem.
Note: This report is available if you have installed the Extended CSQ component only.
This information identifies the report:
Report ID
CSQB02
Report group
CSQ reports
Source
IMS_PSB_ACCOUNT2_D
Attributes
IMS, Resource Utilization, Daily, Overview
Variables
IMS_System_ID, Date

IMS CSQ Resource Utilization, Daily Overview

IMS SUBSYSTEM NAME:'CSSP '
DATE: 2019-03-05

Program DC DC DC
Transaction Total CPU calls calls calls
name Transactions time GN GU ISRT
------------ ------------ ------------ ---------- ---------- ----------
$BMP 0.00000 2.60417E-05 0.000 0.000 0.000
HC9APIA0 4.00000 4.53125E-03 0.000 6.000 4.000
H9FDIS 2.00000 2.96875E-03 0.000 4.000 3.600E+01
IVTNO 2.00000 2.63021E-03 0.000 3.000 2.000
IVTNV 3.00000 3.67187E-03 0.000 4.000 3.000

DB DB DB DB DB DB
calls calls calls calls calls calls
DLET GHN GHNP GHU GN GNP
---------- ---------- ---------- ---------- ---------- ----------
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 1.000 0.000 0.000
0.000 0.000 0.000 1.000 0.000 0.000

DB DB DB
calls calls calls
GU ISRT REPL
----- ---------- ----------
0.000 0.000 0.000
0.000 0.000 0.000
0.000 0.000 0.000
1.000 0.000 1.000
2.000 0.000 1.000

IBM Z Performance and Capacity Analytics Report: CSQB02

Figure 31. Example of an IMS CSQ Resource Utilization, Daily Overview report

112 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem.
Transaction Name
The name of the transaction.
Total Transactions
The total number of transactions.
Program CPU time
The total dependent region CPU seconds.
DC calls GN
The total number of DL/I message queue GN calls.
DC calls GU
The total number of DL/I message queue GU calls.
DC calls ISRT
The total number of DL/I message queue ISRT calls.
DB calls DLET
The total number of DL/I database DLET calls.
DB calls GHN
The total number of DL/I database GHN calls.
DB calls GHNP
The total number of DL/I database GHNP calls.
DB calls GHU
The name of the transaction.
DB calls GN
The total number of DL/I database GN calls.
DB calls GNP
The name of the transaction.
DB calls GU
The total number of DL/I database GU calls.
DB calls ISRT
The name of the transaction.
DB calls REPL
The total number of DL/I database REPL calls.

IMS Message Queue reports

These reports show statistics of the IMS Message Queue utilization.

IMS Message Queue Utilization, Date report

This report shows the Input and Output message queue utilization for the selected IMS system and
Queue type, at 15-minute intervals during a specified time period. Figure 32 on page 114 shows an
example of a report.
This information identifies the report:
Report ID
CSQTQ01
Report group
CSQ reports

Chapter 5. Report reference 113

Reports

Source
IMS_TRAN_QUEUE_QV
Attributes
IMS, Transaction, Queue, Date
Variables
Origin_IMS, Date, Queue_Type

IMS Message Queue Utilization, Date report

Date: 2019-12-05
IMS System: 'CSSD'
Queue Type: 'MSGQ LOCAL'

<------------Input Message----------------------> <---------------Output Message------------------>

Time FpMsg ShMsg LgMsg Processed AvgTime FpMsg ShMsg LgMsg Processed AvgTime
------ ------ ------ ------ --------- -------- ------ ------ ------ --------- -------
15.15 - 0.16 1.38 251 0.21 - 0.00 0.00 217 0.00
15.30 - 0.15 9.06 481 0.45 - 0.00 0.00 393 0.00
15.45 - 1.39 20.84 463 0.43 - 41.90 0.00 421 0.00
16.00 - 0.49 2.41 459 0.18 - 6.21 0.00 438 0.00
16.15 - 0.12 5.08 368 0.23 - 0.00 0.00 362 0.00
16.30 - 0.10 14.42 278 0.29 - 0.00 0.00 266 0.00

IBM Z Performance and Capacity Analytics Report: CSQTQ01

Figure 32. Example of IMS Message Queue Utilization, Date report

The report contains this information:

Date
The date of the day for the measurement.
Time
The 15-minute interval for which the data applies.
IMS System
The name of the IMS subsystem.
Queue Type
The queue type.
Input Message
Statistics on inbound message queues activities.
FpMsg
Average Fast Path messages on queue before processing.
ShMsg
Average Full Function Short messages on queue before processing.
LgMsg
Average Full Function Long messages on queue before processing.
Processed
Total number of messages processed.
AvgTime
Average time (in seconds) each message remains on queue.
Output Message
Statistics on outbound message queues activities.
FpMsg
Average Fast Path messages on queue before processing.
ShMsg
Average Full Function Short messages on queue before processing.
LgMsg
Average Full Function Long messages on queue before processing.
Processed
Total number of messages processed.

114 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

AvgTime
Average time (in seconds) each message remains on queue.

IMS Msg Queue Utilization by Transaction, Date report

This report shows the Input and Output message queue utilization by transaction name for the selected
IMS system and Queue type, at 15-minute intervals during a specified time period. Figure 33 on page 115
shows an example of a report.
This information identifies the report:
Report ID
CSQTQ02
Report group
CSQ reports
Source
IMS_TRAN_QUEUE_QV
Attributes
IMS, Transaction, Queue, Date
Variables
Origin_IMS, Date, Queue_Type

IMS Message Queue Utilization by Transaction, Date report

Date: 2019-12-05
IMS System: 'CSSD'
Queue Type: 'MSGQ LOCAL'

<----------------Input Message----------------> <------------------Output Message----------------->

Transaction
Time Name FpMsg ShMsg LgMsg Processed AvgTime FpMsg ShMsg LgMsg Processed AvgTime
------- ----------- ------ ------- ------ ---------- -------- ------ ------- -------- ---------- -------
15.15 HC9CHKRT - 1.00 0.00 1 0.00 - 0.00 0.00 0 0.00
HC9PMUT0 - 1.00 0.00 2 0.40 - 0.00 0.00 0 0.00
HC9SWFT0 - 0.00 4.00 2 0.10 - 0.00 0.00 2 0.00
HC9T100T - 0.00 1.00 2 0.00 - 0.00 0.00 2 0.00
HC9T200T - 0.00 3.50 2 0.00 - 0.00 0.00 0 0.00

15.30 HC9CHKRT - 0.00 8.00 2 0.00 - 0.00 0.00 0 0.00

HC9CRIT0 - 0.00 13.00 3 0.30 - 0.00 0.00 0 0.00
HC9T100T - 0.00 10.83 5 0.00 - 0.00 0.00 5 0.00
HC9T11T0 - 0.00 13.25 3 0.00 - 0.00 0.00 0 0.00
HEQAAUU2 - 0.00 5.00 2 0.00 - 0.00 0.00 2 0.00
HEQBW104 - 0.00 9.00 1 0.00 - 0.00 0.00 1 0.00

IBM Z Performance and Capacity Analytics Report: CSQTQ02

Figure 33. Example of IMS Message Queue Utilization by Transaction, Date report

The report contains this information:

Date
The date of the day for the measurement.
Time
The 15-minute interval for which the data applies.
IMS System
The name of the IMS subsystem.
Queue Type
The queue type.
Transaction Name
The name of the IMS transaction.
For the description of the other columns in this report, see “IMS Message Queue Utilization, Date report”
on page 113.

IMS Message Queue Utilization Overview, Daily Report

This report shows an overview of the different queue types utilization for the selected IMS system
between the FROM_DATE and TO_DATE specified. Figure 34 on page 116 shows an example of a report.

Chapter 5. Report reference 115

Reports

This information identifies the report:

Report ID
CSQTQ03
Report group
CSQ reports
Source
IMS_TRAN_QUEUE_DV
Attributes
IMS, Transaction, Queue, Date
Variables
Origin_IMS, From_Date, To_Date

IMS Msg Queue Utilization Overview, Daily Report

Date: '2019-07-17' to '2019-07-17'
IMS System: 'CSSD '

<--------------Input Message------------> <---------------Output Message----------->

Date Queue Type FpMsg ShMsg LgMsg Processed AvgTime FpMsg ShMsg LgMsg Processed AvgTime
---------- ----------- -------- -------- -------- --------- -------- -------- -------- -------- --------- --------
2019-07-17 EMHQ LOCAL 1.00 - - 1 0.00 1.00 - - 1 0.63
MSGQ LOCAL - 0.24 10.56 3841 0.36 - 5.20 0.00 3609 0.00

IBM Z Performance and Capacity Analytics Report: CSQTQ03

Figure 34. Example of an IMS Message Queue Utilization Overview, Daily Report

The report contains this information:

Date
The date of the day for the measurement.
IMS System
The name of the IMS subsystem.
Queue Type
The queue type.
For the description of the other columns in this report, see “IMS Message Queue Utilization, Date report”
on page 113.

IMS Transaction Arrival Rate and Msg Queue Usage, Daily Trend report
This report shows a daily trend on how the selected IMS system performs with the message queue
resources of the selected queue type, compared with the transaction arrival rate, between the
FROM_DATE and TO_DATE specified.
Figure 35 on page 117 shows an example of a report.
This information identifies the report:
Report ID
CSQTQ04
Report group
CSQ reports
Source
IMS_TRAN_QUEUE_DV
Attributes
IMS, Transaction, Queue, Date
Variables
Origin_IMS, From_Date, To_Date, Queue Type

116 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

IMS Transaction Arrival Rate and Message Queue Utilization Overview, Daily Trend
Date: 2019-11-06 to 2019-11-09
IMS System: 'IMS2'
Queue Type: 'MSGQ LOCAL'

<--Transaction--> <--------------Input Message------------> <--------------Output Message---------------->

Date Total Rate FpMsg ShMsg LgMsg Processed AvgTime FpMsg ShMsg LgMsg Processed AvgTime
---------- ----------- ----- ------ ------ --------- -------- ------- ------- ------ ---------- --------
2019-07-17 3841 0.0445 - 0.24 10.56 3841 0.36 - 5.20 0.00 3609 0.00
2019-07-18 7000 0.0810 - 1.00 20.30 7000 0.70 - 6.30 4.00 6500 0.40
2019-07-19 4511 0.0522 - 0.50 11.20 4511 0.46 - 3.20 2.00 4211 0.20

IBM Z Performance and Capacity Analytics Report: CSQTQ04

Figure 35. Example of an IMS Transaction Arrival Rate and Message Queue Usage, Daily Trend report

The report contains this information:

Date
The date of the day for the measurement.
IMS System
The name of the IMS subsystem.
Queue Type
The queue type.
Transaction Total
The total number of transactions processed.
Transaction Rate
The total arrival date (transactions per second).
For the description of the other columns, see “IMS Message Queue Utilization, Date report” on page 113.

IMS CSQ Transaction Transit Time reports

These reports show statistics of the IMS Transaction Transit Time performance.

IMS CSQ Transit Time Analysis By Transaction Name report

This daily report shows the transaction transit time metrics for the selected IMS systems, date, and
transaction name.
Figure 36 on page 118 shows an example of a report.
This information identifies the report:
Report ID
CSQTQ05
Report group
CSQ reports
Source
IMS_TRAN_D
Attributes
IMS, Transaction, Utilization, Performance, Daily
Variables
Origin_IMS, Process_IMS, Date, Transaction_Name

Chapter 5. Report reference 117

Reports

IMS CSQ Transit Time Analysis By Transaction Name

ORIGIN IMS NAME:'IMSE'
PROCESS IMS NAME:'IMSE'
DATE: 2019-03-11

CPU Average Input Input

Transaction Total Utilization Response Local CQS
name Transactions Approximate time Queue queue
+------------++------------++------------++----------++----------++----------+
A70200 6.83000E+02 3.903E+00 1.878E-01 6.830E+02 0.000E+00
A70210 2.00000E+00 0.000E+00 1.500E-01 2.000E+00 0.000E+00
A7041030 4.20000E+01 4.748E-03 3.333E-02 4.200E+01 0.000E+00
A7042030 4.60000E+01 4.845E-03 3.804E-01 4.600E+01 0.000E+00
A7051030 1.36000E+02 1.299E-02 1.390E-01 1.360E+02 0.000E+00

Average Average
Average Average Output Output Output Output
Input Process Local CQS Local CQS
time time Queue queue time time
----------++----------++----------++----------++----------++----------+
1.025E-03 1.114E-01 6.630E+02 0.000E+00 3.017E-04 0.000E+00
0.000E+00 0.000E+00 2.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 3.333E-02 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 2.913E-01 4.600E+01 0.000E+00 0.000E+00 0.000E+00
0.000E+00 6.912E-02 1.340E+02 0.000E+00 0.000E+00 0.000E+00

Average
Total Network
Responses time
---------++----------
6.630E+02 7.738E-02
2.000E+00 1.500E-01
0.000E+00 0.000E+00
4.600E+01 8.913E-02
1.340E+02 7.090E-02

IBM Z Performance and Capacity Analytics Report: CSQTQ05

Figure 36. Example of IMS CSQ Transit Time Analysis By Transaction Name, Daily report

The report contains this information:

ORIGIN IMS
The IMS subsystem ID defined in the origin part of the UOW token. It identifies the activity origin. In a
non-CSQ configuration it always matches the PROCESS_IMS value.
PROCESS IMS
The IMS subsystem ID defined in the processing part of the UOW token. It identifies the activity
processor. In a non-Shared Queue configuration it always matches the ORIGIN_IMS value.
Date
The date of the day for the measurement.
Transaction Name
The name of the IMS transaction the user requested.
Total Transactions
Total number of transactions.
CPU Utilization Approximate
This column represents the sum of approximate number of CPU seconds of program execution time
while the transactions were active. This value is not provided for WFI or PWFI transactions (to get the
correct value look at the PSB_ACCOUNT_x table or CSQA04 report_ID).
Average Response time
The average time in seconds, needed to process a transaction from the beginning to the end. It should
be considered as the sum of the host time plus network time.
Input Local Queue
The total number of input messages issued by transactions, BMP programs, not using CSQ.
Input CQS queue
The total number of input messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non-CSQ configuration.
Average Input time
The average time, in seconds, that transactions and BMP programs spent on the IMS input message
queue, including input queue time for program-to-program switch transactions. In a CSQ
configuration it also includes the time the transaction spent in SQ before being processed.

118 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Average Process time

The average elapsed time in seconds that transactions and BMP programs spent processing in the
dependent regions, in seconds.
Output Local Queue
The total number of output messages issued by transactions and BMP programs, not using CSQ.
Output CQS Queue
The total number of output messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non-CSQ configuration.
Average Output Local time
The average time that responding transactions spent on the IMS output queue waiting for
transmission to the ultimate network destination, in seconds.
Average Output CQS time
The time between when the completed output transaction was put on the queue and when it was
routed as output to the terminal. Always zero in a non-CSQ configuration.
Total Responses
The total number of SNA definite responses or exception responses for which the message is
dequeued.
Average Network time
The average time that responding transactions spent in network transmission to the ultimate
destination, in seconds, as measured using SNA definite response. This may also include user think
time to the next transaction, if the transaction defined as such in IMS.

IMS CSQ Transit Time Analysis By LTERM and Userid report

This daily report shows the transaction transit time metrics for the selected IMS systems, date, origin
LTERM, and userid.
Note: The summary average transit time values (in bold) will be provided only if you are using IBM Z
Performance and Capacity Analytics with QMF (QMFUSE=YES coded in the IBM Z Performance and
Capacity Analytics initialization member DRLFPROF).
Figure 37 on page 120 shows an example of a report.
This information identifies the report:
Report ID
CSQTQ06
Report group
CSQ reports
Source
IMS_TRAN_D
Attributes
IMS, Lterm, Utilization, Performance, Daily, User_ID
Variables
Origin_IMS, Process_IMS, Date, origin_Lterm, User_ID

Chapter 5. Report reference 119

Reports

IMS CSQ Transit Time Analysis By Lterm and Userid

ORIGIN IMS NAME:'IMSE'
PROCESS IMS NAME:'IMSE'
DATE: 2019-03-11

Origin CPU Average

Logical Transaction Total Utilization Response
Userid Terminal name Transactions Approximate time
+--------++--------++------------++------------++------------++----------+
IN1004 BIN1004 A71400 2.00000E+00 1.412E-04 5.000E-02
A71410 6.00000E+00 3.272E-04 1.000E-01
A71500 1.00000E+00 0.000E+00 0.000E+0
F7150010 2.00000E+00 1.532E-04 4.000E-01
I7450030 1.20000E+01 7.845E-03 1.750E-01
I7451030 6.00000E+00 4.190E-03 2.167E-01
I8010030 6.00000E+00 4.322E-03 1.333E-01
I8020030 1.20000E+01 7.457E-03 1.667E-01
I8600030 1.00000E+01 1.263E-03 1.200E-01
K70500 1.80000E+01 2.739E-03 1.111E-01
------------ ------------ ----------
7.90000E+01 2.743E-02 1.430E-01

Input Input Average Average Output Output

Local CQS Input Process Local CQS
Queue Queue time time Queue Queue
+----------++----------++----------++----------++----------++----------
2.000E+00 0.000E+00 0.000E+00 5.000E-02 0.000E+00 0.000E+00
6.000E+00 0.000E+00 0.000E+00 8.333E-02 4.000E+00 0.000E+00
1.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
2.000E+00 0.000E+00 0.000E+00 3.000E-01 1.000E+00 0.000E+00
1.200E+01 0.000E+00 0.000E+00 1.417E-01 6.000E+00 0.000E+00
6.000E+00 0.000E+00 0.000E+00 1.833E-01 3.000E+00 0.000E+00
6.000E+00 0.000E+00 0.000E+00 8.333E-02 3.000E+00 0.000E+00
1.200E+01 0.000E+00 0.000E+00 1.250E-01 6.000E+00 0.000E+00
1.000E+01 0.000E+00 0.000E+00 6.000E-02 5.000E+00 0.000E+00
1.800E+01 0.000E+00 0.000E+00 7.778E-02 1.800E+01 0.000E+00
---------- ---------- ---------- ---------- ---------- ----------
7.900E+01 0.000E+00 0.000E+00 1.025E-01 4.800E+01 0.000E+00

Average Average
Output Output Average
Local CQS Total Network
time time Responses time
----------++----------++----------++----------
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 4.000E+00 2.500E-02
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 1.000E+00 2.000E-01
0.000E+00 0.000E+00 6.000E+00 6.667E-02
0.000E+00 0.000E+00 3.000E+00 6.667E-02
0.000E+00 0.000E+00 3.000E+00 1.000E-01
0.000E+00 0.000E+00 6.000E+00 8.333E-02
0.000E+00 0.000E+00 5.000E+00 1.200E-01
5.556E-03 0.000E+00 1.800E+01 2.778E-02
---------- ---------- ---------- ----------
1.266E-03 0.000E+00 4.800E+01 3.924E-02

IBM Z Performance and Capacity Analytics Report: CSQTQ06

Figure 37. Example of IMS CSQ Transit Time Analysis By LTERM and Userid, Daily report

The report contains this information:

ORIGIN IMS
The IMS subsystem ID defined in the origin part of the UOW token. It identifies the activity origin. In a
non-CSQ configuration it always matches the PROCESS_IMS value.
PROCESS IMS
The IMS subsystem ID defined in the processing part of the UOW token. It identifies the activity
processor. In a non-Shared Queue configuration it always matches the ORIGIN_IMS value.
Date
The date of the day for the measurement.
Userid
The user identifier used to gain authorized access to IMS resources. This column contains the logical
terminal name if security is not being managed by the IMS-supported /SIGNON.
Origin Logical Terminal
The IMS-defined logical name for the terminal used to request the transaction or OTMA Tpipe name.
Transaction Name
The name of the IMS transaction the user requested.
Total Transactions
Total number of transactions.

120 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

CPU Utilization Approximate

This column represents the sum of approximate number of CPU seconds of program execution time
while the transactions were active. This value is not provided for WFI or PWFI transactions (to get the
correct value look at the PSB_ACCOUNT_x table or CSQA04 report ID).
Average Response time
The average time in seconds, needed to process a transaction from the beginning to the end. It should
be considered as the sum of the host time plus network time.
Input Local Queue
The total number of input messages issued by transactions, BMP programs, not using CSQ.
Input CQS queue
The total number of input messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non-CSQ configuration.
Average Input time
The average time, in seconds, that transactions and BMP programs spent on the IMS input message
queue, including input queue time for program-to-program switch transactions. In a CSQ
configuration it also includes the time the transaction spent in SQ before being processed.
Average Process time
The average elapsed time in seconds that transactions and BMP programs spent processing in the
dependent regions, in seconds.
Output Local Queue
The total number of output messages issued by transactions and BMP programs, not using CSQ.
Output CQS Queue
The total number of output messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non- CSQ configuration.
Average Output Local time
The average time that responding transactions spent on the IMS output queue waiting for
transmission to the ultimate network destination, in seconds.
Average Output CQS time
The time between when the completed output transaction was put on the queue and when it was
routed the output to the terminal. Always zero in a non CSQ configuration.
Total Responses
The total number of SNA definite responses or exception responses for which the message is
dequeued.
Average Network time
The average time that responding transactions spent in network transmission to the ultimate
destination, in seconds, as measured using SNA definite response. This may also include user think
time to the next transaction, if the transaction is defined as such in IMS.

IMS CSQ Transit Time Analysis by Region report

This daily report shows the transaction transit time metrics for the selected IMS systems, date, and
region job name.
Note: The summary average transit time values (in bold) will be provided only if you are using IBM Z
Performance and Capacity Analytics with QMF (QMFUSE=YES coded in the IBM Z Performance and
Capacity Analytics initialization member DRLFPROF).
Figure 38 on page 122 shows an example of a report.
This information identifies the report:
Report ID
CSQTQ07
Report group
CSQ reports

Chapter 5. Report reference 121

Reports

Source
IMS_TRAN_D
Attributes
IMS, Region, Utilization, Performance, Daily
Variables
Origin_IMS, Process_IMS, Date, Region_Job_Name

IMS CSQ Transit Time Analysis By Region

ORIGIN IMS NAME:'IMSE'
PROCESS IMS NAME:'IMSE'
DATE: 2019-03-11

Region CPU Average Input Input

Job Program Total Utilization Response Local CQS
name name Transactions Approximate time Queue Queue
+--------++--------++------------++------------++----------++----------++----------+
MPPEROB1 DO31 3.10000E+01 0.000E+00 2.484E-01 3.100E+01 0.000E+00
DO35 1.00000E+01 2.845E-03 2.070E+00 1.000E+01 0.000E+00
DO40 3.10000E+01 7.964E-03 1.097E-01 3.100E+01 0.000E+00
DO41 9.00000E+00 2.658E-03 2.444E-01 9.000E+00 0.000E+00
DO675 1.80000E+01 3.441E-03 6.111E-01 1.800E+01 0.000E+00
DO676 3.50000E+01 8.235E-03 6.000E-01 3.500E+01 0.000E+00
DO700 1.60000E+01 3.127E-03 5.062E-01 1.600E+01 0.000E+00
FSRASF1I 1.97000E+02 4.350E-02 1.229E+00 1.970E+02 0.000E+00
------------ ------------ ---------- ---------- -----------
3.47000E+02 7.177E-02 9.115E-01 3.470E+02 0.000E+00

Average
Average Average Output Output Output
Input Process Local CQS Local
time time Queue Queue time
+----------++----------++----------++----------++----------+
2.484E-01 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 2.040E+00 5.000E+00 0.000E+00 0.000E+00
1.032E-01 6.452E-03 0.000E+00 0.000E+00 0.000E+00
1.333E-01 1.111E-01 0.000E+00 0.000E+00 0.000E+00
5.556E-02 5.556E-01 0.000E+00 0.000E+00 0.000E+00
1.429E-02 5.857E-01 0.000E+00 0.000E+00 0.000E+00
4.687E-01 3.750E-02 0.000E+00 0.000E+00 0.000E+00
1.015E-03 7.142E-01 1.580E+02 0.000E+00 0.000E+00
---------- ---------- ---------- ---------- ----------
6.138E-02 5.573E-01 1.630E+02 0.000E+00 0.000E+00

Average
Output Average
CQS Total Network
time Responses time
+----------++----------++----------
0.000E+00 0.000E+00 0.000E+00
0.000E+00 5.000E+00 6.000E-02
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 1.580E+02 6.411E-01
---------- ---------- ----------
0.000E+00 1.630E+02 2.928E-01

IBM Z Performance and Capacity Analytics Report: CSQTQ07

Figure 38. Example of IMS CSQ Transit Time Analysis By Region, Daily report

The report contains this information:

ORIGIN IMS
The IMS subsystem ID defined in the origin part of the UOW token. It identifies the activity origin. In a
non-CSQ configuration it always matches the PROCESS_IMS value.
PROCESS IMS
The IMS subsystem ID defined in the processing part of the UOW token. It identifies the activity
processor. In a non-Shared Queue configuration it always matches the ORIGIN_IMS value.
Date
The date of the day for the measurement.
Region Job Name
The MVS- and JES-identified job name for the IMS dependent region.
Program Name
The name of the IMS application program used to process the transaction.
Total Transactions
Total number of transactions.

122 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

CPU Utilization Approximate

This column represents the sum of approximate number of CPU seconds of program execution time
while the transactions were active. This value is not provided for WFI or PWFI transactions (to get the
correct value look at the PSB_ACCOUNT_x table or CSQA04 report ID).
Average Response time
The average time in seconds, needed to process a transaction from the beginning to the end. It should
be considered as the sum of the host time plus network time.
Input Local Queue
The total number of input messages issued by transactions, BMP programs, not using CSQ.
Input CQS queue
The total number of input messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non-CSQ configuration.
Average Input time
The average time, in seconds, that transactions and BMP programs spent on the IMS input message
queue, including input queue time for program-to-program switch transactions. In a CSQ
configuration it also includes the time transaction spent in SQ before being processed.
Average Process time
The average elapsed time in seconds that transactions and BMP programs spent processing in the
dependent regions, in seconds.
Output Local Queue
The total number of output messages issued by transactions and BMP programs, not using CSQ.
Output CQS Queue
The total number of output messages issued by transactions and BMP programs queued through IMS
CSQ. Always zero in a non-CSQ configuration.
Average Output Local time
The average time that responding transactions spent on the IMS output queue waiting for
transmission to the ultimate network destination, in seconds.
Average Output CQS time
The time between when the completed output transaction was put on the queue and when it was
routed as the output to the terminal. Always zero in a non-CSQ configuration.
Total Responses
The total number of SNA definite responses or exception responses for which the message is
dequeued.
Average Network time
The average time that responding transactions spent in network transmission to the ultimate
destination, in seconds, as measured using SNA definite response. This may also include user think
time to the next transaction, if the transaction is defined as such in IMS.

IMS CSQ Utilization reports

IMS utilization reports show you the metrics of system resource utilization.

IMS CSQ Resource Utilization, Daily Overview report

IMS CSQ utilization reports show you the metrics of system resource utilization for the selected IMS ID
and date by transaction code.
Figure 40 on page 125 shows an example of a report.
This information identifies the report:
Report ID
CSQTQ08
Report group
CSQ reports

Chapter 5. Report reference 123

Reports

Source
IMS_PSB_ACCOUNT_D
Attributes
IMS, Accounting, Utilization, Daily, Overview
Variables
IMS_System_ID, Date

IMS CSQ Resource Utilization, Daily Overview

IMS SUBSYSTEM NAME:'IMSE'
DATE: 2019-03-11

Program DC DC DC
Transaction Total CPU calls calls calls
name Transactions time GN GU ISRT
------------ ------------ ------------ ---------- ---------- ----------
$BMP 0.00000E+00 2.00521E-01 0.000E+00 0.000E+00 1.000E+00
AUSWAHL 4.44000E+02 1.02109E+00 0.000E+00 4.440E+02 4.440E+02
A5760010 7.20000E+01 1.51792E+00 0.000E+00 1.440E+02 0.000E+00
A5770010 3.00000E+00 6.69531E-02 0.000E+00 6.000E+00 0.000E+00
A5780010 1.00000E+00 2.27839E-01 0.000E+00 3.000E+00 0.000E+00
A5790010 1.00000E+00 5.07031E-02 0.000E+00 2.000E+00 0.000E+00
A5920010 1.68000E+02 3.31427E+00 0.000E+00 3.320E+02 0.000E+00
A70200 6.83000E+02 3.90340E+00 0.000E+00 6.830E+02 6.830E+02

DB DB DB DB DB DB
calls calls calls calls calls calls
DLET GHN GHNP GHU GN GNP
---------- ---------- ---------- ---------- ---------- ----------
0.000E+00 2.161E+03 0.000E+00 1.500E+01 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 4.440E+02
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 5.430E+02 5.430E+02 0.000E+00 5.940E+02

DB DB DB
calls calls calls
GU ISRT REPL
---------- ---------- ----------
9.000E+00 1.500E+01 1.800E+01
4.440E+02 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00
6.830E+02 0.000E+00 1.086E+03

IBM Z Performance and Capacity Analytics Report: CSQTQ08

Figure 39. Example of IMS CSQ Resource Utilization, Daily Overview report

The report contains this information:

Date
The date of the day for the measurement.
IMS System ID
The name of the IMS subsystem effectively processing the transaction.
Transaction name
The name of the IMS transaction.
Total Transactions
Total number of transactions processed by a PSB.
Program CPU time
The total dependent region CPU seconds.
DC calls GN
The total number of DL/I message queue GN calls.
DC calls GU
The total number of DL/I message queue GU calls.
DC calls ISRT
The total number of DL/I message queue ISRT calls.
DB calls DLET
The total number of DL/I database DLET calls.

124 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

DB calls GHN
The total number of DL/I database GHN calls.
DB calls GHNP
The total number of DL/I database GHNP calls.
DB calls GHU
The total number of DL/I database GHU calls.
DB calls GN
The total number of DL/I database GN calls.
DB calls GNP
The total number of DL/I database GNP calls.
DB calls GU
The total number of DL/I database GU calls.
DB calls ISRT
The total number of DL/I database ISRT calls.
DB calls REPL
The total number of DL/I database REPL calls.

IMS HALDB OLR reports

IMS HALDB OLR reports show information about the High Availability Large Databases Online
Reorganization (HALDB OLR) process.

IMS HALDB OLR Unit Of Reorganization, Details report

This report shows details about High Availability Large Databases Online Reorganization (HALDB OLR)
activity at Unit of Reorganization (UOR) level.
This information identifies the report:
Report ID
CSQO01
Report group
CSQ reports
Source
IMS_HALDB_OLR_T
Attributes
IMS, HALDB, OLR, UOR
Variables
EVENT_DATE, PERIOD_NAME, SYSPLEX_NAME, MVS_SYSTEM_ID, DATABASE, PARTITION

IMS HALDB OLR Unit of Reorganization, Details

Sysplex='PLEX1' System='MVS1'
Date: '2019-05-11'

Period Moved
Time name Database Partition segments Moved bytes Moved roots Lock count
-------- -------- --------- ---------- ---------- ----------- ----------- ----------
22.04.18 NIGHT JPDSWIH JPIHRA1 8.9200E+02 6.0346E+40 1.3170E+03 1.3900E+03

IBM Z Performance and Capacity Analytics Report: CSQO01

Figure 40. Example of an IMS HALDB OLR Unit of Reorganization, Details report

The report contains this information:

Time
Time of the measurement. From TIME.
Period name
Period name. From PERIOD_NAME.

Chapter 5. Report reference 125

Reports

Database
Database. From DBD_NAME.
Partition
Database partition name. From PARTITION_NAME.
Moved segments
The number of moved segments. From SEG_MOVED.
Moved bytes
The number of moved bytes. From SIZE_MOVED.
Moved roots
The number of moved roots. From ROOTS_MOVED.
Lock count
The count of locks. From LOCK_COUNT.

IMS HALDB OLR Statistics report

This report shows details about High Availability Large Databases Online Reorganization (HALDB OLR)
activity.
This information identifies the report:
Report ID
CSQO02
Report group
CSQ reports
Source
IMS_HALDB_OLR_H
Attributes
IMS, HALDB, OLR
Variables
FROM_DATE, TO_DATE, PERIOD_NAME, SYSPLEX_NAME, MVS_SYSTEM_ID, DATABASE, PARTITION

IMS HALDB OLR Statistics

Sysplex='PLEX1' System='MVS1'
From: '2019-05-11' To: '2019-05-12'

Period Moved
Date Time name Database Partition UOR count segments
---------- -------- -------- --------- ---------- ---------- ----------
2019.05.11 22.00.00 NIGHT JPDSWIH JPIHRA1 18 1.2351E+04

Total exec Total wait

Moved bytes Moved roots Lock count time (sec) time (sec)
----------- ----------- ---------- ----------- ----------
8.3559E+05 5.2870E+03 6.5190E+03 4.8000E+01 1.1600E+01

IBM Z Performance and Capacity Analytics Report: CSQO02

Figure 41. Example of an IMS HALDB OLR Statistics report

The report contains this information:

Date
Date of the measurement. From DATE.
Time
Hour of the measurement. From TIME.
Period name
Period name. From PERIOD_NAME.
Database
Database. From DBD_NAME.
Partition
Database partition name. From PARTITION_NAME.

126 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

UOR count
The number of UORs. From UOR_COUNT.
Moved segments
The number of moved segments. From SEG_MOVED.
Moved bytes
The number of moved bytes. From SIZE_MOVED.
Moved roots
The number of moved roots. From ROOTS_MOVED.
Lock count
The count of locks. From LOCK_COUNT.
Total exec time (sec)
Total execution time, in seconds. From EXEC_TIME.
Total wait time (sec)
Total wait time, in seconds. From WAIT_TIME.

Key Performance Metrics IMS reports

IMS Key Performance Metrics (KPM) reports show Processing Times by Transaction by Hour.

KPM IMS Processing Times by Transaction by Hour report

This information identifies the report:
Report ID
CSQTL01
Report Group
KPM IMS reports
Source
KPM_IMS_TRAN_H
Attributes
IMS, Transactions, Utilization, Performance, Hourly
Variables
ORIGIN_IMS, PROCESS_IMS, DATE

Chapter 5. Report reference 127

Reports

KPM IMS Processing Times by Transaction by Hour

Origin IMS Name:'IV01'
Process IMS Name:All
Date: 2019-05-15

Trans Total Average Max Min Average Max

Hour Name Trans CPU Time CPU Time CPU Time Elapsed Elapsed
----- ---------- -------- ----------- ----------- ----------- ----------- -----------
0300 $UNKNOWN 2 0.022448 0.022531 0.022365 0.024902 0.024902
-------- ----------- -----------
2 0.022448 0.024902

0900 $UNKNOWN 155 0.049941 3.665114 0.000120 0.000064 0.002368

ACSC 1 0.006096 0.006096 0.006096 0.000000 0.000000
ACSD 4 0.015014 0.020731 0.008550 0.000000 0.000000
ACSI 1 0.023724 0.023724 0.023724 0.000000 0.000000
ALPC 2 0.054317 0.103725 0.004909 0.000000 0.000001
ALPP 12 0.010634 0.021458 0.002382 0.000000 0.000000

Min Average DB Max DB Min DB Average DB Max DB Min DB

Elapsed IO Time IO Time IO Time Lock Time Lock Time Lock Time
----------- ----------- ----------- ----------- ----------- ----------- -----------
0.024901 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
----------- -----------
0.000000 0.000000

0.000000 0.084121 12.305070 0.000000 0.032270 4.272377 0.000000

0.000000 0.000000 0.000000 0.000000 0.012567 0.012567 0.012567
0.000000 0.001720 0.003757 0.000000 0.001877 0.003498 0.000511
0.000000 0.000000 0.000000 0.000000 0.001251 0.001251 0.001251
0.000000 0.006486 0.012972 0.000000 0.000327 0.000653 0.000000
0.000000 0.000000 0.000000 0.000000 0.000016 0.000108 0.000000

IBM Z Performance and Capacity Analytics Report: CSQTL01

Figure 42. Example of a KPM IMS Processing Times by Transaction by Hour report

KPM IMS Processing Times by PSB Name by Hour report

This information identifies the report:
Report ID
CSQTL02
Report Group
KPM IMS reports
Source
KPM_IMS_TRAN_H
Attributes
IMS, PSB, Utilization, Performance, Hourly
Variables
ORIGIN_IMS, PROCESS_IMS, DATE

128 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

KPM IMS Processing Times by PSB Name by Hour

Origin IMS Name:'IV01'
Process IMS Name:All
Date: 2019-05-15

Total Average Max Min Average Max

Hour PSB Name Trans CPU Time CPU Time CPU Time Elapsed Elapsed
----- ---------- -------- ----------- ----------- ----------- ----------- -----------
0300 CQX002 1 0.022531 0.022531 0.022531 0.024901 0.024901
RXX007 1 0.022365 0.022365 0.022365 0.024902 0.024902
-------- ----------- -----------
2 0.022448 0.024902

0900 ALP001 8 0.001381 0.001718 0.001190 0.000300 0.000300

ALP005 76 0.031569 0.388543 0.001413 0.000000 0.000005
ALX005 185 0.006624 0.128647 0.001277 0.000000 0.000002
ALX008 21 0.015122 0.103725 0.002382 0.000000 0.000001
AMP900 4 0.935952 3.665114 0.001105 0.000006 0.000025

Min Average DB Max DB Min DB Average DB Max DB Min DB

Elapsed IO Time IO Time IO Time LOCK Time LOCK Time LOCK Time
----------- ----------- ----------- ----------- ----------- ----------- -----------
0.024901 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
0.024902 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
----------- -----------
0.000000 0.000000

0.000300 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000

0.000000 0.000759 0.014693 0.000000 0.001207 0.046265 0.000000
0.000000 0.000183 0.015206 0.000000 0.000055 0.002894 0.000000
0.000000 0.000618 0.012972 0.000000 0.000069 0.000653 0.000000
0.000000 3.087085 12.305070 0.000000 1.068497 4.272377 0.000000

IBM Z Performance and Capacity Analytics Report: CSQTL02

Figure 43. Example of a KPM IMS Processing Times by PSB Name by Hour report

KPM IMS Processing Times by Region Type by Hour report

This information identifies the report:
Report ID
CSQTL03
Report Group
KPM IMS reports
Source
KPM_IMS_TRAN_H
Attributes
Attributes
Variables
ORIGIN_IMS, PROCESS_IMS, DATE

Chapter 5. Report reference 129

Reports

KPM IMS Processing Times by Region Type by Hour

Origin IMS Name:'IV01'
Process IMS Name:All
Date: 2019-05-15

Region Total Average Max Min Average Max

Hour Type Trans CPU Time CPU Time CPU Time Elapsed Elapsed
----- ---------- -------- ----------- ----------- ----------- ----------- -----------
0300 IFP 2 0.022448 0.022531 0.022365 0.024902 0.024902
-------- ----------- -----------
2 0.022448 0.024902

0900 BMP 611 0.014355 3.665114 0.000028 0.000012 0.000300

IFP 29 0.002579 0.009025 0.000120 0.000163 0.002368
MPP 370 0.005728 0.038287 0.000888 0.000000 0.000001
WMPP 1,311 0.003908 0.128647 0.000887 0.000000 0.000003
-------- ----------- -----------
2,321 0.006932 0.000005

Min Average DB Max DB Min DB Average DB Max DB Min DB

Elapsed IO Time IO Time IO Time Lock Time Lock Time Lock Time
----------- ----------- ----------- ----------- ----------- ----------- -----------
0.024901 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
----------- -----------
0.000000 0.000000

0.000000 0.021340 12.305070 0.000000 0.008186 4.272377 0.000000

0.000000 0.000650 0.015173 0.000000 0.000011 0.000157 0.000000
0.000000 0.002763 0.080835 0.000000 0.000384 0.017589 0.000000
0.000000 0.000729 0.189007 0.000000 0.001757 1.998152 0.000000
----------- -----------
0.006478 0.003209

IBM Z Performance and Capacity Analytics Report: CSQTL03

Figure 44. Example of a KPM IMS Processing Times by PSB Name by Hour report

KPM IMS Average Enqueues by Transaction by Hour report

This information identifies the report:
Report ID
CSQTL04
Report Group
KPM IMS reports
Source
KPM_IMS_TRAN_H
Attributes
IMS, Enqueues, Utilization, Performance, Hourly
Variables
ORIGIN_IMS, PROCESS_IMS, DATE

130 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

KPM IMS Average Enqueues by Transaction by Hour

Origin IMS Name:'IV01'
Process IMS Name:All
Date: 2019-05-15

Ave Ave Ave Ave Ave Ave Ave

Trans Total Test Test Test Queue Queue Queue Updt
Hour Name Trans Enqueues Enqwaits Dequeues Enqueues Enqwaits Dequeues Enqueues
----- ---------- -------- -------- -------- -------- -------- -------- -------- --------
0300 $UNKNOWN 2 0.000 0.000 0.000 0.000 0.000 0.000 0.000
-------- -------- -------- -------- -------- -------- -------- --------
2 0.000 0.000 0.000 0.000 0.000 0.000 0.000

0900 $UNKNOWN 155 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ACSC 1 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ACSD 4 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ACSI 1 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ALPC 2 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ALPP 12 0.000 0.000 0.000 0.000 0.000 0.000 0.000
ALPQ 4 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Ave Ave Ave Ave Ave

Updt Updt Excl Excl Excl
Enqwaits Dequeues Enqueues Enqwaits Dequeues
-------- -------- -------- -------- --------
0.000 0.000 0.000 0.000 0.000
-------- -------- -------- -------- --------
0.000 0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000

IBM Z Performance and Capacity Analytics Report: CSQTL04

Figure 45. Example of a KPM IMS Average Enqueues by Transaction by Hour report

KPM IMS DB Activity by Region by Program Name by Hour report

This information identifies the report:
Report ID
CSQTL05
Report Group
KPM IMS reports
Source
KPM_IMS_TRAN_H
Attributes
IMS, Data Base Activity, Program, Performance, Hourly
Variables
ORIGIN_IMS, PROCESS_IMS, DATE

Chapter 5. Report reference 131

Reports

KPM IMS DB Activity by Region by Program Name by Hour

Origin IMS Name:'IV
Process IMS Name:A
Date: 2019-05-15

Average
Transact PSB Region Total Approx CPU Response Db2 Read Db2 Write
Hour Name Name Type Trans Utilization Time Calls Calls
----- -------- -------- -------- ----------- ----------- ----------- ---------- -----------
0300 $UNKNOWN CQX002 IFP 1 0.022531 0.024901 0 0
$UNKNOWN RXX007 1 0.022365 0.024902 0 0
----------- ----------- ----------- ---------- -----------
2 0.044896 0.024902 0 0

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

2 0.044896 0.024902 0 0

0900 $UNKNOWN ALP001 BMP 8 0.011050 0.000300 0 0

$UNKNOWN ALP005 76 2.399230 0.000000 223 55
----------- ----------- ----------- ---------- -----------
84 2.410280 0.000029 223 55

ALVS ALX005 WMPP 185 1.225476 0.000000 247 0

ALPC ALX008 2 0.108633 0.000000 7 0
ALPP ALX008 12 0.127605 0.000000 52 0
ALPQ ALX008 4 0.052412 0.000000 16 0
ALRP ALX008 1 0.012439 0.000000 0 0
ALTH ALX008 2 0.016463 0.000000 0 0
----------- ----------- ----------- ---------- -----------
206 1.543029 0.000000 322 0

Db2 Delete Db2 Total

Calls Calls VSAM Reads VSAM Writes OSAM Reads OSAM Writes Total IOS
----------- ----------- ----------- ----------- ----------- ----------- -----------
0 0 0 0 0 0 0
0 0 0 0 0 0 0
----------- ----------- ----------- ----------- ----------- ----------- -----------
0 0 0 0 0 0 0

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

0 0 0 0 0 0 0

0 0 0 0 0 0 0
0 278 14 0 0 0 14
----------- ----------- ----------- ----------- ----------- ----------- -----------
0 278 14 0 0 0 14

0 247 16 0 0 0 16
0 7 3 0 0 0 3
0 52 0 0 0 0 0
0 16 0 0 0 0 0
0 0 0 0 0 0 0
0 0 0 0 0 0 0
----------- ----------- ----------- ----------- ----------- ----------- -----------
0 322 19 0 0 0 19

IBM Z Performance and Capacity Analytics Report: CSQTL05

Figure 46. Example of a KPM IMS DB Activity by Region by Program Name by Hour report

KPM IMS ATF Transaction Statistics report

This report shows ATF statistics on transaction level. It can be filtered by transaction name, and the
output can be limited to specific hourly intervals only.
This information identifies the report:
Report ID
CSQATF01
Report Group
IMS KPM reports
Source
KPM_IMS_ATF_T
Attributes
IMS, TRANSACTION, ATF
Variables
DATE, FROM_HOUR, TO_HOUR, TRANSACTION_NAME

132 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

KPM IMS ATF Transaction Statistics

Date: 19.08.2019 Transaction: All

+~~~~~~~~~~~~~~~~~~~~Elapsed Seconds~~~~~~~~~~~~~~~~>
Start Date/Time Tran Total DLI Db2 MQ
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~ ~~~~~~~~~~~~ ~~~~~~~~~~~~
2019-08-19-18.02.11.681753 ABC00030 0.540473 0.538344 0.000000 0.000000
2019-08-19-18.02.11.683860 ABC00110 0.537017 0.536584 0.000000 0.000000
2019-08-19-18.02.12.222528 ABC00180 0.244198 0.241669 0.000000 0.000000
2019-08-19-18.02.12.223214 ABC00050 0.237932 0.236555 0.000000 0.000000
2019-08-19-18.02.12.466206 ABC00050 0.069113 0.066197 0.000000 0.000000
2019-08-19-18.02.12.467445 ABC00180 0.058003 0.057599 0.000000 0.000000
2019-08-19-18.02.12.526499 ABC00060 0.002929 0.000849 0.000000 0.000000
2019-08-19-18.02.12.529914 ABC00060 0.023332 0.015366 0.000000 0.000000
2019-08-19-18.02.12.535319 ABC00050 0.002768 0.002069 0.000000 0.000000
2019-08-19-18.02.12.538409 ABC00020 0.004063 0.003069 0.000000 0.000000

+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~CPU Seconds~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~>
Control DLISAS Other
Total DLI Db2 MQ Region Region Region
~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~
0.013989 0.013250 0.000000 0.000000 0.000000 0.000000 0.000000
0.014002 0.013772 0.000000 0.000000 0.000000 0.000000 0.000000
0.007740 0.007021 0.000000 0.000000 0.000000 0.000000 0.000000
0.007613 0.007070 0.000000 0.000000 0.000000 0.000000 0.000000
0.005628 0.005423 0.000000 0.000000 0.000000 0.000000 0.000000
0.002319 0.002089 0.000000 0.000000 0.000000 0.000000 0.000000
0.000626 0.000163 0.000000 0.000000 0.000000 0.000000 0.000000
0.002794 0.002738 0.000000 0.000000 0.000858 0.000000 0.000000
0.000391 0.000209 0.000000 0.000000 0.000000 0.000000 0.000000
0.000597 0.000316 0.000000 0.000000 0.000000 0.000000 0.000000

+~~~~~~~~~~~~~~~~~~~~~~~~~~~Transaction Details~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~>
PSB Logical User Region Tran
Name Terminal ID Jobname Type Class
~~~~~~~~~~~~~~~ ~~~~~~~~~~~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001

IBM Z Performance and Capacity Analytics Report: CSQATF01

Figure 47. Example of a KPM IMS ATF Transaction Statistics report

The report contains this information:

Start Date/Time
The start date and time of the transaction.
Tran
The name of the IMS transaction.
Elapsed Seconds Total
Total elapsed time, in seconds.
Elapsed Seconds DLI
Total DLI elapsed time, in seconds.
Elapsed Seconds Db2
Total Db2 elapsed time, in seconds.
Elapsed Seconds MQ
Total MQ elapsed time, in seconds.
CPU Seconds Total
Total CPU time, in seconds.
CPU Seconds DLI
Total DLI CPU time, in seconds.
CPU Seconds Db2
Total Db2 CPU time, in seconds.
CPU Seconds MQ
Total MQ CPU time, in seconds.
CPU Seconds Control Region
Total CPU time, in seconds, in CTL region.
CPU Seconds DLISAS Region
Total CPU time, in seconds, in DLISAS region.

Chapter 5. Report reference 133

Reports

CPU Seconds Other Region

Total CPU time, in seconds, in all other regions.
PSB Name
The name of the IMS program used to process the transaction. This is the program specification block
(PSB).
Logical Terminal
The name of the logical terminal on which the activity occurred.
User ID
The user ID.
Jobname
The job name.
Region Type
The type of IMS Region.
Tran Class
The transaction class.

KPM IMS ATF Transaction Events report

This report shows ATF statistics on transaction event level. It can be filtered by transaction name, and the
output can be limited to specific hourly intervals only.
This information identifies the report:
Report ID
CSQATF02
Report Group
IMS KPM reports
Source
KPM_IMS_ATF_EVT_T
Attributes
IMS, TRANSACTION, ATF, EVENT
Variables
DATE, FROM_HOUR, TO_HOUR, TRANSACTION_NAME

134 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

KPM IMS ATF Transaction Events

Date: 19.08.2019 Transaction: All

+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Elapsed Seconds~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
Start Date/Time Tran Total DLI Db2 MQ
~~~~~~~~~~~~~~~~~~~~~~~~~~~- ~~~~~~~~~~~~ ~~~~~~~~~~- ~~~~~~~~~~ ~~~~~~~~~~- ~~~~~~~~~
2019-08-19-18.02.11.681753 ABC00030 0.540473 0.538344 0.000000 0.000000
2019-08-19-18.02.11.683860 ABC00110 0.537017 0.536584 0.000000 0.000000
2019-08-19-18.02.12.222528 ABC00180 0.244198 0.241669 0.000000 0.000000
2019-08-19-18.02.12.223214 ABC00050 0.237932 0.236555 0.000000 0.000000
2019-08-19-18.02.12.466206 ABC00050 0.069113 0.066197 0.000000 0.000000
2019-08-19-18.02.12.467445 ABC00180 0.058003 0.057599 0.000000 0.000000
2019-08-19-18.02.12.526499 ABC00060 0.002929 0.000849 0.000000 0.000000
2019-08-19-18.02.12.529914 ABC00060 0.023332 0.015366 0.000000 0.000000
2019-08-19-18.02.12.535319 ABC00050 0.002768 0.002069 0.000000 0.000000
2019-08-19-18.02.12.538409 ABC00020 0.004063 0.003069 0.000000 0.000000

+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~CPU Seconds~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~+
Control DLISAS Other
Total DLI Db2 MQ Region Region Region
~~~~~~~~~- ~~~~~~~~~- ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~- ~~~~~~~~~- ~~~~~~~~~~-
0.013989 0.013250 0.000000 0.000000 0.000000 0.000000 0.000000
0.014002 0.013772 0.000000 0.000000 0.000000 0.000000 0.000000
0.007740 0.007021 0.000000 0.000000 0.000000 0.000000 0.000000
0.007613 0.007070 0.000000 0.000000 0.000000 0.000000 0.000000
0.005628 0.005423 0.000000 0.000000 0.000000 0.000000 0.000000
0.002319 0.002089 0.000000 0.000000 0.000000 0.000000 0.000000
0.000626 0.000163 0.000000 0.000000 0.000000 0.000000 0.000000
0.002794 0.002738 0.000000 0.000000 0.000858 0.000000 0.000000
0.000391 0.000209 0.000000 0.000000 0.000000 0.000000 0.000000
0.000597 0.000316 0.000000 0.000000 0.000000 0.000000 0.000000

+~~~~~~~~~~~~~~~~-Transaction Details~~~~~~~~~~~~~~~~~+
PSB Logical User Region Tran
Name Terminal ID Jobname Type Class
~~~~~~~~~- ~~~~~~~~- ~~~~~~~~- ~~~~~~~~~~ ~~~~~~ ~~~~~-
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS100 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX2 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001
MPSYS000 VM001XY USER99D IMSREGX1 MPP 0001

Start Date/Time Tran Event Type Count

~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~ ~~~~~~~~~~~~~~~~~ ~~~~~~
2019-08-19-18.02.11.681753 ABC00030 DLI DB GHNP 2
DLI DB GHU 2

+~~~-Elapsed Seconds~~~~> +~~~-CPU Seconds~~~~>

Total Average Total Average
~~~~~~~~~~~- ~~~~~~~~~~~- ~~~~~~~~~- ~~~~~~~~~-
0.076725 0.038363 0.001763 0.000881
0.098235 0.049117 0.002573 0.001287
0.004768 0.004768 0.000173 0.000173
0.352112 0.176056 0.008549 0.004275
0.000010 0.000003 0.000010 0.000003
0.000010 0.000010 0.000010 0.000010
0.006488 0.003244 0.000177 0.000088
0.000067 0.000067 0.000023 0.000023
0.506789 0.000693 0.008224 0.000013
0.533698 0.053820 0.013497 0.001380
0.000013 0.000004 0.000012 0.000004
0.000012 0.000012 0.000012 0.000012
0.002868 0.001434 0.000256 0.000128
0.504625 0.000542 0.008567 0.000014
0.142141 0.035535 0.003529 0.000882
0.099529 0.041935 0.003494 0.001456
0.000002 0.000002 0.000002 0.000002
0.001641 0.001641 0.000044 0.000044
0.228324 0.000625 0.004408 0.000013
0.000671 0.000671 0.000010 0.000010
0.000016 0.000008 0.000016 0.000008
0.236536 0.060509 0.007052 0.001878
0.000006 0.000006 0.000005 0.000005

IBM Z Performance and Capacity Analytics Report: CSQATF02

Figure 48. Example of a KPM IMS ATF Transaction Events report

The report contains this information:

Start Date/Time
The start date and time of the transaction.
Tran
The name of the IMS transaction.
Event
The event name.
Type
The event type.
Count
The number of events.

Chapter 5. Report reference 135

Reports

Elapsed Seconds Total

Total elapsed time, in seconds.
Elapsed Seconds Average
Average elapsed time, in seconds.
CPU Seconds Total
Total CPU time, in seconds.
CPU Seconds Average
Average CPU time, in seconds.

KPM IMS ATF Transaction Database Activity report

This report shows database activity on transaction level. It can be filtered by transaction name, and the
output can be limited to specific hourly intervals only.
This information identifies the report:
Report ID
CSQATF03
Report Group
IMS KPM reports
Source
KPM_IMS_ATF_EVT_T
Attributes
IMS, TRANSACTION, ATF, DATABASE
Variables
DATE, FROM_HOUR, TO_HOUR, TRANSACTION_NAME
KPM IMS ATF Transaction Database Activity
Date: 19.08.2019 Transaction: All

Start Date/Time Tran PSB Name Database Call Type Count

-------------------------- -------- -------- ---------- ---------- ------
2019-08-19-18.02.11.681753 ABC00030 MPSYS030 B01X1O0 READ 1
DBC1ABO0 READ 3
DBC1ABO0 UPDATE 1
DB12DHO0 READ 1
DB12DHO0 UPDATE 1
2019-08-19-18.02.11.683860 ABC00110 MPSYS110 DBPRDLOA READ 3
DB$TESTX READ 5
2019-08-19-18.02.12.222528 ABC00180 MPSYS180 DBPRDLOA READ 1
DBXHDHO0 READ 6
2019-08-19-18.02.12.223214 ABC00050 MPSYS050 DBDRL01K READ 3
DBXHDHO0 READ 3
2019-08-19-18.02.12.466206 ABC00050 MPSYS050 DB01X1O0 DELETE 1
DB01X1O0 INSERT 1
DB01X1O0 READ 3
2019-08-19-18.02.12.467445 ABC00180 MPSYS180 DBMIGRAT READ 2
DBXHDHO0 READ 8
2019-08-19-18.02.12.526499 ABC00060 MPSYS060 DBDRL01K READ 5
DBDRL01K UPDATE 3
DBXHDHO0 READ 1
2019-08-19-18.02.12.529914 ABC00060 MPSYS060 DBMIGRTS READ 1
DBXHDHO0 READ 4
DBXHDHO0 UPDATE 2
2019-08-19-18.02.12.535319 ABC00050 MPSYS050 DB01X1O0 READ 1
DBDRL01K READ 1
DBXHDHO0 READ 3

IBM Z Performance and Capacity Analytics Report: CSQATF03

Figure 49. Example of a KPM IMS ATF Transaction Database Activity report

The report contains this information:

Start Date/Time
The start date and time of the transaction.
Tran
The name of the IMS transaction.
PSB Name
The name of the IMS program used to process the transaction. This is the program specification block
(PSB).
Database
The database name.

136 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Reports

Call Type
The database call type.
Count
The number of calls.

Chapter 5. Report reference 137

Reports

138 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Chapter 6. Customization reference
This section describes some sample jobs, utilities, and tailoring examples.

Creating IMS log record DSECTs

This section can help you create log record dummy control sections (DSECTs) for use with the IMS
Performance Feature.
Figure 50 on page 139 shows an example of a job that you can use to assemble the log records mapping
macro ILOGREC that is provided with IMS. The sample shown is the most inclusive method of extracting
DSECTs. You can modify it to suit your own needs. You can write your own record procedures that use the
composite record created by the IMS Performance Feature.

//USERIDA JOB (ACC000,001),'ILOGREC',

// NOTIFY=USERID,MSGCLASS=A,CLASS=A,REGION=0M
//DSECT PROC
//HASM EXEC PGM=IEV90,PARM='NODECK,NOXREF,LIST,NORLD,NOOBJECT'
//SYSLIB DD DISP=SHR,DSN=IMS71.GENLIBB,DCB=SYS1.MACLIB
// DD DISP=SHR,DSN=IMS71.GENLIBA
// DD DISP=SHR,DSN=IMS71.GENLIB
// DD DISP=SHR,DSN=SYS1.MACLIB
//SYSUT1 DD UNIT=SYSDA,SPACE=(CYL,(2,1))
//SYSUT2 DD UNIT=SYSDA,SPACE=(CYL,(2,1))
//SYSUT3 DD UNIT=SYSDA,SPACE=(CYL,(2,1))
//SYSPRINT DD SYSOUT=*
// PEND
//D1 EXEC DSECT /* Full function record DSECTs */
//SYSIN DD *
ILOGREC RECID=ALL
END
/*
//D2 EXEC DSECT /* Fast path record DSECTs */
//SYSIN DD *
DBFLGRIM
EJECT
DBFLGROM
EJECT
DBFBMSDB
EJECT
DBFDOCL
EJECT
DBFLGRDQ
EJECT
DBFLGSYN
EJECT
DBFLGRIC
EJECT
DBFLSRT RECID=50
EJECT
DBFLGRSD
EJECT
DBFLGRRE
END
/*

Figure 50. Sample JCL for assembling IMS log record DSECTs

//D3 EXEC DSECT /* IMS control block DSECTs */

//SYSIN DD *
IBPOOL
EJECT
IDLIVSAM BFSP
EJECT
ICLI CTTBASE=1
EJECT
ICLI CNTBASE=1
EJECT
ICLI SPQBASE=1
EJECT
ICLI CTBBASE=1

© Copyright IBM Corp. 1994, 2017 139

 EJECT
ICLI CIBBASE=1
EJECT
ICLI PCIBASE=1
EJECT
ICLI CRBBASE=1
EJECT
ICLI CLBBASE=1
EJECT
ICLI CVBBASE=1
EJECT
ICLI CCBBASE=1
EJECT
ICLI BUFBASE=1
EJECT
ICLI POOBASE=1
EJECT
ICLI BFRBASE=1
EJECT
ICLI QDSBASE=1
EJECT
ICLI QPCBASE=1
EJECT
ICLI QLGBASE=1
EJECT
ICLI FREBASE=1
EJECT
ICLI FBPBASE=1
EJECT
ICLI SPABASE=1
EJECT
ICLI FEIBBASE=1
EJECT
IAPS SMBBASE=1
EJECT
IAPS TCTBASE=1
EJECT
IDLI DBSBASE=1
EJECT
IDLI DDRBASE=1
EJECT
IDLI DMBBASE=1
EJECT
IDLI DPCBASE=1
EJECT
IDLI DSGBASE=1
EJECT
IDLI FDBBASE=1
EJECT
IDLI FLDBASE=1
EJECT
IDLI FSBBASE=1
EJECT
IDLI FUNBASE=1
EJECT
IDLI JCBBASE=1
EJECT
IDLI LEVBASE=1
EJECT
IDLI PDRBASE=1
EJECT
IDLI PDVBASE=1
EJECT
IDLI PSBBASE=1
EJECT
IDLI PSHBASE=1
EJECT
IDLI PSPBASE=1
EJECT
IDLI PSTBASE=1
EJECT
IDLI PS24BASE=1
EJECT
IDLI SDBBASE=1
EJECT
IDLI SMBASE=1
EJECT
IDLI TPCBASE=1
EJECT
IDLI CALLER=1
EJECT
ISCD

140 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 END
//

Sample archive exit

This appendix shows a description from the sample IMS archive exit, DRLJXIMS. The source for this
sample exit can also be found in member DRLJXIMS in the product CNTL library. Figure 51 on page 141
shows the sample exit.
To assemble this exit, you need access to both the MVS and the IMS macro libraries. Refer to your IMS
system documentation for detailed information regarding the archive exit.

TITLE 'DRLJXIMS - IMS ARCHIVE EXIT'

DRLJXIMS CSECT
SPACE
***********************************************************************
* *
* MODULE NAME: DRLJXIMS *
* *
* DESCRIPTION: IMS ARCHIVE EXIT FOR EPDM/IMS *
* *
* COPYRIGHT: NONE *
* *
* STATUS: IMS/ESA V3R1 *
* *
* FUNCTION: *
* WRITES THE RECORDS USED BY EPDM/IMS TO THE FILE DEFINED BY THE *
* DDNAME IMSLOG. *
* *
* THE RECORD TYPES CONCERNED ARE AS FOLLOWS (ALL IN HEX) *
* *
* 01, 02, 03, 06, 07, 08, 0A, 11, 12, 13, 16, 24, *
* 31, 32, 33, 34, 35, 36, 37, 38, 42, 45, *
* 47, 55 AND 56 *
* 40, SUBTYPES 01, 04, AND 98. *
* 59, SUBTYPES 01, 03, 36, 37, AND 38. *
* *
* THIS MINIMIZES THE AMOUNT OF DATA PASSED TO EPDM/IMS. *
* *
* LOGIC: *
* CASE INIT (DRLJXIMS CALL CODE 1). *
* GETMAIN STORAGE FOR WORK AREAS AND ANCHOR IT IN THE USER *
* WORD. *
* OPEN OUTPUT FILE. *
* END CASE INIT. *
* *
* CASE NORMAL (DRLJXIMS CALL CODE 2). *
* *
* SUBCASE RECORD TYPES *
* 01, 02, 03, 06, 07, 08, 0A, 11, 12, 13, *
* 16, 24, 31, 32, 33, 34, 35, 36, 37, 38, *
* 42, 45, 47, 55 AND 56 *
* COPY RECORD. *
* END SUBCASE RECORD TYPES *
* 01, 02, 03, 06, 07, 08, 0A, 11, 12, 13, *
* 16, 24, 31, 32, 33, 34, 35, 36, 37, 38, *
* 42, 45, 47, 55 AND 56 *
* *
* SUBCASE RECORD TYPE *
* 40, SUBTYPES 01, 04, AND 98. *
* COPY RECORD. *
* END SUBCASE RECORD TYPE *
* 40, SUBTYPES 01, 04, AND 98. *
* *
* SUBCASE RECORD TYPE *
* 59, SUBTYPES 01, 03, 36, 37, AND 38. *
* COPY RECORD. *
* END SUBCASE RECORD TYPE *
* 59, SUBTYPES 01, 03, 36, 37, AND 38. *
* *
* END CASE NORMAL. *
* *
* CASE TERMINATE (DRLJXIMS CALL CODE 3). *
* CLOSE OUTPUT FILE. *
* FREEMAIN STORAGE FOR WORK AREAS AND RESET ANCHOR POINTER. *
* END CASE TERMINATE. *

Figure 51. Description from the sample IMS archive exit

DFSLTMG0 log merge utility

The DFSLTMG0 log merge utility produces one data set by merging the system log data sets (SLDS) from
two or more IMS systems. The log merge utility can merge up to nine IMS system logs from the same IMS
release. Each log is the output of a uniquely identified IMS system running during the same time span.
The order of input to the log merge utility is LOG01, LOG02... LOG09. DFSLTMG0 is placed in IMS.RESLIB
during IMS system definition.

Chapter 6. Customization reference 141

 //STEP0 EXEC PGM=DFSLTMG0
//STEPLIB DD DSNAME=IMS.RESLIB,DISP=SHR
//SYSPRINT DD SYSOUT=*
//PRINT DD SYSOUT=*
//LOG01 DD DSNAME=xxx.IMSLOG01,DISP=OLD
//LOG02 DD DSNAME=xxx.IMSLOG02,DISP=OLD
//LOGOUT DD DSNAME=xxx.IMSLOG.MERGE,DISP=(NEW,CATLG)
// VOL=SER=yyyy,UNIT=3390,SPACE=(CYL,(100,5)),
// DCB==(RECFM=VB,LRECL=27994,BLKSIZE=32760)
//SYSIN DD *
START 01157,0001
STOP 01157,2359
MSG
//*

Figure 52. DFSLTMG0 log merge utility

Following is the description of the statements:

STEPLIB DD
This statement points to IMS.RESLIB, which contains the IMS nucleus and required action modules.

//STEPLIB DD DSNAME=IMS.RESLIB,DISP=SHR

PRINT DD
Indicates the SYSPRINT data set used for control statements and error messages.

//PRINT DD SYSOUT=A

LOG01 DD
Describes the first input log data set.

//LOG01 DD DSNAME=IMS.LOGA,DISP=OLD,
// VOL=SER=XXXXXX,UNIT=TAPE

LOG02 DD
Describes the second input log data set.

//LOG02 DD DSNAME=IMS.LOGB,DISP=OLD,
// VOL=SER=XXXXXX,UNIT=TAPE

LOGOUT DD
Describes the output data set.

//LOGOUT DD DSNAME=IMS.LOGOUT,DISP=(,PASS),
// VOL=SER=YYYYYY,UNIT=TAPE,
// DCB=(RECFM=VBS,LRECL=6000,BLKSIZE=6008)

SYSIN DD
Describes the control statement data set.

//SYSIN DD *

START
Used to specify a start time. This statement must be present (yyddd, hhmmsstt).

START

STOP
You must specify a stop time, which must be relative to the time field in LOG01 (yyddd, hhmmsstt)

STOP

Log Record Selection

Use this control statement to merge only certain types of log records.

142 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Controlling the log merge

About this task

To control the log output, you need to:

Procedure
1. Choose the required systems to participate in the logical link paths you want to examine.
2. Coordinate the series of input logs for each system so that they cover a similar time span.
3. Specify a start and stop time for the Log Merge utility control statements if you want to sample the
cross-system processing for a particular interval.
4. Specify MSG to select log records that are suitable for the transaction analysis step. Records is the
default, but this means the DL/I activity for several systems is included in the utility input, and this can
cause extended processing time.

Tailoring example using MSGTEXT for IMS_TRAN_x tables

About this task

The MSGTEXT log procedure parameter allows you to extract up to 60 bytes of user data from the
MSGXSTXT field of type X'01' and type X'03' log records and store that data in the CQS feature
transaction data tables. To store the data, you must customize at least some of your record, update, and
table definitions. This is an example customization. For more information about specifying the MSGTEXT
parameter see “Specifying log procedure parameters” on page 25.
To store the extracted data into Db2 you will need to customize your IMS feature tables. The following
instructions describe how to store the data in the IMS_TRAN_x tables. Similar changes would be required
for the other transaction-oriented tables. You can give the column any name you like, but you should be
consistent.
This example covers IMS V12 only (additional members would need to be customized to support other
versions of IMS). In this example the DRLIPARM parameter MSGTEXT=1,10 is used to extract 10 bytes of
user data from the MSGXSTXT field of the IMS X'01' and X'03' records. The new 10 byte field will be a
new key field in the database. These are the Db2 table definition changes required for the IMS_TRAN_x
tables.
Copy the members indicated to your LOCAL.DEFS data set and perform the customization as indicated:

Procedure
1. Customize SDRLDEFS(DRLTCSQR).
This contains the definitions of the IMS_TRAN_x tables and indexes.
a) Add USER_DATA CHAR(10) NOT NULL to the list of key columns in each of the table definitions
following the TRANSACTION_NAME column.
b) Add USER_DATA to the primary key definition in each of the table definitions and to the list of key
columns in the index definitions.
2. Customize SDRLDEFS(DRLRSC1C). This contains the CSQ_VC10_R2 and CSQ_VC10_R2_LIGHT record
definitions.
a) At the start of the definitions for CSQ_VC10_R2 and CSQ_VC10_R2_LIGHT, change the string in the
VERSION statement following the DEFINE to a unique string for your modification.
b) At the start of the definition for CSQ_VC10_R2_LIGHT, either remove the lineIDENTIFIED BY
DRL_LENGTH = 246, or change it to include the new record length IDENTIFIED BY
DRL_LENGTH = 246 OR DRL_LENGTH = nnn, where nnn is the record length of the extended
record containing the USER_DATA. In this example nnn will be 258 (246 + 2 (length prefix) + 10
(user-data)).

Chapter 6. Customization reference 143

 c) At the end of the definitions for CSQ_VC10_R2 and CSQ_VC10_R2_LIGHT replace the following:

DRL_TXCLS LENGTH 2 HEX); -- Transaction class

with

DRL_TXCLS LENGTH 2 HEX, -- Transaction class

DRL_MSGTXL LENGTH 2 BINARY, - User data length
DRL_MSGTXT CHAR(10)); - User data

3. Customize SDRLDEFS(DRLUIC1C).
This contains the CSQVC10TRANH and CSQVC10TRANLH update definitions. At the start of the
definitions:
a) Change the string in the VERSION statement following the DEFINE to a unique string for your
modification.
b) In the LET clause for each definition following TRANSACTION_NAME add:

W_USERDATA = CASE DRL_MSGTXL

WHEN 10 THEN DRL_MSGTXT
ELSE '$BLANK '
END

c) Following the TRANSACTION_NAME assignment in the GROUP BY clause for each definition add:

USER_DATA = W_USERDATA ,

4. Customize SDRLDEFS(DRLUIMSC). This contains the IMS_TRAN_D, IMS_TRAN_H, and IMS_TRAN_W

update definitions. At the start of the definitions:
• Change the string in the VERSION statement following the DEFINE to a unique string for your
modification.
• In the GROUP BY clause for each definition following the TRANSACTION_NAME assignment add:

USER_DATA = USER_DATA ,

5. Install or Reinstall the Component to activate the local modifications.

a) If the IMS 12.1 CSQ Collect Component has not been installed previously:
• Install the IMS 12.1 CSQ Collect Component using the new definitions created in your
LOCAL.DEFS data set.
• The IMS_TRAN_H, IMS_TRAN_D, and IMS_TRAN_W tables will contain the new definition of your
USER_DATA field.
b) If the IMS 12.1 CSQ Collect Component is already installed:
• UNLOAD the data you have stored in the IMS_TRAN_H, IMS_TRAN_D, and IMS_TRAN_W tables.
• DROP the IMS_TRAN_H, IMS_TRAN_D, and IMS_TRAN_W tables.
• Reinstall the IMS 12.1 CSQ Collect Component using the new definitions stored in your
LOCAL.DEFS data set.
• The IMS_TRAN_H, IMS_TRAN_D, and IMS_TRAN_W tables will be recreated and now contain the
new definition of your USER_DATA field.
• Re-LOAD your data to the IMS_TRAN_H, IMS_TRAN_D, and IMS_TRAN_W tables. You will need to
add the new field names to the UNLOAD SYSPUNCH control statements created by the UNLOAD,
these are used as SYSIN to re-LOAD the tables.

144 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Accessing the extracted information

About this task

By specifying a valid MSGTEXT parameter, a length-prefixed data field is appended to the IMS feature
transaction (or R2) record produced when the IMS log is processed. You can use the DRLSLOGP program
to produce a file containing the R2 records for review.

Procedure
In your DRLSLOGP JCL specify a DRLIRPT2 data set with a RECFM=F or RECFM=V, with an LRECL of at
least 246 + the length from the MSGTEXT parameter, plus two bytes for the user data length prefix, plus
the length from the MSGTEXT parameter you coded in your DRLIPARM.
Where data is available, in the R2 record, you will see the 2 byte hex length at column 247, followed by
the selected data starting at column 249.

Chapter 6. Customization reference 145

146 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
 Support information

Appendix A. Support information

If you have a problem with your IBM software, you want to resolve it quickly. IBM provides a number of
ways for you to obtain the support you need.
• Searching knowledge bases: You can search across a large collection of known problems and
workarounds, Technotes, and other information.
• Obtaining fixes: You can locate the latest fixes that are already available for your product.
• Contacting IBM Software Support: If you still cannot solve your problem, and you need to work with
someone from IBM, you can use a variety of ways to contact IBM Support. See “Contacting IBM
Support” on page 147 for more information.

Contacting IBM Support

This topic describes how to contact IBM Support if you have been unable to resolve a problem with IBM Z
Performance and Capacity Analytics.
Before contacting IBM Support, your company must have an active IBM software maintenance contract,
and you must be authorized to submit problems to IBM. The type of software maintenance contract that
you need depends on the type of product you have. For more information, refer to the IBM Support
website at the following links:
IBM Support
https://www.ibm.com/mysupport/s/
IBM Z Support
https://www.ibm.com/support/pages/ibm-enterprise-support-and-preferred-care-options-ibm-z
To contact IBM Support to report a problem (open a case), follow these steps:
1. Determine the business impact.
2. Describe the problem and gather information.
3. Submit the problem report.

Determining the business impact

When you report a problem to IBM, you are asked to supply a severity level. Therefore, you need to
understand and assess the business impact of the problem that you are reporting. Use the following
criteria:
Severity 1
The problem has a critical business impact. You are unable to use the program, resulting in a critical
impact on operations. This condition requires an immediate solution.
Severity 2
The problem has a significant business impact. The program is usable, but it is severely limited.
Severity 3
The problem has some business impact. The program is usable, but less significant features (not
critical to operations) are unavailable.
Severity 4
The problem has minimal business impact. The problem causes little impact on operations, or a
reasonable circumvention to the problem was implemented.

Describing the problem and gathering information

When describing a problem to IBM, be as specific as possible. Include all relevant background
information so that IBM Support specialists can help you solve the problem efficiently. To save time,
know the answers to the following questions:

© Copyright IBM Corp. 1994, 2017 147

Support information

• What software versions were you running when the problem occurred?
• Do you have logs, traces, and messages that are related to the problem symptoms? IBM Support is
likely to ask for this information.
• Can you re-create the problem? If so, what steps were performed to re-create the problem?
• Did you make any changes to the system? For example, did you make changes to the hardware,
operating system, networking software, product-specific customization, and so on.
• Are you currently using a workaround for the problem? If so, be prepared to explain the workaround
when you report the problem.

Submitting the problem

You can submit your problem to IBM Support in either of the following ways:
Online
Go to https://www.ibm.com/mysupport/s/, click on Open a case, and enter the relevant details into
the online form.
By email or phone
For the contact details in your country, go to the IBM Support website at https://www.ibm.com/
support/. Look for the tab on the right and click Contact and feedback > Directory of worldwide
contacts for a list of countries by geographic region. Select your country to find the contact details for
general inquiries, technical support, and customer support.
If the problem you submit is for a software defect or for missing or inaccurate documentation, IBM
Support creates an Authorized Program Analysis Report (APAR). The APAR describes the problem in
detail. Whenever possible, IBM Support provides a workaround that you can implement until the APAR is
resolved and a fix is delivered. IBM publishes resolved APARs on the IBM Support website, so that other
users who experience the same problem can benefit from the same resolution.

148 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Notices
This information was developed for products and services offered in the U.S.A. IBM may not offer the
products, services, or features discussed in this document in other countries. Consult your local IBM
representative for information on the products and services currently available in your area. Any
reference to an IBM product, program, or service is not intended to state or imply that only that IBM
product, program, or service may be used. Any functionally equivalent product, program, or service that
does not infringe any IBM intellectual property right may be used instead. However, it is the user's
responsibility to evaluate and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this
document. The furnishing of this document does not give you any license to these patents. You can send
license inquiries, in writing, to:

IBM Director of Licensing

IBM Corporation
North Castle Drive
Armonk, NY 10504-1785 U.S.A.

For license inquiries regarding double-byte (DBCS) information, contact the IBM Intellectual Property
Department in your country or send inquiries, in writing, to:

IBM World Trade Asia Corporation

Licensing
2-31 Roppongi 3-chome, Minato-ku
Tokyo 106, Japan

The following paragraph does not apply to the United Kingdom or any other country where such
provisions are inconsistent with local law:
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS"
WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A
PARTICULAR PURPOSE.
Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore,
this statement might not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically
made to the information herein; these changes will be incorporated in new editions of the publication.
IBM may make improvements and changes in the product(s) and/or the program(s) described in this
publication at any time without notice.
Any references in this information to non-IBM websites are provided for convenience only and do not in
any manner serve as an endorsement of those websites. The materials at those websites are not part of
the materials for this IBM product and use of those websites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes appropriate without
incurring any obligation to you.
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the
exchange of information between independently created programs and other programs (including this
one) and (ii) the mutual use of the information which has been exchanged, should contact:

IBM Corporation
2Z4A/101
11400 Burnet Road
Austin, TX 78758 U.S.A.

© Copyright IBM Corp. 1994, 2017 149

 Such information may be available, subject to appropriate terms and conditions, including in some cases
payment of a fee.
The licensed program described in this document and all licensed material available for it are provided by
IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement or any
equivalent agreement between us.
Any performance data contained herein was determined in a controlled environment. Therefore, the
results obtained in other operating environments may vary significantly. Some measurements may have
been made on development-level systems and there is no guarantee that these measurements will be the
same on generally available systems. Furthermore, some measurement may have been estimated
through extrapolation. Actual results may vary. Users of this document should verify the applicable data
for their specific environment.
Information concerning non-IBM products was obtained from the suppliers of those products, their
published announcements or other publicly available sources. IBM has not tested those products and
cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM
products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of
those products.
All statements regarding IBM's future direction or intent are subject to change or withdrawal without
notice, and represent goals and objectives only.
All IBM prices shown are IBM's suggested retail prices, are current and are subject to change without
notice. Dealer prices may vary.
This information is for planning purposes only. The information herein is subject to change before the
products described become available.
This information contains examples of data and reports used in daily business operations. To illustrate
them as completely as possible, the examples include the names of individuals, companies, brands, and
products. All of these names are fictitious and any similarity to the names and addresses used by an
actual business enterprise is entirely coincidental.
COPYRIGHT LICENSE:
This information contains sample application programs in source language, which illustrate programming
techniques on various operating platforms. You may copy, modify, and distribute these sample programs
in any form without payment to IBM, for the purposes of developing, using, marketing or distributing
application programs conforming to the application programming interface for the operating platform for
which the sample programs are written. These examples have not been thoroughly tested under all
conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these
programs. You may copy, modify, and distribute these sample programs in any form without payment to
IBM for the purposes of developing, using, marketing, or distributing application programs conforming to
IBM‘s application programming interfaces.
If you are viewing this information in softcopy form, the photographs and color illustrations might not
display.

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be
trademarks of IBM or other companies. For a current list of IBM trademarks, refer to the Copyright and
trademark information at https://www.ibm.com/legal/copytrade.

150 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Bibliography

IBM Z Performance and Capacity Analytics publications

The IBM Z Performance and Capacity Analytics library contains the following publications and related
documents.
The publications are available online in the IBM Knowledge Center at the following link, from where you
can also download the associated PDF:
https://www.ibm.com/support/knowledgecenter/SSPNK7_3.1.0
• Administration Guide and Reference, SC28-3211
Provides information about initializing the IBM Z Performance and Capacity Analytics database and
customizing and administering IBM Z Performance and Capacity Analytics.
• Capacity Planning Guide and Reference, SC28-3213
Provides information about the capacity planning, forecasting, and modeling feature of IBM Z
Performance and Capacity Analytics, intended for those who are responsible for monitoring system
capacity and key performance metrics to help ensure that sufficient resources are available to run the
business and meet expected service levels.
• CICS Performance Feature Guide and Reference, SC28-3214
Provides information for administrators and users about collecting and reporting performance data
generated by Customer Information Control System (CICS®).
• Distributed Systems Performance Feature Guide and Reference, SC28-3215
Provides information for administrators and users about collecting and reporting performance data
generated by operating systems and applications running on a workstation.
• Guide to Reporting, SC28-3216
Provides information for users who display existing reports, for users who create and modify reports,
and for administrators who control reporting dialog default functions and capabilities.
• IBM i System Performance Feature Guide and Reference, SC28-3212
Provides information for administrators and users about collecting and reporting performance data
generated by IBM i systems.
• IMS Performance Feature Guide and Reference, SC28-3217
Provides information for administrators and users about collecting and reporting performance data
generated by Information Management System (IMS).
• Language Guide and Reference, SC28-3218
Provides information for administrators, performance analysts, and programmers who are responsible
for maintaining system log data and reports.
• Messages and Problem Determination, GC28-3219
Provides information to help operators and system programmers understand, interpret, and respond to
IBM Z Performance and Capacity Analytics messages and codes.
• Network Performance Feature Installation and Administration, SC28-3221
Provides information for network analysts or programmers who are responsible for setting up the
network reporting environment.
• Network Performance Feature Reference, SC28-3222
Provides reference information for network analysts or programmers who use the Network Performance
Feature.

© Copyright IBM Corp. 1994, 2017 151

 • Network Performance Feature Reports, SC28-3223
Provides information for network analysts or programmers who use the Network Performance Feature
reports.
• Resource Accounting for z/OS, SC28-3224
Provides information for users who want to use IBM Z Performance and Capacity Analytics to collect
and report performance data generated by Resource Accounting.
• System Performance Feature Guide, SC28-3225
Provides information for performance analysts and system programmers who are responsible for
meeting the service-level objectives established in your organization.
• System Performance Feature Reference Volume I, SC28-3226
Provides information for administrators and users with a variety of backgrounds who want to use IBM Z
Performance and Capacity Analytics to analyze z/OS, z/VM®, zLinux, and their subsystems, performance
data.
• System Performance Feature Reference Volume II, SC28-3227
Provides information for administrators and users with a variety of backgrounds who want to use IBM Z
Performance and Capacity Analytics to analyze z/OS, z/VM, zLinux, and their subsystems, performance
data.
• Usage and Accounting Collector User Guide, SC28-3228
Provides information about the functions and features of the Usage and Accounting Collector.

152 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Glossary
A
administration
A IBM Z Performance and Capacity Analytics task that includes maintaining the database, updating
environment information, and ensuring the accuracy of data collected.
administration dialog
A set of host windows used to administer IBM Z Performance and Capacity Analytics.
C
collect
A process used by IBM Z Performance and Capacity Analytics to read data from input log data sets,
interpret records in the data set, and store the data in Db2 tables in the IBM Z Performance and
Capacity Analytics database.
component
An optionally installable part of a IBM Z Performance and Capacity Analytics feature. Specifically in
IBM Z Performance and Capacity Analytics , a component refers to a logical group of objects used to
collect log data from a specific source, to update the IBM Z Performance and Capacity Analytics
database using that data, and to create reports from data in the database.
control table
A predefined IBM Z Performance and Capacity Analytics table that controls results returned by some
log collector functions.
D
data table
A IBM Z Performance and Capacity Analytics table that contains performance data used to create
reports.
E
environment information
All of the information that is added to the log data to create reports. This information can include data
such as performance groups, shift periods, installation definitions, and so on.
I
IBM Z Performance and Capacity Analytics database
A set of Db2 tables that includes data tables, lookup tables, system tables, and control tables.
IMS
See “Information Management System” on page 153.
Information Management System
A transactional and hierarchical database management system for critical on-line operational and on-
demand business applications and data, enabling information integration, management, and
scalability.
internal data type
A data type used within IBM Z Performance and Capacity Analytics during the processing of data.
K
key columns
The columns of a Db2 table that together constitute the key.
key value
Value that is used to sort records into groups.
L

© Copyright IBM Corp. 1994, 2017 153

 log collector
A IBM Z Performance and Capacity Analytics program that processes log data sets and provides other
TIBM Z Performance and Capacity Analytics services.
log collector language
IBM Z Performance and Capacity Analytics statements used to supply definitions to and invoke
services of the log collector.
log data set
Any sequential data set that is used as input to IBM Z Performance and Capacity Analytics.
log definition
The description of a log data set processed by the log collector.
log procedure
A program module that is used to process all record types in certain log data sets.
lookup expression
An expression that specifies how a value is obtained from a lookup table.
lookup table
A IBM Z Performance and Capacity Analytics Db2 table that contains grouping, translation, or
substitution information.
P
PCB
See Program Control Block.
Program Control Block
A control block that contains pointers to IMS databases.
purge condition
Instruction for purging old data from the IBM Z Performance and Capacity Analytics database.
Program Specification Block
A control block that identifies the destinations and databases used by the application program. A PSB
consists of one or more program communication blocks (PCBs).
PSB
See Program Specification Block.
R
record definition
The description of a record type contained in the log data sets used by IBM Z Performance and
Capacity Analytics, including detailed record layout and data formats.
record procedure
A program module that is called to process some types of log records.
record type
The classification of records in a log data set.
repeated section
A section of a record that occurs more than once, with each occurrence adjacent to the previous one.
report definition language
IBM Z Performance and Capacity Analyticsstatements used to define reports and report groups.
report group
A collection of IBM Z Performance and Capacity Analytics reports that can be referred to by a single
name.
reporting dialog
A set of host or workstation windows used to request reports.
resource group
A collection of network resources that are identified as belonging to a particular department or
division. Resources are organized into groups to reflect the structure of an organization.

154 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
resource information
Environment information that describes the elements in a network.
S
section
A structure within a record that contains one or more fields and may contain other sections.
source
In an update definition, the record or Db2 table that contains the data used to update a IBM Z
Performance and Capacity Analytics Db2 table.
sysplex
A set of systems communicating and cooperating with each other, through certain multisystem
hardware components and software services, in order to process customer workloads.
system table
A Db2 table that stores information that controls log collector processing, TIBM Z Performance and
Capacity Analytics dialogs, and reporting.
T
target
In an update definition, the Db2 table in which IBM Z Performance and Capacity Analytics stores data
from the source record or table.
threshold
The maximum or minimum acceptable level of usage. Usage measurements are compared with
threshold levels.
U
update definition
Instructions for entering data into Db2 tables from records of different types or from other Db2 tables.
V
view
An alternative representation of data from one or more tables. A view can include all or some of the
columns contained in the table on which it is defined.

Glossary 155
156 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
Index

A CSQ data tables

IMS_PSB_ACCOUNT_H, _D, _W 60
accessing extracted information 145 IMS_PSB_ACCOUNT2_H, _D, _W, _M 65
ACCOUNT parameter 25 IMS_SYSTEM_TRAN_H,_D 70
additional capabilities 37 IMS_SYSTEM_TRAN2_H,_D, _M 73
administering IMS Performance Feature 21 IMS_TRAN_QUEUE_Q, _D 82
AOI user exit initiated transaction special log procedure case KPM_IMS_ATF_EVT_T 90
10 KPM_IMS_ATF_T 89
APAR (Authorized Program Analysis Report) 147 KPM_IMS_TRAN_H_ D_W 85
application reports CSQ feature, administering 21
response time overview graphical 107–109, 113, CSQ lookup table
115–117, 119, 121 IMS_AVAILABILITY_T 57
archive exit for IMS 141 CSQ records
attributes of reports 98 mapping to table fields 93
Authorized Program Analysis Report (APAR) 147 CSQ views
availability for IMS resources 6 IMS_TRAN_QUEUE_QV, _DV 84
Average Enqueues by Transaction by Hour reports CSQ_Vnnn_COLLECL log definition 39
KPM IMS 130 CSQ_Vnnn_COLLECT log definition 39
customer support
contacting IBM Support 147
B customization reference 139
books for IBM Z Performance and Capacity Analytics 151
buffer pool reports D
detail for OSAM/ISAM by date 105
detail for VSAM by date 106 data flow
Buffered Fast Path 50 through DRL2LOGP 18
through IMS CSQ light feature 19
through log collector 17
C data sets
changes in this edition xv input and output 29
chart report 99 data table
checkpoint EEQE 46 IMS_AVAILABILITY_D, _W 56
checkpoint facility data table reference 39
recovery using 37 data tables
choosing between standard and extended subcomponents 5 IMS Performance Feature 55
collect IMS_CHKPT_IOSAM_T 58
recovering from abends during 37 IMS_CHKPT_POOLS_T 58
collect components 5 IMS_CHKPT_REGION_T 59
collecting IMS log data 3 IMS_CHKPT_STATS_T 59
collection IMS_CHKPT_VSAM_T 59
preparing IMS log data for 1 IMS_HALDB_OLR_T,_H, _D, _W 59
setting IMS ATF summary record type 1 IMS_PSB_ACCOUNT_H, _D, _W 60
comparison of performance programs 51 IMS_PSB_ACCOUNT2_H, _D, _W, _M 65
components IMS_SYSTEM_TRAN_H,_D 70
collect 5 IMS_SYSTEM_TRAN2_H,_D, _M 73
components, standard and extended IMS_TRAN_QUEUE_Q, _D 82
choosing between 5 KPM_IMS_ATF_EVT_T 90
composite records and subtypes 9 KPM_IMS_ATF_T 89
control tables KPM_IMS_TRAN_H_D, _W 85
DAY_OF_WEEK 55 KPM_IMS_TRAN_H, _D, _W 77
PERIOD_PLAN 55 data tables and lookup tables 54
controlling the log merge 143 DAY_OF_WEEK control table 55
conversational transaction special log procedure case 10 DB Activity by Region by Program Name by Hour report
CSQ KPM IMS 131
reports samples 100 Db2 tables
CSQ data table built by IMS Performance Feature 54
IMS_AVAILABILITY_D, _W data table 56 DDNAMEs for DRL2LOGP

Index 157
DDNAMEs for DRL2LOGP (continued) FLUSHTYPE parameter 25
DRLDUMP 22 format of reports 97
DRLICHKI 22 front end switching special log procedure case 11
DRLICHKO 22
DRLICOMP 30
DRLIPARM 22
G
DRLIRPT0n 30 glossary 153
DRLIRPTn 30 graphical report 99
DRLOUT 22
dependency
release 11 H
descriptions in tables 55
HALDB OLR
descriptions of record definitions 39
considerations when updating tables 35
detail report 99
data tables 59
DFSLTMG0
handling of special IMS cases within the IMS CSQ 9
log merge utility 141
documentation
IBM Z Performance and Capacity Analytics 151 I
DRL2LOGP
data flow 18 IBM Knowledge Center
data format 21, 29 publications xiii
DDNAME 21 IBM Z Performance and Capacity Analytics IMS
DDNAMEs 29 Performance Feature components
description 21 selecting 4
running 30 ID for report 97
using 29 IEC1301 issued for SQNLOGS+1 34, 141
DRLDUMP DDNAME 22 ILOGREC sample JCL 139
DRLICHKI DDNAME 22 IMS
DRLICHKO DDNAME 22 log record DSECTs 139
DRLICOMP IMS ATF summary record
DDNAME 30 setting IMS ATF summary record type 1
DRLIPARM DDNAME 22 IMS Availability reports 107
DRLIRPT0n DDNAME 30 IMS CSQ
DRLIRPTn DDNAME 30 descriptions of record definitions 39
DRLJXIMS archive exit 141 IMS CSQ feature
DRLOUT object definitions 20
DDNAME 22 record definitions 39
log procedure report 12 using 8
DRLSLOGP IMS CSQ feature object definitions 20
parameter descriptions 32 IMS CSQ feature, administering 21
using on remote systems 34 IMS CSQ light feature data flow 19
DRLSLOGP data flow 18 IMS CSQ Resource Utilization, Daily Overview 123
DRLSLOGP parameter descriptions 32 IMS CSQ Utilization reports 123
DRLSLOGP parameters IMS CSQ Utilization Reports
specifying 31 IMS CSQ Resource Utilization, Daily Overview 123
DRLSLOGPs IMS detail reports 104
specifying parameters 31 IMS HALDB OLR reports 125
DSECT IMS HALDB OLR Reports
for log record 139 IMS HALDB OLR Statistics 126
IMS HALDB OLR Unit Of Reorganization, Details 125
IMS HALDB OLR Statistics 126
E IMS HALDB OLR Unit Of Reorganization, Details 125
IMS light feature
EEQE record 43
considerations when updating IMS_HALDB_OLR_x
example table contents
tables 35
lookup table IMS_AVAIL_RESOURCE 91
considerations when updating KPM_IMS_ATF_x tables
lookup table IMS_SYSTEM_NAMES 91
35
extended components 5
setting up a load library 32
setting up the collect 34
F using 32
IMS log data
Fast Path record definition 50 collecting 3
Fast Path special log procedure case 11 preparing for collection 1
FLUSHMSGS parameter 25 IMS logs

158 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
IMS logs (continued) KPM IMS ATF Transaction Database Activity report (continued)
shared-queue environment 21 KPM IMS 136
IMS logs in a shared-queue environment 21 KPM IMS ATF Transaction Events report
IMS Message Queue Pool Detal by Date report 104 KPM IMS 134
IMS Message Queue reports 113 KPM IMS ATF Transaction Statistics report
IMS Message Queue Utilization, Date report KPM IMS 132
TLS 113 KPM IMS Average Enqueues by Transaction by Hour report
IMS Performance Feature 130
introducing 1 KPM IMS DB Activity by Region by Program Name by Hour
log and record procedures 8 report 131
understanding data flow 17 KPM IMS Processing Times by PSB Name by Hour report 128
IMS Performance Feature data tables 55 KPM IMS Processing Times by Region Type by Hour report
IMS resources, availability 6 129
IMS_ATFsummarydata, considerations on the IMS light KPM IMS Processing Times by Transaction by Hour report
feature 35 127
IMS_AVAIL_RESOURCE KPM_IMPS_ATF_x
lookup table 91 considerations when updating tables 35
IMS_AVAILABILITY_D, _W data table 56 KPM_IMS_ATF_EVT_T 90
IMS_AVAILABILITY_T lookup table 57 KPM_IMS_ATF_T 89
IMS_CHKPT KPM_IMS_TRAN_H_D_W 85
data tables 58, 59 KPM_IMS_TRAN_H, _D, _W tables 77
IMS_CHKPT_IOSAM_T table 58
IMS_CHKPT_POOLS_T table 58
IMS_CHKPT_REGION_T table 59
L
IMS_CHKPT_STATS_T table 59 log and record definitions
IMS_CHKPT_VSAM_T table 59 IMS Performance Feature 39
IMS_HALDB_OLR_T,_H, _D, _W tables 59 log and record procedures
IMS_HALDB_OLR_x tables, considerations on the IMS light IMS Performance Feature 8
feature 35 log collector
IMS_PSB_ACCOUNT_H, _D, _W data tables 60 data flow 17
IMS_PSB_ACCOUNT_x (H, D, W) 94 running 23
IMS_PSB_ACCOUNT2_H, _D, _W, _M data tables 65 specifying parameters 24
IMS_SYSTEM_NAMES using 21
lookup table 91 log collector data flow 17
IMS_SYSTEM_TRAN_H,_D tables 70 log definition 39
IMS_SYSTEM_TRAN_x (H, D) 94 log merge
IMS_SYSTEM_TRAN2_H,_D, _M tables 73 controlling 143
IMS_TRAN_QUEUE_Q, _D tables 82 log merge utility
IMS_TRAN_QUEUE_QV, _DV views 84 DFSLTMG0 141
IMS_TRAN_QUEUE_x (Q, D) 93 log procedure
IMS_TRAN_x (H, D, W) 93 composite records and subtypes 9
IMS_Vnnn_SLDS log definition 39 description 8
input and output data sets 21, 29 DRLOUT reports 12
installation parameter descriptions 26
testing 8 special cases 9
ISAM/OSAM buffer pool detail by date 105 log procedure parameter descriptions 26
ISC special log procedure case 11 log procedures parameters
specifying 25
J log record DSECTs 139
lookup and control tables
JCL samples updating 7
ILOGREC 139 lookup table
IMS_AVAIL_RESOURCE 91
IMS_AVAILABILITY_T 57
K IMS_SYSTEM_NAMES 91
key columns lookup tables 91
TRANS_TYPE 81
Key Performance Metrics IMS reports 127 M
keys
recovery token 8 manuals
unit of work 8 IBM Z Performance and Capacity Analytics 151
Knowledge Center, IBM publications xiii mapping between table fields and CSQ records 93
KPM IMS ATF Transaction Database Activity report MAXFREE

Index 159
MAXFREE (continued) preparing IMS log data for collection 1
using to allow more pending data to be held 34 problem determination, IBM Support
MAXFREE parameter 25 determining business impact 147
message procedure
IEC1301 issued for SQNLOGS+1 34 record 15
message-driven BMP program special log procedure case 10 Processing Times by PSB Name by Hour reports
MSGTEXT for IMS_TRAN_x tables KPM IMS 128
accessing extracted information 145 Processing Times by Region Type by Hour reports
MSGTEXT for IMSPTRAN_x tables KPM IMS 129
customizing, example 143 Processing Times by Transaction by Hour report
MSGTEXT parameter 25 KPM IMS 127
multiple outputs special log procedure case 10 program schedule record definition 42
multiple segment input special log procedure case 9 program-to-program switch
multiple segment output special log procedure case 10 special log procedure case 10
multiple system coupling special log procedure case 11 publications
multiple transactions per schedule of PSB special log accessing online xiii
procedure case 10 IBM Knowledge Center xiii
IBM Z Performance and Capacity Analytics 151
N
Q
naming standard for tables 54
non-message-driven BMP program special log procedure quick reschedule special log procedure case 11
case 10
R
O
record definitions
object definitions descriptions of 39
IMS Performance Feature 20 IMS Performance Feature 39
OSAM/ISAM buffer pool detail by date report 105 record grouping
OTMATRANCODE parameter 25 OUOW group 9
output message reenqueue special log procedure case 10 PSB group 9
overview report 98 PUOW group 9
record procedure 15
recovering from abends during collect 37
P recovery token keys 8
parameter descriptions recovery using the log procedure checkpoint facility 37
log procedure 26 RECTYPE parameter 25
parameter report from log procedure 12 release dependency 11
parameters report
ACCOUNT 25 attribute 98
FLUSHMSGS 25 format 99
FLUSHTYPE 25 ID 97
MAXFREE 25 IMS Message Queue Pool Detal by Date report 104
MSGTEXT 25 title 97
OTMATRANCODE 25 type 98
PASSLOGREC 25 report format and general description 97
PHASE2ENDTIME 25 report title 97
RECTYPE 25 reports
SECONDARY 25 IMS Availability 107
SET07BUF 25 IMS CSQ Utilization 123
SQNLOGS 25 IMS detail 104
START 25 IMS Message Queue 113, 125
STATISTIC 26 Key Performance Metrics IMS 127
STOP 26 KPM IMS ATF Transaction Database Activity 136
TABLEFLUSH 26 KPM IMS ATF Transaction Events 134
VERIFY 26 KPM IMS ATF Transaction Statistics 132
WRITEPENDING 26 KPM IMS Average Enqueues by Transaction by Hour 130
PASSLOGREC parameter 25 KPM IMS DB Activity by Region by Program Name by
performance programs Hour 131
comparison 51 KPM IMS Processing Times by PSB Name by Hour 128
PERIOD_NAME key column 59 KPM IMS Processing Times by Region Type by Hour 129
PERIOD_PLAN control table 55 KPM IMS Processing Times by Transaction by Hour 127
PHASE2ENDTIME parameter 25 resources, availability 6

160 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
response time report tabular report 99
application overview graphical 107–109, 113, 115–117, tailoring example using MSGTEXT for IMS_TRAN_x tables
119, 121 143
running the log collector 23 terminal message switch special log procedure case 10
terms defined 153
TRANS_TYPE key column 81
S trend report definition 98
sample archive exit 141 type of report 98
samples of reports across non-CSQ amd CSQ tables 100
SECONDARY parameter 25 U
selecting IBM Z Performance and Capacity Analytics IMS
Performance Feature components 4 understanding data flow through the IMS Performance
SET07BUF parameter 25 Feature 17
setting IMS ATF summary record type 1 unit of work keys 8
setting up a load library for the IMS light feature 32 updating other lookup and control tables 7
setting up the IMS light feature collect 34 using DRLSLOGP on remote systems 34
severity using log and record procedures
contacting IBM Support 147 IMS CSQ feature 8
determining business impact 147 using MAXFREE to allow more pending data to be held 34
shared-queue environment using TABLEFLUSH to improve performance 36
IMS logs 21 using TABLEFLUSH to minimize virtual storage use 36
single segment input with operator logical Paging output using TABLEFLUSH to prevent performance degradation 35
special log procedure case 10 using TABLEFLUSH to prevent storage shortage failures 35
special IMS cases using the IMS Performance Feature 8
handling of 9 using the log collector 21
specifying DRLSLOGP parameters 31
specifying log procedure parameters 25
specifying the log collector parameters 24
V
SQNLOGS parameter 25 VERIFY parameter 26
standard components 5 views
standard report format 99 IMS_TRAN_QUEUE_QV, _DV 84
START parameter 25 Transaction Transit Time subcomponent 84
STATISTIC parameter 26 VSAM buffer pool detail by date report 106
STOP parameter 26
storage shortage failures
using TABLEFLUSH 35 W
summary of changes xv
wait for input program special log procedure case 11
Support
what's new in this edition xv
contacting IBM 147
worst case report 99
describing problems 147
WRITEPENDING parameter 26
determining business impact 147
submitting problems 147
system generated output special log procedure case 10

T
table
IMS_PSB_ACCOUNT_x (H, D, W) 94
IMS_SYSTEM_TRAN_x (H, D) 94
IMS_TRAN_QUEUE_x (Q, D) 93
IMS_TRAN_x (H, D, W) 93
table descriptions 55
table fields
mapping to CSQ records 93
TABLEFLUSH
using to improve performance 36
using to minimize virtual storage use 36
using to prevent performance degradation 35
using to prevent storage shortage failures 35
TABLEFLUSH parameter 26
tables
built by IMS Performance Feature 54
naming standard 54

Index 161
162 IBM Z Performance and Capacity Analytics : IMS Performance Feature Guide and Reference
IBM®

SC28-3217-00