100606

MAINVIEW for CICS
PERFORMANCE REPORTER User Guide
Supporting
Version 6.3.00 of MAINVIEW for CICS
August 2009
www.bmc.com
Contacting BMC Software

You can access the BMC Software website at http://www.bmc.com. From this website, you can obtain information about the company, its products, corporate offices, special events, and career opportunities.
United States and Canada

Address BMC SOFTWARE INC 2101 CITYWEST BLVD HOUSTON TX 77042-2827 USA Telephone 713 918 8800 or 800 841 2031 Fax 713 918 8000
Outside United States and Canada

Telephone (01) 713 918 8800 Fax (01) 713 918 8000
Copyright 1987 2009 BMC Software, Inc. BMC, BMC Software, and the BMC Software logo are the exclusive properties of BMC Software, Inc., are registered with the U.S. Patent and Trademark Office, and may be registered or pending registration in other countries. All other BMC trademarks, service marks, and logos may be registered or pending registration in the U.S. or in other countries. All other trademarks or registered trademarks are the property of their respective owners. CICS, DB2, IBM, MVS, OS/390, VTAM, and z/OS are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. BMC Software considers information included in this documentation to be proprietary and confidential. Your use of this information is subject to the terms and conditions of the applicable End User License Agreement for the product and the proprietary and restricted rights notices included in this documentation.
Restricted rights legend

U.S. Government Restricted Rights to Computer Software. UNPUBLISHED -- RIGHTS RESERVED UNDER THE COPYRIGHT LAWS OF THE UNITED STATES. Use, duplication, or disclosure of any data and computer software by the U.S. Government is subject to restrictions, as applicable, set forth in FAR Section 52.227-14, DFARS 252.227-7013, DFARS 252.227-7014, DFARS 252.227-7015, and DFARS 252.227-7025, as amended from time to time. Contractor/Manufacturer is BMC SOFTWARE INC, 2101 CITYWEST BLVD, HOUSTON TX 77042-2827, USA. Any contract notices should be sent to this address.
Customer support
You can obtain technical support by using the BMC Software Customer Support website or by contacting Customer Support by telephone or e-mail. To expedite your inquiry, see Before contacting BMC.
Support website
You can obtain technical support from BMC 24 hours a day, 7 days a week at http://www.bmc.com/support_home. From this website, you can
s s s s s s s s
read overviews about support services and programs that BMC offers find the most current information about BMC products search a database for issues similar to yours and possible solutions order or download product documentation download products and maintenance report an issue or ask a question subscribe to receive proactive e-mail alerts when new product notices are released find worldwide BMC support center locations and contact information, including e-mail addresses, fax numbers, and telephone numbers
Support by telephone or e-mail

In the United States and Canada, if you need technical support and do not have access to the web, call 800 537 1813 or send an e-mail message to [email protected]. (In the subject line, enter SupID:<yourSupportContractID>, such as SupID:12345). Outside the United States and Canada, contact your local support center for assistance.
Before contacting BMC

Have the following information available so that Customer Support can begin working on your issue immediately:
s
product information product name product version (release number) license number and password (trial or permanent)
operating system and environment information machine type operating system type, version, and service pack or other maintenance level such as PUT or PTF system hardware configuration serial numbers related software (database, application, and communication) including type, version, and service pack or maintenance level
s s s
sequence of events leading to the issue commands and options that you used messages received (and the time and date that you received them) product error messages messages from the operating system, such as file system full messages from related software
License key and password information

If you have questions about your license key or password, contact BMC as follows:
s
(USA or Canada) Contact the Order Services Password Team at 800 841 2031, or send an e-mail message to [email protected]. (Europe, the Middle East, and Africa) Fax your questions to EMEA Contracts Administration at +31 20 354 8702, or send an e-mail message to [email protected]. (Asia-Pacific) Contact your BMC sales representative or your local BMC office.
MAINVIEW for CICS PERFORMANCE REPORTER User Guide
Contents
About this book 15 Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Online services documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Part 1
Introduction and report data

Chapter 1 Introduction to PERFORMANCE REPORTER
19
21 23 24 24 26 27 27 28 32 34 37 38 41 41 43 44 45 45 46 48 49 50 51 64 66 69
Running report programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . An overview of PERFORMANCE REPORTER reports . . . . . . . . . . . . . . . . . . . . . . . . . Predefined reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Custom PRL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 2 Collecting and archiving report data
Collecting report data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transaction detail data (CMRDETL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CICS operational statistics (CMRSTATS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Archiving data (CMRPURG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Purging summary records from the CMRSUMD data set . . . . . . . . . . . . . . . . . . . Converting CMRDETL records into CICS 110 SMF format . . . . . . . . . . . . . . . . . . . . . Chapter 3 Managing report data
Creating statistical records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SYSPRINT error messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Console error messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reformatting compressed detail transaction records . . . . . . . . . . . . . . . . . . . . . . . . . . . CMRCMPW and CMRCMPWN batch decompression programs. . . . . . . . . . . . . CMRCMPR and CMRCMPRN decompression programs . . . . . . . . . . . . . . . . . . . Creating reports from summarized data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summarization utility JCL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summarization utility control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Running the summarization utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Creating PRL reports from summary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notes on using the summarization utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents
Part 2
MAINVIEW for CICS reports

Chapter 4 Resource Tracking and Analysis report
77
79
Creating a Resource Tracking and Analysis report . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Sample CMRRAPM macro assembly statements . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Report control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Report structure and conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Field descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Chapter 5 Graphic histogram reports 91
Creating a histogram report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Keyword=value statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 CLASS and DATA control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 END statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Sample reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Response time for all transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 CPU used by CICS execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 CICS paging rate/second . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 CICS transaction rate/second . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Chapter 6 Service Level Analysis report 111
Creating a Service Level Analysis report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Field descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Part 3
COBOL and SAS reports

Chapter 7 COBOL reports
115
117
Accessing detail transaction records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 CMRFRED and CMRFREDN programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 CMRCOBCP and CMRCOBCPN (COBOL copybook) programs . . . . . . . . . . . . 120 Customizing sample COBOL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Copying and modifying the program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Compiling and linking the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Running the program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 Sample programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 CMRCOBJ2sample execution JCL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 COBOL report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Performance Analysis for All Transactions report (CMRCOB11 and CMRCOB11N) . 128 Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Performance Analysis by Program report (CMRCOB12 and CMRCOB12N) . . . . . 129
Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by Terminal report (CMRCOB13 and CMRCOB13N) . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by Netname report (CMRCOB14 and CMRCOB14N) . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis for All Users report (CMRCOB15 and CMRCOB15N) . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . File Analysis for All Files report (CMRCOB21 and CMRCOB21N) . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rogues report (CMRCOB23 and CMRCOB23N) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abended Transaction report (CMRCOB25 and CMRCOB25N) . . . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 8 SAS reports
130 130 131 131 132 132 133 133 134 135 135 136 136 137 138 140 140 141 142 142 143 143 144 144 145 145 146 146 147 147 148 148 149 149 150 150
Sample SAS programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SAS report fields. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis for All Transactions report (CMRSAS11 and CMRSAS11N) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by Program report (CMRSAS12 and CMRSAS12N) . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by Terminal report (CMRSAS13 and CMRSAS13N) . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by Node Name report (CMRSAS14 and CMRSAS14N) . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Performance Analysis by User ID report (CMRSAS15 and CMRSAS15N) . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . File Analysis for All Files report (CMRSAS21 and CMRSAS21N) . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rogues report (CMRSAS23 and CMRSAS23N) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abended Transaction Roster report (CMRSAS25 and CMRSAS25N) . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Operational statistics report (CMRSAS6F and CMRSAS6FN) . . . . . . . . . . . . . . . . . . Sample report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Column descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Part 4
Performance Reporting Language reports

Chapter 9 User-defined reports (CMRPRL and CMRPRLN)
151
153 154 155 156 156 158
Performance Reporting Language (PRL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Expressions and operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Substrings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Symbolic variables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents
Predefined variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Generating PRL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Job control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Basic PRL verbs used to create a report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 PRL processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Logic flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 Record processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 Example: Reporting average terminal response time. . . . . . . . . . . . . . . . . . . . . . . 165 Chapter 10 PRL control statement verbs 169
PRL control statement syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 PRL reserved words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 COMPUTE (calculate mathematical expression). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 DEFINE (define variable attributesoptional) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 DETAIL (tabulate and summarize each record) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 DIAGNOSE (perform diagnostics) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 END (stop execution) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 FOR and NEXT (execute a set of statements on a condition). . . . . . . . . . . . . . . . . . . . 180 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 GOTO (continue processing at label with : suffix) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 IF and THEN (do statement to test conditions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 LET (set variable length) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 ORDER BY (sort selected records) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Syntax. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
8 MAINVIEW for CICS PERFORMANCE REPORTER User Guide
Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PRINT (print one-line message). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REPORT (report selected record). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SCAN (scan a variable field) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SELECT (find record type) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SET (assign a value) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUMMARY (total and average selected records) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . USING (designate column headings and record fields) . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHERE (qualify record selection). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 11 PRL programming techniques
186 187 187 187 188 188 188 188 188 189 189 190 191 191 191 191 192 195 195 196 196 196 197 197 198 198 199 199 199 201 203 203 204 205 205 205 206 207 207 208 210 210 211 213 217 218 219 220
9
Routing report output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Defining output with JCL and PRL SET REPORTID . . . . . . . . . . . . . . . . . . . . . . . Defining output with PRL SET CLASS and SET FORM . . . . . . . . . . . . . . . . . . . . Formatting report output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Predefined data dictionary descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . User-defined data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selecting archive data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Expediting record retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indexing detail (CMRDETL) records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indexing by qualified variable name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Producing graphs with PRL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Time versus activity graph. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Service level analysis graph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summarizing data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sorting by more than one field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reporting 4GL database activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Creating longest transaction response time reports . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents
Index
223
10
Figures
Report creation process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Typical batch job to create a custom report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 MAINVIEW for CICS transaction detail records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 CICS statistical records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Sample JCL to run to merge, archive, and purge CMRDETL records . . . . . . . . . . . . 34 Sample recorder file purge report listings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 JCL for CICS statistical reports (CMRSTJCL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 JCL for CMRCMPWN batch decompression program . . . . . . . . . . . . . . . . . . . . . . . . . 46 Assembler routine to dynamically call CMRCMPRN . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Summarizing CMRDETL records with the CMRSUMD utility . . . . . . . . . . . . . . . . . . 49 Sample JCL for creating summary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Recommended control statement order in CMRSUMD JCL . . . . . . . . . . . . . . . . . . . . 53 Summarization utility JCL and control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Summarization utility workload summary report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 PRL report JCL and control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 PRL report using summary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Selecting records for summarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 HISTORY Service Display of summary records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Resource Tracking and Analysis reportinput JCL . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 CMRRAPM macro assembly statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Transaction Resource Tracking and Analysis report . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Example JCL to create histogram reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Response time for all transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 CPU used by CICS execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 CICS paging rate/second (page-in + page-out) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 CICS transaction rate/second . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Service Level Analysis reportinput JCL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Service Level Analysis report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 JCL Execution statements for all COBOL reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Performance Analysis for All Transactions report . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 Performance Analysis by Program report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Performance Analysis by Terminal report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Performance Analysis by Netname report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 Performance Analysis for All Users report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 File Analysis for All Files report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 COBOL rogues report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Abended Transaction report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Performance Analysis for All Transactions report . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Performance Analysis by Program report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Performance Analysis by Terminal report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
Figures 11
Performance Analysis by VTAM Symbolic Node Name report . . . . . . . . . . . . . . . . . 145 Performance Analysis by User ID report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 File Analysis for All Files report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 SAS rogues report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 Abended Transaction Roster report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Operational statistics report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 JCL used for PRL batch execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 Example PRL verbs for creating a report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Detail file control table analysissample PRL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 Sample PRL using the COMPUTE statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 Terminal-attached transaction response time report . . . . . . . . . . . . . . . . . . . . . . . . . 167 Detail report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Summary report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 DD statement change to point to a tape data set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Qualified variable names PRL statement example . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Time versus activity PRL statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Time versus activity within one day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 Time versus activity over several months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Service level analysis PRL statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Service level analysis output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 SUMMARY PRL statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Transaction Analysis for All Transactions summary report . . . . . . . . . . . . . . . . . . . 218 Set variables to sort by more than one field PRL statement . . . . . . . . . . . . . . . . . . . . 219 NATURAL performance PRL statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Longest transaction response time PRL statements . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Longest transaction response times report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
12
Tables
CMRSTATS control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Commonly used control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 CMRSUMD workload summary report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Changes to CMRDETL record fields after summarization . . . . . . . . . . . . . . . . . . . . . . 73 Processing summary record fields during summarization . . . . . . . . . . . . . . . . . . . . . 74 Required JCL for Resource Tracking and Analysis report . . . . . . . . . . . . . . . . . . . . . . 80 Description of the CMRRAPM macro assembly statements . . . . . . . . . . . . . . . . . . . . 81 Report control statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Histogram control statement set structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Sample histogram reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Service Level Analysis report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Sample COBOL programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 Sample SAS programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 SAS report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 Operational statistics report fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 PRL statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 PRL samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 PRL verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 PRL control statement verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 T6E and T6D fields for indexing detail records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Tables
13
14
About this book

PERFORMANCE REPORTER is the MAINVIEW for CICS batch report facility. This book describes how to submit predefined PERFORMANCE REPORTER batch programs to produce reports about IBM CICS performance and resource usage. This book also explains how to use the PERFORMANCE REPORTER Performance Reporting Language (PRL) to create your own custom reports. This book contains examples of the reports you can produce with the PERFORMANCE REPORTER batch programs. A description of the data shown in the report accompanies each example. This book is intended for anyone who needs to produce printed reports about the performance of their sites CICS systems. Readers are expected to understand CICS concepts; they should be able to write JCL jobs that produce printed reports. This book is a companion to the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. The Data Reference lists sample members in the BBSAMP data set and the data fields of MAINVIEW for CICS records. Like most BMC documentation, this book is available in printed and online formats. To request additional printed books or to view online books and notices (such as release notes and technical bulletins), see the Customer Support website at http://www.bmc.com/support_home. Some product shipments also include the online books on a documentation CD.
NOTE
Online books are formatted as PDF or HTML files. To view, print, or copy PDF books, use the free Adobe Reader from Adobe Systems. If your product installation does not install the reader, you can obtain the reader at http://www.adobe.com.
The software also offers online Help. To access Help, press F1 within any product or click the Help button in graphical user interfaces (GUIs).
About this book
15
Related publications
The following tables describes the related publications (including this book) for MAINVIEW for CICS:
Category installation documents Document OS/390 and z/OS Installer Guide

Description provides information about the installation of BMC products on IBM OS/390 and z/OS systems provides recommendations for the installation, implementation, and maintenance of MAINVIEW products explains how to create a MAINVIEW event management strategy and includes information about identifying metrics; defining alarm and workload definitions; creating rules for automation; and managing applications provides instructions for manually customizing the MAINVIEW environment for your products provides procedures for preparing the MAINVIEW subsystems to support MAINVIEW for CICS describes the basics of how to define security for MAINVIEW products with an external security manager (ESM) provides complete information about the MAINVIEW security interfaces for windows mode, full-screen mode, and MAINVIEW Alternate Access provides information about configuring and administering the Runtime Component System (RTCS) explains how to configure, start, and stop IBM VTAM and EXCP AutoLogon sessions to access MAINVIEW products without TSO communication support
MAINVIEW Best Practices: Installation, Implementation, and Maintenance MAINVIEW Best Practices: Event Management
MAINVIEW Common Customization Guide
MAINVIEW for CICS Customization Guide
MAINVIEW Security Guide
MAINVIEW Security Reference Manual
MAINVIEW Runtime Component System Configuration and Administration Guide MAINVIEW Alternate Access Implementation and User Guide
16
Category core documents
Document MAINVIEW for CICS Getting Started
Description provides an introduction to the product and offers exercises to help you get started with the online services describes how to monitor and manage CICS regions with the online services This book is intended for anyone responsible for maintaining adequate CICS performance at their site.
MAINVIEW for CICS Online Services Reference Manual
describes how to produce a variety of batch reports, including

s s
standard CICS performance and resource reports custom reports written with the Performance Reporting Language (PRL)
MAINVIEW for CICS PERFORMANCE REPORTER Data Reference MAINVIEW for CICS Monitors Guide MAINVIEW Administration Guide
describes the data fields of MAINVIEW for CICS records and lists sample members in the BBSAMP data set describes how to use the data collection monitors provides information about MAINVIEW operations, targets, single system image contexts, data sets, and diagnostic facilities provides information about working with MAINVIEW products in windows mode, in full-screen mode, and from MAINVIEW Explorer explains how to create and install alarm definitions that indicate when exceptions occur in a sysplex provide information about enhancements, fixes, and new features of the product provides a self-directed multimedia application of interactive tutorials, exercises, and tips
MAINVIEW User Guide
MAINVIEW Alarm Management Guide
notices and other documents
MAINVIEW for CICS release notes, flashes, and technical bulletins MAINVIEW for CICS Interactive Guide
About this book
17
Online services documentation
Online services documentation

The MAINVIEW for CICS Help system consists of a tutorial, specific service help, and an online message library.
s
An online tutorial provides an introduction to general services that are available to MAINVIEW for CICS with the MAINVIEW subsystem. The tutorial is listed as an option on the Primary Option Menu. All of the MAINVIEW for CICS services and views include online Help. The online help extends also to the fields within a view or service. By placing the cursor directly over a field within a view and pressing the HELP key, you can get a functional description of the field. MAINVIEW messages are documented in the Messages and Codes online display, which you can access by typing MSG on the COMMAND line of any MAINVIEW display.
Conventions
This book uses the following special conventions:
s
All syntax, operating system terms, and literal examples are presented in this typeface. Variable text in path names, system messages, or syntax is displayed in italic text:
testsys/instance/fileName
The symbol => connects items in a menu sequence. For example, Actions => Create Test instructs you to choose the Create Test command from the Actions menu.
18
Part
Part 1

This part provides an overview of the MAINVIEW for CICS PERFORMANCE REPORTER batch report facility and explains how to manage the input data used to create reports. This part presents the following topics: Chapter 1, Introduction to PERFORMANCE REPORTER Chapter 2, Collecting and archiving report data Chapter 3, Managing report data
Part 1
19
20
Chapter
Introduction to PERFORMANCE REPORTER

1
PERFORMANCE REPORTER is the MAINVIEW for CICS batch report facility. This chapter gives a short introduction to PERFORMANCE REPORTER and the various types of reports you can create. Figure 1 on page 22 shows the process used to create reports with PERFORMANCE REPORTER. CICS transaction and statistical data are gathered by data collectors and stored on specialized VSAM data sets. Frequently, long-term reports are created from data that has been archived to tape.
21
Figure 1
Report creation process
Reports are created by running standard JCL batch programs. Samples of these batch programs are located in your sites BBSAMP data set. As a user, you edit copies of these samples and modify them to your sites conditions. Normally, control statements are added to a report job. Some statements select the data from the VSAM data sets that you want to appear in a report. Other control statements specify how a report should be formatted.
22
Running report programs
Figure 1 on page 22 shows the various types of reports you can create with the sample programs provided by PERFORMANCE REPORTER. The following parts of this book give examples from each report type and the control statements that are needed to produce specific reports.
Running report programs

PERFORMANCE REPORTER provides the capability to create predefined or custom reports. A predefined report has a fixed format that shows values that apply to a specific range of CICS performance parameters. Your BBSAMP data set contains sample jobs to create predefined CICS, COBOL, and SAS statistical reports. In most cases, you merely specify the input data set and several control statements in your JCL to produce predefined reports. The control statements select the input data and specify whether the calculations produce a summary or detail report. Other times, you may want to write programs to create custom reports. You use PERFORMANCE REPORTERs English-like Performance Reporting Language (PRL) to create custom reports. Figure 2 shows an example of a PRL program to create a custom report that summarizes transaction response times by CICS region. Figure 2 Typical batch job to create a custom report
After modifying the CMRPRL batch job to your sites conditions, you include short PRL routines to select data and the format of your report.
23
An overview of PERFORMANCE REPORTER reports
An overview of PERFORMANCE REPORTER reports

This book describes the reports you can create with PERFORMANCE REPORTER, organized into two basic groups: predefined reports and custom reports.
Predefined reports
As shown in Figure 1 on page 22, PERFORMANCE REPORTER includes sample predefined programs to create CICS, COBOL, and SAS statistical reports. Each of these groups includes sample programs to create functionally similar reports. Typically, predefined report programs include short-term and long-term reports.
Short-term reports
Short-term reports show resource usage for the current or immediately preceding CICS execution, as follows:
s
The Destination Control Table Report provides either detailed or summarized usage of intra- and extrapartition transient data. The File Control Table Report provides either detailed file usage at specific times of the day or total file activity.
NOTE
The File Control Table Report includes only active table entries.
The Journal Control Table Report provides either detailed journal and journal buffer usage at specific times of the day or total journal activity. The Processing Program Table Report provides table-entry statistics that can start with the program name, the highest fetch count, or the highest use count.
NOTE
The Processing Program Table Report includes only active table entries.
24
Predefined reports
The Program Control Table Report provides either detailed transaction usage at specific times of the day or total activity for each transaction.
NOTE
The Program Control Table Report includes only active table entries.
The Temporary Storage Table Report provides the usage of temporary storage and the auxiliary temporary storage data set. The Terminal Control Table Report provides either terminal activity at specific times of the day or total terminal activity.
NOTE
The Terminal Control Table Report includes only active table entries.
The Transaction Class Usage Report provides class usage within the CICS region
NOTE
Short-term reports are generated by using PRL input with distributed sample report statements.
Long-term reports
Long-term report programs report about tabulated transaction or program resource usage, either individually or grouped as application systems or as histograms of CICS performance. These reports are as follows:
s
The Resource Analysis Report tabulates resource usage and run-time statistics on transactions or programs individually or as a grouped application. The control statements submitted with the batch job control language determine time and date ranges, report input, chronological ordering of the data, and the report output type (a summarized or detailed report or a report of only the resources not meeting service levels). The Graphic Histogram Report produces histograms of selected CICS performance parameters or resource usage. The control statements submitted with the batch job define the CICS performance parameters that appear in a report. The Service Level Analysis Report shows the distribution of transactions into service level ranges.
25
Custom PRL reports
Custom PRL reports

With PRL control statements, you can create custom reports from data stored on the CMRDETL and CMRSTATS data sets. Chapter 10, PRL control statement verbs, describes the usage of PRL programs to produce short-term batch reports for CICS users. Using control statements, you specify and format the following information:
s
type of data selected from the records stored on CMRDETL or CMRSTATS data sets date and time range of selected records interval counts for each data item or summarized totals you have specified type of CMRDETL or CMRSTATS record processing resulting from a PRL computation or condition report titles and column headings organization of the report
26
Chapter
Collecting and archiving report data

PERFORMANCE REPORTER batch reports are created from data gathered by the MAINVIEW for CICS data collectors. Depending upon the sampling parameters specified when MAINVIEW for CICS was customized, CICS data is collected over periodic intervals from monitored targets and then written as records to VSAM databases. This chapter gives a description of the various data records that are used to create batch reports. An outline of the various steps you must complete to prepare this data is also part of the discussion. Chapter 3, Managing report data, gives a more complete description of some of the procedures introduced in this chapter. This chapter concludes with a section that describes the procedure to archive and purge records from data sets.
Collecting report data

CICS transaction data collected by MAINVIEW for CICS is written as records to CMRDETL data sets. CICS statistical data collected is written as SMF 110 statistic records to SMF data sets. The SMF 110 statistic records can be formatted as CMRSTATS records by using batch program CMRSTATS. PERFORMANCE REPORTER uses data dictionaries (CMRDETL and CMRSTATS) that are located in the BBPLIB data set to reformat the stored records for reports. The MAINVIEW for CICS Customization Guide describes how to allocate and initialize the CMRDETL data sets. The MAINVIEW for CICS PERFORMANCE REPORTER Data Reference describes the format of each type of record that can be stored in a CMRDETL data set, as well as the records that can be stored on a CMRSTATS data set.
Chapter 2
27
Transaction detail data (CMRDETL)

A detail record is written to the CMRDETL data set for every transaction that occurs in a CICS region monitored by MAINVIEW for CICS. This record contains a history of the transactions life span. CICS Transaction Server version 3.2 expanded all performance record timing fields from 8 to 12 bytes. The main field that is expanded is the actual time field. The old time field was 4 bytes and contained a value expressed in 16-microsecond units (providing a capacity of about 19 hours). The new timing field is 8 bytes and contains a value expressed in single microsecond units (providing a capacity of several years). The reason for this expansion is to provide greater granularity for CPU and suspension times resulting from the speed of today's processors. As a result, MAINVIEW for CICS 6.1 constructed a new T6E history (CMRDETL) record to support the CICS/TS 3.2 modifications. MAINVIEW for CICS carries two formats of its transaction history record on the same CMRDETL data set:
s
F6 is for CICS/TS 3.1 and earlier and is formatted by the BBSAMP member FACTRECM when decompressed F7 is for CICS/TS 3.2 and is formatted by the BBSAMP member CMRQT6E when decompressed
MAINVIEW for CICS processes both record formats as necessary based on the release of the CICS and no user modifications are necessary. MAINVIEW for CICS provides internal compression and external decompression of these records based on programs being executed. These decompression programs can decompress F7 records to the F6 level and vice versa. MAINVIEW for CICS knows the CICS release that is being processed and constructs the T6E history record as necessary. When the history record is decompressed before being written to the CMRDETL data set, the proper compression routine is called to compress the record. For record processing through the history views and displays, both record formats are decompressed and built to the format desired by the presentation facilities. You can install MAINVIEW for CICS 6.2 without making any modifications to online facilities or batch utilities. All existing views and displays and all batch reporting and summarization utilities operate as in previous releases. When a CICS/TS 3.2 target is made available to MAINVIEW for CICS 6.2, the proper decompression utility prepares the history record. Since the existing supplied and user-written batch reporting and summarization utilities in COBOL, SAS, or PRL currently process the F6 level of the history record, the existing decompression routinesCMRCMPR and CMRMCPWdetect an F7 level record and decompress it to the F6 format. No user modificatons are necessary to use these facilities to process CICS/TS 3.2 performance data.
28
However, to take advantage of the greater granularity provided by the enlarged time fields, modifications are required to all user-written batch reporting programs (COBOL, PRL, SAS and assembler). MAINVIEW for CICS provides a new pair of decompression programs, CMRCMPRN and CMRCMPWN, that decompress T6E history records to the F7 level. These programs decompress all T6E history records (F6 or F7) to the F7 level. These new routines are automatically called based on the programs you execute in the batch reporting and summarization JCL. You must alter your programs, using the appropriate dictionaries, copybooks and/or macros, then execute the proper programs in order for MAINVIEW for CICS to call the proper decompression routine. For example, program CMRPRL will decompress T6E records to the F6 level using the CMRDETL dictionary, but program CMRPRLN will decompress T6E records to the F7 level using the CMRDETLN dictionary. The following sections explains both the existing and new processing available for the PRL, COBOL and SAS batch reporting facilities as well as the summarization utilities.
Note on a decompression discrepancy

When F7 records are converted to the F6 format, the time fields are built by taking the middle 4 bytes from the 8-byte time field and saving them in the F6 record. Because of this, a processing discrepancy can exist when converting F7 format records to the F6 format because of the loss of granularity. This discrepancy exists when the time fields in the individual resource entries are totaled and compared to a single total time provided in the performance record field. Here is an example using the file resource entries and the total TMRFCWT (FCIOWTT) time from a CICS/TS 3.2 target and converting the F7 record to the F6 format:
FC wait time in F7 format File1: File2: Total: 0000 0000 05F7 F87E 0000 0000 0552 30BB 0000 0000 0B4A 2939 FC wait time in decompressed F6 format 0000 05F7 0000 0552 0000 0B49 0000 0B4A
TMRFCWT: 0000 0000 0B4A 2939
As you can see for the F7 format, the total of the two file-wait times match the actual time provided in the performance record. However, since the F6 record is constructed by using just the middle 4 bytes of the F7 field, the total of the two file entries does not match the total of the middle 4 bytes from the performance record field. BMC recommends that you convert programs as soon as practical after migrating to CICS/TS 3.2. Figure 3 on page 30 gives an overview of the various steps to prepare detail transaction records for reports.
Chapter 2
29
Figure 3
MAINVIEW for CICS transaction detail records
30
Two types of detail transaction records exist. A 6E record is written for every transaction that completes successfully. The CMRDETL data set also contains 6D abended transaction records. They are kept separate from 6E records because their data can be unreliable due to the abended transaction. Batch jobs that produce the reports select data from the CMRDETL data set by the record ID (the hexadecimal code 6E or 6D) located in the TYPE field of the record. The following example shows a PRL control statement to select a 6E record as input data for a report
SELECT TYPE 6E The MAINVIEW for CICS PERFORMANCE REPORTER Data Reference gives a complete description of every field of 6E and 6D transaction records. This section includes two tables that list the fields of the record by offset order and by function. Before 6E or 6D records are written to the CMRDETL data set, data for application trace (if any), online graphs, and workload services are extracted from it and stored in buffer pools. You can view this data with the MAINVIEW for CICS online trace, graph, and workload services. After that, the 6E or 6D records are compressed and written to a CMRDETL data set. You can view the data through MAINVIEW for CICS history views and full screen displays You can use single or dual CMRDETL data sets. Dual data sets allow you to monitor almost continuously without having to stop and archive the data set when it becomes full. CMRDETL data set switching allows archives to occur without interrupting the data collection process. The MAINVIEW for CICS Customization Guide describes how to use the control statements located in member CMRDTL00 of your BBPARM data set to specify the switching conditions for dual CMRDETL data sets. PERFORMANCE REPORTER batch reports can be prepared from compressed detail transaction records held in the CMRDETL data set or a tape archive. If you are going to use this data to prepare SAS reports or other third-party vendor reports, the records must be decompressed before running the batch report programs. Sample members CMRCJCL and CMRSAS10 located in your BBSAMP data set can be edited to run batch jobs that decompress 6E or 6D transaction records. CMRSUMD is a utility that summarizes selected detail transaction records. You can summarize transaction activity over periods from minutes to days to create uncompressed summary records stored to tape or a VSAM database. In turn, these summary records can be used to create specialized summary reports. If you store summary records to a VSAM data set, you can display them online with the MAINVIEW for CICS HISTORY service. Creating reports from summarized data on page 49 describes how to create summary records from detail transaction records.
Chapter 2
31
CICS operational statistics (CMRSTATS)

MAINVIEW for CICS collects CICS statistics according to parameters specified in the CMRSOPT startup table. CICS statistical records are first written to the system management facility (SMF) data set with CICS SMF 110 records. Statistical records can then be reformatted to CMRSTATS records by the CMRSTATS batch routine. Figure 4 on page 33 shows how CICS statistical records are processed for batch reports.
32
Figure 4
CICS statistical records
Chapter 2
33
Archiving data (CMRPURG)
Predefined statistical reports select data from the CMRSTATS data set. If you want to create CICS SMF or other third-party vendor reports from data held in the SMF data set, you must first filter out MAINVIEW for CICS statistical records. A sample member named CMRSTATX located in your BBSAMP data set removes MAINVIEW for CICS records, leaving SMF 110 records for other reports. All statistical records contain a field in the header area that identifies the record type. Records are selected from the CMRSTATS data set by the two-digit hexadecimal code entered into the TYPE field of the record. Global Performance records (Type CC) are written to the data set at five-minute intervals. Other types of records are written according to the startup parameters specified in the CMRSOPT table.

CMRPURG is a program that purges and archives data held in CMRDETL data sets. If specified, CMRPURG can also archive and purge selected data to a tape data set and merge it with other archive tapes. Refer to the MAINVIEW for CICS Customization Guide for information about using CMRPURG to archive CMRDETL records in a single or dual data set environment. The execution JCL for CMRPURG is shown in Figure 5 on page 34. Descriptions corresponding to the numbered statements follow the JCL.
NOTE
Sample JCL for the CMRPURG program is in the CMRPURGE member of your BBSAMP data set.
Figure 5
Sample JCL to run to merge, archive, and purge CMRDETL records

user parameters parameters DISP=SHR,DSN=CMR.CMRV5.BBLINK PGM=CMRPURG DISP=OLD,DSN=PREV.HISTORY.FILE DISP=(NEW,CATLG),DSN=CMR.MERGED.ARCHIVE,UNIT=TAPE DSN=CMR.CMRWRK1,SPACE=(CYL,(10,10)),UNIT=SYSDA DISP=(NEW,DELETE,CATLG) DISP=SHR,DSN=CMR.CMRV5.CMRDETL SYSOUT=* SYSOUT=* *
//jobname JOB /*JOBPARM user //JOBLIB DD //ARCHIVE EXEC //TAPEIN DD //TAPEOUT DD //CMRWRK1 DD // //CMRDETL DD //REPORT DD //SYSPRINT DD //SYSIN DD DATE=mmddyyyy HISTORY=YES
(1) (2) (3) (4)
(5) (6)
The numbered items in the following list explain the JCL statements of the example CMRPURG job shown in Figure 5:
1 identifies a data set that is to be merged and archived with the purged CMRDETL
records.
2 identifies the output tape data set for the purged CMRDETL records. The default
DCB is DSORG=PS,RECFM=VB,LRECL=12256,BLKSIZE=12260.
3 (specifies a data set, CMRWRK1, to hold all records that are not archived during
the reorganization phase. The recorder data sets also must be defined with REUSE so that reorganization can occur. The records can be written to either tape or disk.
4 identifies the CMRDETL recorder data sets to be processed. Archiving each data
set to a separate tape as separate jobs increases access to statistical data because there is less volume. More days can then be kept online without requiring a purge.
5 specifies the date for purging. This statement must start in column 1. All data
stamped with this or a previous date is purged and, if requested, archived to the output tape data set. DATE=12319999 purges all the data from the recorder data set.
NOTE
If 12319999 is specified, the CMRWRK1 data set is not required.
The date for purging can also be specified as a number of days preceding today. The format is DATE=*-nn where nn is a two-digit number of days before today. Note that the date calculated can represent a day in the previous year. The actual date calculated by the program is displayed in an informational message on the report output.
6 specifies the type of history tape manipulation to be performed. This statement

must start in column 1. Valid values are as follows: NO specifies that no input or output history tapes are to be used. All data that has a date earlier than the DATE value will be purged. OUT specifies that no input tape is to be used. Only an output tape is to be created containing the data to be purged. YES specifies that both input and output tapes are to be used. The data on the input tape is to be merged with the data to be purged from the MAINVIEW for CICS recorder data sets and written to the output tape.
7 LEAVE= specifies whether to leave data in CMRDETL or purge it after archival.

This statement must start in column 1. Valid values are as follows:
Chapter 2 Collecting and archiving report data 35
NO specifies that data is removed from CMRDETL. The default is NO (it is not specified in the sample). NO is required for single CMRDETL environment. It is not recommended in dual data set environment because data will not be available for online access after the switch. YES specifies that data remains in CMRDETL after archival. This value is recommended only for dual CMRDETL data set environment. It provides HISTORY access to the data after a data set switch occurs.
Archive-purge process
The CMRPURG job performs the following operations: 1. The job reads data from the recorder data sets, merges it with another archive data set (HISTORY=YES), and then writes it to the history output tape (HISTORY=YES or HISTORY=OUT). If HISTORY=NO is specified, the data is deleted and not archived. 2. The job purges data from the recorder data sets.
NOTE
Purging is performed here so that abends (such as tape errors) do not leave the recorder data sets in a state of reduced integrity.
The data that is purged is determined by the DATE= control statement. The entire data set can be purged by specifying DATE=12319999, or a portion of the recorder data set can be purged as follows: A. If only part of a data set is to be purged, you can define the VSAM REUSE parameter when the MAINVIEW for CICS recorder data set is defined, as described in the MAINVIEW for CICS Customization Guide, and allocate a temporary data set, as shown by CMRWRK1 in the previous sample JCL. CMRPURG uses CMRWRK1 to contain temporarily the records that are to remain on the recorder data set. REUSE reorganizes and resets the CMRDETL recorder data sets so that records written to CMRWRK1 from these data sets can be reloaded. B. CMRPURG processes the CMRDETL recorder data set. The records stamped with the DATE value or a date previous to the specified date are written as specified by the HISTORY= control statement. C. The selected records are deleted from CMRDETL if NOREUSE is specified, which can be a time-consuming update to the CMRDETL data set.
36
Purging summary records from the CMRSUMD data set
If an abend occurs while archiving data, check the messages on the report output to determine at what point it occurred. CMRPURG does not write the same record twice if it exists on both the history and the recorder data sets. Each phase of the archive-purge process is recorded in the Recorder File Purge Report. Figure 6 shows a typical listing for CMRDETL recorder data sets. Figure 6 Sample recorder file purge report listings
Report Listing for CMRDETL Recorder File C I C S M A N A G E R R E C O R D E R F I L E P U R G E
FT290I-DETAIL TRANSACTION RECORDER FILE BEING PROCESSED FT293I-BEGINNING ARCHIVE OF REQUESTED RECORDS FT294I-ARCHIVE OF REQUESTED RECORDS COMPLETED FT295I-BEGINNING PURGE OF REQUESTED RECORDS FT296I-PURGE OF REQUESTED RECORDS COMPLETED FT297I-CMRWRK1 DATASET FOUND - BEGINNING REORGANIZATION FT298I-ENTERING SECOND PHASE OF REORGANIZATION FT301I-REORGANIZATION OF RECORDER FILE COMPLETED ------ RECORDER FILE -----INPUT PURGED 000001667 000001666 -- HISTORY FILE ACTIVITY -INPUT OUTPUT 000000000 000001666
Archive-purge completion
When the archive-purge process is complete, activate data collection from CICS with an SMN2 transaction or from an IBM MVS console using FST2. This action reopens the CMRDETL data set.
Purging summary records from the CMRSUMD data set

The CMRSUMD utility summarizes performance data over selected intervals and places uncompressed summary records in an output VSAM data set. You can create batch reports from the summary records or view the records online with the HISTORY service. CMRPURG can purge selected summary records from the CMRSUMD data set. This purging allows you to keep some summary records required for long-term reports and remove others that are no longer needed. The JCL shown in Figure 5 on page 34 can be used to run the CMRPURG program. An additional WORKLOAD control statement must be added to the job stream beneath the SYSIN DD statement. The WORKLOAD control statement selects which summary records are purged.
Chapter 2 Collecting and archiving report data 37
Converting CMRDETL records into CICS 110 SMF format
The WORKLOAD control statement has the following positional parameters: WORKLOAD = wkld, group, time where
wkld is the name of the workload used to summarize records All records are purged by default. group is the name of the group of CICS regions combined to create a summary workload All groups are purged by default. The group name is specified with the SGROUP control statement of the CMRSUMD utility. time is the summarization interval of summary records All summary intervals are purged by default.
The following examples show how to use the WORKLOAD control statement with the CMRPURG program to purge summary records from the CMRSUMD data set:
EXAMPLE
WORKLOAD=IBMTRANS,*,1440
In the preceding example, daily summary records of all groups of the IBMTRANS workload are purged from a VSAM data set.
EXAMPLE
WORKLOAD=,60
In the preceding example, all hourly records from all workloads and groups are purged from a VSAM data set.

MAINVIEW for CICS enables you to convert transaction history records (CMRDETL) into CICS 110 SMF format. Any MAINVIEW for CICS CMRDETL record can be converted regardless of which release of CICS or MAINVIEW for CICS produced the record.
38
Support is provided to create the SMF 110 Performance records for CICS/TS 2.2 CICS/TS 2.3 CICS/TS 3.1 CICS/TS 3.2 CICS/TS 4.1 Support is provided by the following programs: CMRCMPW for CICS/TS 2.2 only support CMRCMPWN for CICS/TS 2.2 and later support Conversion is performed by the CMRCMPWN utility program, which was new with MAINVIEW for CICS 6.1. CMRCMPWN enables you to control the output format by using one of several input parameters. CMRCMPWN supports the input parameters that are supported by the older CMRCMPW program along with several other ones: FORMAT=2.3 or 2.3Y FORMAT=3.1 or 3.1Y FORMAT=3.2 or 3.2Y FORMAT=4.1 or 4.1Y The output format will be constructed regardless of the CICS release used. In other words, if the T6E input is from CICS/TS 2.2 and the user specifies FORMAT=3.1, the CICS/TS 3.1 SMF format data records are built. In this case, however, all CICS/TS 2.3 and CICS/TS 3.1 data fields are initialized to binary zero or blank as appropriate. The Y in the second format sets the date format to YYYYDDDF. The default format is CCYYDDDF. The converted data contains SMF formatted records: the SMF header followed by the SMF product section followed by the blocked data records. The data records are blocked to fit below 32K with each blocked area preceded by the dictionary field connectors. A CICS dictionary record is written that is equivalent to the complete dictionary for the CICS specified on the FORMAT= statement. This record is followed by the SMF transaction data record. It is followed by the MAINVIEW for CICS user counts (10) and clocks (4) followed by the MAINVIEW for CICS data area, followed by the MAINVIEW for CICS IBM DB2 data area. The latter two areas are defined in the SMF record as if the your site had specified an MCT, including the MAINVIEW for CICS MCT definitions in the BBSAMP members CMRNDMCT and CMRN2MCT, respectively.
Chapter 2
39
The SMF transaction data record format is described by the DFHMNTDS dsect starting at offset 8. The MAINVIEW for CICS user counts/clocks areas consist of 4 twelve-byte clocks followed by 10 four-byte counts. These areas contain some MAINVIEW for CICS calculated DBCTL information. The dictionary entries for these fields are identified by USER fields 001004 for the clocks and USER fields 001010 for the counts. No macro or copy book is available for the format of the USER fields. The clock fields are (refer to the DFSDSTA control block) as follows:
USER 001 USER 002 USER 003 USER 004 number of seconds of elapsed time for DBCTL requests (DBC$SCHT/16) delta between PSB schedule start and stop (DBC$DATNDBC$DATS) not used elapsed time for database I/O (DBC$TMIO/16)
The count fields are as follows:

USER USER USER USER USER USER USER USER USER USER 001 002 003 004 005 006 007 008 009 010 database GU calls (DBC$GU1) database GN calls (DBC$GH) database GNP calls (DBC$GNP) database GHU calls (DBC$GHU) database GHN calls (DBC$GHN) database GHNP calls (DBC$GHNP) database ISRT calls (DFH$ISRT) database DLET calls (DFH$DLET) database REPL calls (DFH$REPL) total database calls (DBC$TOTC)
The MAINVIEW for CICS data area is formatted by the CMRNDATA assembler dsect or the CMRNDSAS SAS field identifications for CICS/TS 3.2 and later. The dictionary entry is identified by the name of CMRDATA. The MAINVIEW for CICS DB2 data area is formatted by the CMRNDB2 assembler dsect or the CMRN2SAS SAS field identifications for CICS/TS 3.2 and later. The dictionary entry is identified by the name of CMMRDB2.
40
Chapter
Managing report data

This chapter describes several procedures to prepare data that appears in batch reports. Separate procedures describe how to
s s s
create statistical records reformat compressed detail transaction records summarize detail transaction records
Creating statistical records

To create statistical records from data collected by a CICS region, you must run the CMRSTATS program. The CMRSTATS program reads data directly from an SMF data set or a sequential extract from an SMF data set. The program determines the type of SMF data set that is allocated to the job. Only one VSAM data set is read at a time, and the required SMF data set (SYS1.MAN1, SYS1.MAN2) must be specified. Sample JCL to run this program is provided in BBSAMP as member CMRSTJCL. You can use the sample JCL along with distributed sample report statements to generate CICS statistical reports. Figure 7 on page 42 shows the JCL for running the CMRSTATS program.
Chapter 3
41
Creating statistical records
Figure 7
JCL for CICS statistical reports (CMRSTJCL)
//* This sample JCL can be used to extract CMRSTATS statistical //* data from the SMF data produced by MAINVIEW for CICS. //* //* You must make some changes to this JCL. The lines that may //* require changes are indicated by a '<===' marker. //* //JOBLIB DD DSN=MAINVIEW.CMRxx.BBLINK, <=== change // DISP=SHR //* The following step (STATS) extracts the CMRSTATS data from //* the SMF file and places it into a file that can be used //* by the Performance Reporter Language (PRL), or other user //* written programs to produce reports on the performance of //* your CICS regions. //* //STATS EXEC PGM=CMRSTATS //SYSPRINT DD SYSOUT=* //SMFIN DD DSN=smf.input.data, <=== Input SMF data // DISP=SHR //CMRSTATS DD DSN=cmrstats.data.set, <=== Output CMRSTATS // DISP=(,CATLG,DELETE), <=== change // VOL=SER=VOLSER, <=== change // UNIT=SYSDA, <=== change // SPACE=(CYL,(15,15),RLSE), <=== change // DCB=BLKSIZE=32760 //SYSIN DD * CMRSTATS control statements should be inserted here. <=== change (See the Performance Reporter Guide for details.) <=== change //* //* The following step (PRL) uses the statistical data extracted //* by the first step (STATS) to provide a Performance Reporter //* Language report. (Service Level Analysis Report). //* //PRL EXEC PGM=CMRPRL //SYSPRINT DD SYSOUT=* //SYSUDUMP DD SYSOUT=* //SYSOUT DD SYSOUT=* //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,5) //SORTWK02 DD UNIT=SYSDA,SPACE=(CYL,5) //SORTWK03 DD UNIT=SYSDA,SPACE=(CYL,5) //ISYPROG DD SYSOUT=*, // DCB=(LRECL=80,BLKSIZE=3120,RECFM=FBA) //ISYSOUT DD SYSOUT=*, // DCB=(LRECL=80,BLKSIZE=3120,RECFM=FBA) //ISYPDUM DD SYSOUT=*, // DCB=(LRECL=80,BLKSIZE=3120,RECFM=FBA) //PNLLIB DD DSN=MAINVIEW.CMRxx.BBPLIB, <=== change // DISP=SHR //CMRSTATS DD DSN=cmrstats.data.set, <=== Input // DISP=SHR //SYSIN DD DSN=MAINVIEW.CMRxx.BBSAMP(sample), <=== change // DISP=SHR
42
Control statements
NOTE
You must change some of the JCL shown in Figure 7. Lines that might require changes are indicated by <=== markers. See Table 18 on page 155 for a list of sample programs in BBSAMP.
Control statements
The following rules apply to the control statements that you specify with CMRSTATS:
s
Each SYSIN statement contains only one control statement. The control statement can start anywhere from column 1 but cannot be continued on the next line. If you do not supply application names, all CICS data from the SMF data set is processed. You can select data from the SMF data set that was collected from a specific CICS region.
If both generic and specific application IDs are specified, the specific ID is used for selection. You can select only one application ID.
Chapter 3
43
SYSPRINT error messages
Valid control statements are as follows:

:
Table 1
CMRSTATS control statements

Description Include SMF data created after the specified time. Include SMF data created up to and including the specified time. Include SMF data produced from the specified date onward. Note: If you specify a date in the old format (mm/dd/yy), yy is interpreted as 19yy.
Control Statement FROM TIME=hh:mm:ss TO TIME=hh:mm:ss FROM DATE=mm/dd/yyyy
FROM DATE=*-n TO DATE=mm/dd/yyyy
Include the SMF data from today minus n days onward. Yesterday would be *-1. Include SMF data produced up to and including the specified date. Note: If you specify a date in the old format (mm/dd/yy), yy is interpreted as 19yy.
TO DATE=*-n GAPPLID=applid SAPPLID=applid APPLID=applid
Include SMF data up to and including today minus n days. Yesterday would be *-1. Include only SMF data for the specified generic application ID. Include only SMF data for the specified specific application ID. Synonym for SAPPLID. Note: Parameter errors cause the run to terminate with a return code of 8 and an appropriate message on SYSPRINT.
SYSPRINT error messages

SYSPRINT errors messages are accompanied by a return code of 8. Messages are as follows:
FT581I Previous parameter was invalid
Explanation: The parameter statement listed above the message on SYSPRINT is not valid.
FT582I SMFIN DD statement missing or invalid
Explanation: The OPEN for SMFIN failed. Generally, this error means the DD statement is missing.
44
Console error messages
FT583I CMRSTATS DD statement missing or invalid
Explanation: The OPEN for the CMRSTATS DD statement failed. Generally, this error means the DD statement is missing. However, it also could mean an invalid BLKSIZE parameter. BLKSIZE must be at least 6000.
FT584I Insufficient memory to run the program
Explanation: More memory is required to process the data that is used for a report. Either increase the region size or restrict the report to a specific range of data. The date or time ranges normally have little effect on the amount of required memory. Memory is allocated above the 16 MB line.
Console error messages

The following console error message results from a 2048 ABEND:
FT585I SYSPRINT DD statement missing or invalid.
Explanation: The OPEN for the SYSPRINT data set failed. Generally, this error means the SYSPRINT DD statement is missing.
Reformatting compressed detail transaction records

Type 6E and 6D records are compressed when they are stored on the CMRDETL data set. You must decompress these records before they can be read by programs other than PERFORMANCE REPORTER programs. The records can be decompressed into standard MAINVIEW for CICS data or a format compatible with CICS/MVS data or CICS/TS SMF 110 (CMP) data.
s
MAINVIEW for CICS offers four programs to decompress the CMRDETL data set records: CMRCMPW can be used in a separate batch job to decompress CMRDETL records to the F6 level CMRCMPWN can be used in a separate batch job to decompress CMRDETL records to the F7 level CMRCMPR can be called from within a user program to dynamically decompress CMRDETL records to the F6 level
Chapter 3
45
CMRCMPW and CMRCMPWN batch decompression programs
CMRCMPRN can be called from within a user program to dynamically decompress CMRDETL records to the F7 level
NOTE
Type 6E records produced by earlier versions of MAINVIEW for CICS are always decompressed to the record format supported by the current version. Any fields that are new to the current version are initialized to zeros or blanks.

Figure 8 shows a sample batch job to execute the CMRCMPWN program. The CMRCJCL job is in your BBSAMP library. See Reformatting compressed detail transaction records on page 45 for more information. Compressed Type 6E records can reside on a current, active CMRDETL data set or be archived to tape. You select either source with the appropriate DD statement and an INPUT= control statement (INPUT=VSAM or INPUT=TAPE). Figure 8 shows examples of both DD statements. Figure 8 JCL for CMRCMPWN batch decompression program
//S1 EXEC PGM=CMRCMPWN //STEPLIB DD DSN=CMR.CMRV3.BBLINK,DISP=SHR //CMRDETL DD DSN=CMR.CMRV3.CMRDETL,DISP=SHR <- IF INPUT=VSAM //*TAPEIN DD DSN=....ARCHIVE.DATA,DISP=SHR <- IF INPUT=TAPE //SYSUDUMP DD SYSOUT=* //OUTPUT DD DSN=CMR.PROCESS.DATA,DISP=(,CATLG),SPACE=(TRK,(15,15)), // UNIT=SYSDA //SYSIN DD * INPUT=VSAM FORMAT=CMR
Control statements follow the SYSIN DD statement in the job stream. Control statements are as follows:
s
The FORMAT statement specifies the format of the decompressed record. FORMAT=CMR builds decompressed detail records that match the MAINVIEW for CICS format described by FACTRECM for CMRCMPW CMRQT6E for CMRCMPWN
FORMAT=2.1 or 2.1Y builds CICS/MVS 2.1.2 SMF 110 records (valid for CMRCMPW only)
46
FORMAT=2.2 or 2.2Y builds CICS/TS 2.2 SMF 110 records (valid for CMRCMPW and CMRCMPWN) FORMAT=2.3 or 2.3Y builds CICS/TS 2.3 SMF 110 records (valid for CMRCMPWN only) FORMAT=3.1 or 3.1Y builds CICS/TS 3.1 SMF 110 records (valid for CMRCMPWN only) FORMAT=3.2 or 3.2Y builds CICS/TS 3.2 SMF 110 records (valid for CMRCMPWN only) FORMAT=4.1 or 4.1Y builds CICS/TS 4.1 SMF 110 records (valid for CMRCMPWN only) The date is formatted as ccyyddds where: cc yy ddd s is the 2-digit century value (01 for 2000) is the 2-digit year value is the 3-digit day value is a packed decimal sign value
If Y is specified. the date is formatted as yyyyddds where: yyyy ddd s

s
is the 4-digit year value is the 3-digit day value is a packed decimal sign value
The INPUT control statement specifies the type of input data set holding the compressed CMRDETL records. INPUT=VSAM specifies that compressed records are stored on the current, active CMRDETL data set. INPUT=TAPE specifies that compressed records are archived to a tape data set.
NEWAPPL specifies a 1- to 8-character alphanumeric name that identifies the APPLID placed in the SMF header portion of decompressed CMRDETL records. This value replaces existing APPLIDs of compressed input records. Decompressed records retain their existing APPLIDs if the NEWAPPL control statement is not specified. The SELECT control statement selects records to be decompressed by the name of the CICS region that produced the CMRDETL record. If the SELECT control statement is not specified, the first CICS ID encountered is used.
Chapter 3
47
CMRCMPR and CMRCMPRN decompression programs
SELECT=xxxxxxxx specifies a 1- to 8-character alphanumeric name of a CICS region. SELECT=* selects all input records.
CMRCMPR and CMRCMPRN decompression programs

The CMRCMPR and CMRCMPRN programs dynamically decompresses CMRDETL records. Instead of executing CMRCMPW in batch mode, the CMRCMPR program can be called from a user program with an assembler language routine, as shown in Figure 9. See Reformatting compressed detail transaction records on page 45 for more information. Figure 9
LA CALL LTR BNZ . . DS DC DC DC
Assembler routine to dynamically call CMRCMPRN
R1,PARMLIST CMRCMPRN R15,15 HANDLERR 0D A(DFLAG) A(RECORD) C'D'
Q. GOOD DECOMPRESS NO - GO HANDLE ERROR
PARMLIST PARM1 PARM2 DFLAG
Standard linkages include the following registers:

s
Register 1 contains the address of a parameter list that points to a D flag that signifies a decompress request points to the record to be decompressed The field must be large enough to contain the entire decompressed record. LLBB must be the first record field. It must be set to the current length of the record.
Register 13 contains the address of a standard register save area. Register 14 contains the return address. Register 15 contains the return code from the CMRCMPR program.
48
Creating reports from summarized data
Creating reports from summarized data

This section explains how to use PERFORMANCE REPORTERs CMRSUMD and CMRSUMDN utilities to summarize detail transaction records stored on your CMRDETL data set.
s
CMRSUMD decompresses, processes and creates all records at the F6 level formatted by BBSAMP member FACTRECM CMRSUMDN decompresses, processes and creates all records at the F7 level formatted by BBSAMP member CMRQT6E
Note that the summarized output from either utility can only be input to the same utility. The summarization batch utility program creates workloads from CICS performance records. You can summarize data collected from multiple CICS regions in intervals from 1 minute to 24 hours. The summary workloads produced by the summarization utility are compatible with the COBOL, SAS, and Performance Reporting Language (PRL) report programs of MAINVIEW for CICS. Figure 10 illustrates how CMRDETL records are summarized. The summarization utility processes unsummarized records held in a VSAM or sequential data set. Figure 10 Summarizing CMRDETL records with the CMRSUMD utility
CMRDETL Data Set
HISTORY service to view summarized records
OUTPUT Data Set
VSAM Or Sequential CMRSUMD Utility Or
VSAM
Sequential
After the records are summarized over the intervals you requested, the newly created summary workloads are stored on VSAM or sequential output data sets. Summary workloads can be created with data obtained from any supported version of CICS. If you store your summarized records on a VSAM data set, you can view them online with the MAINVIEW for CICS HISTORY service. Viewing summary records on page 71 describes the procedure to use the HISTORY service.
Chapter 3
49
Summarization utility JCL
Summarizing CMRDETL records gives you several benefits. One benefit is that fewer records are required to produce reports that span long periods of time. For example, producing a monthly report of average daily CPU usage requires only one record per day. Another benefit of using summarized records is that you can write simpler report programs. Fewer records must be specified for each report interval and the batch program to create the reports generally requires less processing time for each query. Summarized records are useful for identifying performance trends that become apparent only over long intervals. After you locate CICS problem areas with summarized records, you can use unsummarized data to create reports that give you the detail you need to analyze problems.
Summarization utility JCL

Figure 11 shows an example of JCL to execute the CMRSUMD or CMRSUMDN programs and summarize report data. Figure 11 Sample JCL for creating summary data
//S1 EXEC PGM=CMRSUMD //STEPLIB DD DSN=CMR.BBLINK,DISP=SHR //* //CMRDETL DD DSN=CMR.DETAIL.DATA,DISP=SHR //OUTPUT DD DSN=CMR.SUMMARY.DATA,DISP=SHR //COPYDD DD DSN=CMR.COPY.CMRDETL,DISP=SHR //REPORT DD SYSOUT=* //SYSIN DD *
Figure 11 contains the following JCL statements:

s
EXEC specifies the name of the program that generates the summarized detail data (PGM=CMRSUMD or PGM=CMRSUMDN). //STEPLIB defines the program library containing the MAINVIEW for CICS load modules. //CMRDETL defines the name of the input data set containing the detail records to be summarized. The detail records can be from a single CICS region or multiple CICS regions. For maximum performance, detail records should be sorted in date and time sequence. If necessary, you can use a sort utility to combine multiple detail files into date and time sequence; for example, you can specify SORT FIELDS=(9,20,CH,A) for VSAM SORT FIELDS=(13,20,CH,A) for sequential (to account for the RDW)
50
Summarization utility control statements
//OUTPUT defines the name of the summary output data set. You must specify the BLKSIZE parameter when creating a new sequential output data set. See BBSAMP member CMRDFSUM for guidelines about allocating a VSAM or sequential data set.
NOTE
When you run multiple executions of the CMRSUMD program against the same CICS region, workload, and time frame using a VSAM data set:
s s s
Workloads with the same name as existing records are combined with the records. New workloads are inserted. A maximum of 999 workloads are supported.
//COPYDD, which is optional, defines the name of the selected sequential output data set. See the COPY= parameter on page 62 for more information.
NOTE
When creating a copy data set, specify RECFM=U, LRECL=0, and BLKSIZE=32768 bytes.
//REPORT defines the SYSOUT data set containing the program statistics compiled by CMRSUMD. //SYSIN contains the control statements you must use to specify the input records and summarization options. See Control statement descriptions on page 53 for a description of each control statement.

The control statements described in this section run the summarization utility. They are coded after the SYSIN DD statement in your JCL. Each control statement uses parameters to specify the report data and summarization options. In the descriptions in this section, default parameters are underlined.
Coding control statements

The global control statement parameters affect all the workloads defined in the summarization job. Local parameters apply to only a specific workload. You can code summarization control statements in a predefined order that sets both global and local conditions for each of the workloads defined in the job. BMC recommends coding the control statements in the following order:
Chapter 3 Managing report data 51
1. global control statements Global control statements set conditions for the job itself and are not directly related to workloads. The global control statements are NEWAPPL, REPLACE, and COPYAPPL. 2. default control statements Default control statements set the default conditions for the workloads specified in the summarization job. A default control statement can be any control statement except one that is explicitly defined as global or workload-specific. 3. workload definition control statements The workload definition control statements define the workloads in the summarization job. They are coded immediately before the default override and the workload-specific control statements. Together, the following workload definition control statements specify the conditions for a specific workload in the summarization job: A. default override control statements The workload default override control statements define operating conditions for a specific workload. They may differ from those previously set by the default control statements. A default override control statement can be any control statement except one that is defined as global or workload-specific. B. workload-specific control statements The following workload-specific control statements specify the input data that is used by a specific workload in the summarization job: SKIPFLIO PROCFLIO EXCLxxxx INCLxxxx
52
Figure 12 shows the recommended sequence of control statements. These three sets of control statements are repeated for each workload in the summarization job stream. Figure 12 Recommended control statement order in CMRSUMD JCL
REPLACE=YES TYPE=NOTERM FILES=TIME MINUTES=15 WORKLOAD=PAYROLL INCLTRAN=PAYR MINUTES=5 EXCLUSER=DRP1CICS WORKLOAD=ACCOUNTS MINUTES=10,30 FILES=SUMMARY INCLTRAN=AA
Control statement descriptions

The summarization utility has 27 control statements. These control statements fall into four broad classes based upon how often you are expected to use them. You will use some control statements in almost every summarization job. Other control statements you will use often, but not for every job. A third class is composed of control statements that are used even less frequently. The last class contains control statements that are used rarely, only to set unique conditions for a summarization job. The following four sections discuss the summarization control statements, grouped together by frequency of use.
Chapter 3
53
Commonly used control statements

The following control statements in this class are used for almost every summarization job: Table 2 Commonly used control statements (part 1 of 3)
Description specifies the name assigned to the workload
Control statement WORKLOAD=name
The utility places the workload name in the T6EPGNM (program) field of the summarized record. If you do not explicitly specify a workload name, the default workload name is SYSTEM.
Note: Up to 999 workloads are supported for a specific VSAM data set. MINUTES= {1|5|10|15|30|60|1440} specifies the summarization interval You can specify multiple intervals (for example, MINUTES=5,30,60). The following summarization intervals are permitted:
s s s s s s s
1 specifies increments of 1 minute. 5 specifies increments of 5 minutes. 10 specifies increments of 10 minutes (the default). 15 specifies increments of 15 minutes. 30 specifies increments of 30 minutes. 60 specifies increments of 1 hour. 1440 or DAILY specifies increments of 1 day.
When the MINUTES control statement precedes the WORKLOAD statement, the summarization interval you specify is in effect for all workloads. To specify a unique summarization interval for each workload, place an appropriate MINUTES control statement after each WORKLOAD statement.
54
Table 2
Commonly used control statements (part 2 of 3)

Description specifies whether non-terminal tasks are included in the workload, as follows:
s s
Control statement TYPE= {ALL|NOTERM|TERM}
ALL specifies that all tasks are included in the workload. NOTERM specifies that only non-terminal tasks are included in the workload (the default). TERM specifies that only terminal tasks are included in the workload. Non-terminal tasks are excluded.
When the TYPE control statement precedes the WORKLOAD statement, the type of task you specify is in effect for all workloads. To specify a different type of task for each workload, place an appropriate TYPE control statement after each WORKLOAD statement.
Chapter 3
55
Table 2
Commonly used control statements (part 3 of 3)

Description specifies a list of CICS regions that are combined to create a workload The first name in the SGROUP list identifies the summarized records. Note: SELECT is an alias for SGROUP. That is, you can use SELECT=list to specify a list of CICS regions. Only one SGROUP and its identifier are used in a single execution; subsequent references to the same name use the first definition. For example, if you specify the following statements, you get the same region list for the second SGROUP as you do for the first SGROUP: SGROUP=AORS,CICS1,CICS2,CICS3 ... . and other statements ... SGROUP=AORS ... This situation simplifies the definition of CICS system groups. When the SGROUP control statement precedes the WORKLOAD statement, the list of CICS regions you specify is in effect for all workloads. To specify a different list of CICS regions for each workload, place an appropriate SGROUP control statement after each WORKLOAD statement. Note: It is not necessary to combine all the CICS regions data into a single execution to produce combined workloads. The summarization utility combines data from different executions for the same groups and workloads.
Control statement SGROUP=list
Frequently used control statements

The following control statements in this class are used often but not for every summarization job.
EXCLxxxx=list
Specifies a list of resources to exclude from the summary. The EXCL control statement can be repeated for as many resources as required. EXCLxxxx must be specified within a WORKLOAD.
56
The xxxx variable must be one of the following values:

s s s s s s s s s
PLAN (name of the DB2 plan associated with the transaction) 4GLS (name of the 4GL panel or program associated with the transaction) OPID (three-character OPID associated with the transaction) TERM (four-character CICS terminal ID associated with the transaction) TRAN (transaction ID associated with the transaction) PROG (program name associated with the transaction) NETN (up to an 8-character net name associated with the transaction) USER (up to an 8-character extended user ID associated with the transaction) FILE (name of a file referenced by the transaction)
INCLxxxx=list
Specifies a list of resources to include in the summary. The INCL control statement can be repeated for as many resources as required. INCLxxxx must be specified at the workload level. The xxxx variable must be selected from the following values:
s s s s
PLAN (name of the DB2 plan associated with the transaction) 4GLS (name of the 4GL panel or program associated with the transaction) OPID (three-character OPID associated with the transaction) TERM (four-character CICS terminal ID associated with the transaction)
NOTE
If you want to combine summarized records gathered in intervals of 10 minutes into hourly records, you must specify INCLTERM=10 to ensure records that overlap time frames are not combined to give invalid data.
s s s s s
TRAN (transaction ID associated with the transaction) PROG (program name associated with the transaction) NETN (up to an 8-character net name associated with the transaction) USER (up to an 8-character extended user ID associated with the transaction) FILE (name of a file referenced by the transaction)
NEWAPPL=name
Specifies a new application ID (APPLID) that is written in the key of the output records.
Chapter 3
57
NOTE
The NEWAPPL control statement is used when you want to view records online. The HISTORY service expects the key of the record to match the name specified in the target field of the display panel. SUMMDATA is the default key. Use NEWAPPL when you want to change the default key.
Less frequently used control statements

The following sections discuss the control statements in this class that are used occasionally.
REPLACE={NO | YES}
REPLACE specifies whether data collected during the execution of the job replaces any data from previous executions that have the same workloads This control statement applies to VSAM data sets only. You must use REPLACE to recreate workloads from detail records that have been previously processed. By default, the data is combined with existing data.
FILES={TIME | SUMMARY | COUNT}

FILES specifies the summarization option for the file entries in the records If you specify FILES prior to the first WORKLOAD control statement, it defines the default type of summarization for all workloads, as follows:
s
TIME (summarizes records by file entries, and the 255 files with the longest elapsed I/O time are written to the summarized file) TIME is the default. SUMMARY (summarizes all entries, and only the first 20 entries are written to the summarized file) COUNT (summarizes the file entries, and the 20 files with the largest number of I/Os are written to the summarized file).
DETAIL={NO | YES}
DETAIL specifies whether the summarization utility treats files as individual entities or as a single entity for each file type. If you specify DETAIL prior to the first WORKLOAD control statement, it defines the default value for all workloads.
s
NO summarizes files entries by a common resource, such as CICS, Software AG ADABAS, DBCTL, DB2, and so forth (the default).
58
YES specifies that you want each file entry processed separately.
NOTE
Use DETAIL when you want to determine the overall effect of a given area rather than the overall effect of specific files.
ZERODATA={NO | YES}
ZERODATA specifies whether intervals are written to the summarized file when no tasks are executed for a given workload; for example, when CICS is not active, no tasks are executed. If you specify ZERODATA before the first WORKLOAD control statement, it defines the default value for all workloads. Values are as follows:
s
NO specifies that the utility does not write intervals with zero tasks to the summarized file (the default). YES specifies that the utility writes intervals with zero tasks to the summarized file.
FROM DATE=mm/dd/yyyy
FROM DATE specifies the starting date of the selected records.
NOTE
If you specify a date in the old format (mm/dd/yy), yy is interpreted as 19yy.
You can specify the starting date using a relative format that is an offset from the current date. For example, to start the selected records with yesterdays date, specify
FROM DATE=*-1
If you specify FROM DATE before the first WORKLOAD control statement, it defines the default starting date for all workloads.
TO DATE=mm/dd/yyyy
TO DATE specifies the ending date of the selected records.
Chapter 3
59
NOTE
You can specify the ending date using a relative format that is an offset from the current date. For example, to end the selected records with yesterdays date, specify
TO DATE=*-1
If you specify TO DATE before the first WORKLOAD statement, the utility provides the default ending date for all workloads at the workload level.
FROM TIME=hh:mm:ss
FROM TIME specifies the starting time of the selected records. The format is hh:mm:ss. If you specify FROM TIME before the first WORKLOAD control statement, it defines the default starting time for all workloads.
TO TIME=hh:mm:ss
TO TIME specifies the ending time for selecting records. The format is hh:mm:ss. If you specify TO TIME before the first WORKLOAD control statement, it defines the default ending time for all workloads.
START TIME=hh:mm:ss
START TIME specifies the starting time for the selection interval. You can use this time to specify a period, such as the prime shift for a given workload.
STOP TIME=hh:mm:ss
STOP TIME specifies the ending time for the selection interval. STOP TIME can be used with TO DATE to specify an ending date and time. The following example shows START TIME, STOP TIME, FROM TIME, and TO TIME control statements together in a job stream.
60
FROM DATE=*-5 FROM TIME=12:00:00 TO DATE=*-1 TO TIME=12:00:00 WORKLOAD=TEST1 INCLTRAN=AA* START TIME=08:00:00 STOP TIME=17:00:00 WORKLOAD=TEST2 INCLTRAN=AA* FROM DATE=*-5 TO DATE=*-3
In this example, the default date and time ranges for the selected records are from noon five days ago through yesterday at noon. Workload TEST1 uses records that are collected between 8:00 A.M. and 5:00 P.M. On the first day, records are collected from 12:00 P.M. to 5:00 P.M. On the last day, records are collected from 8:00 A.M. to 12:00 P.M. Workload TEST2 collects records from noon four days ago through the following day at noon.
MINRESP=nnnnn.nnn
MINRESP sets a minimum response time threshold of a transaction record in seconds. Records with transaction response times that exceed the minimum threshold are included in the summarization workload. Records with response times less than the threshold are excluded.
MAXRESP=nnnnn.nnn
MAXRESP sets a maximum response time threshold of a transaction record in seconds. Records with response times less than the maximum threshold are included in the summarization workload. Records that exceed the maximum response time are excluded. The MINRESP and MAXRESP control statements are often used together to filter transaction records by response times. You can create summary workloads that contain only those transactions with response times that fall within the interval set by the two control statements.
MINCPU=nnnnn.nnn
MINCPU sets a minimum CPU usage threshold of a transaction record in seconds. Records with CPU usage that exceed the minimum threshold are included in the summarization workload. Records with CPU usage less than the threshold are excluded.
Chapter 3
61
MAXCPU=nnnnn.nnn
MAXCPU sets a maximum CPU usage threshold of a transaction record in seconds. Records with CPU usage less than the maximum threshold are included in the summarization workload. Records that exceed the maximum threshold are excluded. The MINCPU and MAXCPU control statements can be used together to filter transaction records by CPU usage. You can create summary workloads that contain only those transactions with CPU usage that falls within the interval set by the two control statements.
Rarely used control statements

The control statements in this class set unique conditions that are rarely used for most summarization jobs.
TABLE=name
TABLE specifies the name of a table created using the CMRRAPM macro, which specifies transactions and programs to be included in the workload. The TABLE control statement
s s
is provided for compatibility with the CMRRAPM table has limited flexibility
For example, default values are used for all control statements (other than WORKLOAD and INCLxxxx, which are contained within the CMRRAPM definitions).
COPYAPPL=name
COPYAPPL specifies an application ID (APPLID) that is written in the key of the records copied to the copy data set.
COPY={NO | YES}
COPY specifies whether the selected records are copied to the output file specified with the COPYDD DD statement. When COPY is specified before the first WORKLOAD control statement, it determines whether selected records are copied to the output file for all workloads.
62
Values are as follows:

s s
NO specifies that selected records are not copied. NO is the default. YES specifies that if a record is selected for summarization, it is also copied to the sequential data set specified with the COPYDD DD statement.
NOTE
These copied records are useful when performing detailed analysis of the data without reading the entire detail data set again.
SKIPFLIO=list
SKIPFLIO specifies a list of files that are excluded in the summarized file data. SKIPFLIO must be specified at the workload level.
NOTE
Use of SKIPFLIO does not affect the selection of records, as happens with the EXCLFILE control statement.
PROCFLIO=list
PROCFLIO specifies a list of files to include in the summarized file data. Other files are not summarized. Use PROCFLIO to insure that files you specify are reported for a workload. PROCFLIO must be specified at the workload level.
NOTE
Use of PROCFLIO does not affect the selection of records, as happens with the INCLFILE control statement.
USEREXIT=name
USEREXIT specifies the name of a user-written routine that is given control when a record is processed during the summarization job. The input record is passed to the exit. The record is tested against conditions set by the user-written routine. The user exit either includes or excludes the record in the summarization job based upon the outcome of the test conditions. BBSAMP member CMRUSRD is an example of this exit.
Chapter 3
63
Running the summarization utility

Figure 13 is an example of a job that runs the summarization utility and creates summary workloads. Figure 13 Summarization utility JCL and control statements
//USR1SUM JOB (3831),'USER NAME',CLASS=T,MSGCLASS=R, // NOTIFY=USR1,TIME=60 //* //S1 EXEC PGM=CMRSUMD,REGION=4096K //STEPLIB DD DSN=CMR34.CMR1.LOAD,DISP=SHR //CMRDETL DD DSN=CMR1.DETAIL.DATA,DISP=SHR //* //SYSUDUMP DD SYSOUT=* //OUTPUT DD DSN=CMR.SUMMARY.DATA,DISP=SHR //REPORT DD SYSOUT=* //SYSIN DD * * * SAMPLE CMRSUMD INPUT * REFER TO MEMBER CMRSUMD FOR EXECUTION JCL * THIS MEMBER CREATES DAILY SUMMARY RECORDS FOR 4 WORKLOADS * MAINVIEW for CICS TRANSACTIONS * CICS TRANSACTIONS * ALL OTHER TRANSACTIONS * ALL TERMINAL TRANSACTIONS * * SET DEFAULT FOR DAILY TIME PERIOD AND ALL TRANSACTION TYPES * MINUTES=1440 TYPE=ALL * * CREATE MAINVIEW for CICS WORKLOAD * WORKLOAD=CMRTRANS INCLTRAN=FST2,SMN2,FIC2,FCD2,JNL2 * * CREATE IBM TRANSACTION WORKLOAD * WORKLOAD=IBMTRANS INCLTRAN=C*,DSNC * * CREATE ALL TRANSACTION WORKLOAD (EXCEPT THOSE IN PREVIOUS) * WORKLOAD=ALLTRANS EXCLTRAN=FST2,SMN2,FIC2,FCD2,JNL2 EXCLTRAN=C*,DSNC * * CREATE ALL TERMINAL TRANSACTION WORKLOAD (OVERRIDE TRANSACTION TYPE) * WORKLOAD=ALLTERM TYPE=TERM
The control statements shown in Figure 13 are examples from member CMRSUMD1 of the BBSAMP data set. The job creates four different summary workloads each day. A brief description of each workload is given in the comments section of the job shown in Figure 13. A description of the summarized transaction records created for each workload is shown beneath the SYSIN DD statement.
64
The report shown in Figure 14 is produced by the utility. It shows the input CMRDETL records that were summarized across the four workloads defined in the job. Figure 14
WORKLOAD CMRTRANS IBMTRANS ALLTRANS ALLTERM GROUP ID# 1 2 3 4 MIN DLY DLY DLY DLY
Summarization utility workload summary report

INPUT ACCEPTED 1,583,507 2,346 1,583,507 935,254 1,583,507 645,907 1,583,507 1,471,378 OUTPUT 4 4 4 4 UPDATED MAX-FILES 4 4 4 6 4 6 4 6
Input records are summarized daily, and the OUTPUT column shows four records were created for each workload over the four-day interval.
Description of the report fields

The report fields shown in Figure 14 are as follows: Table 3
Field WORKLOAD GROUP ID # MIN INPUT ACCEPTED OUTPUT UPDATED MAX-FILES
CMRSUMD workload summary report fields

Description workload name assigned to the summarization summarization group of combined CICS regions internal workload ID assigned to the workload and used to make the key unique number of minutes in the summary number of records in the input file that meet the date, time, and CICS system criteria number of records in the input file that were accepted for summarization number of summarized records written number of summary records that were updated with input records from this job number of files accessed by the summarized workload
NOTE
When MAX-FILES is greater than 20, it indicates some files were not written in the summarized records.
Chapter 3
65
Creating PRL reports from summary data
Additional fields appear in the summary report if the REPLACE=YES or USEREXIT=name control statements are included in the job. If REPLACE=YES is included, two fields named DELETED and REPLACED appear in the report. The DELETED field shows the number of records that were deleted because the interval contained no records and ZERODATA=NO was specified. The REPLACED field shows the number of records held in the output data set that were replaced by new records. If USEREXIT=name is included in the job, two fields named INCLUDED and EXCLUDED show the number of records that were included or excluded from the summary workloads due to the test conditions set by the user-written routine. The INCLUDED field shows the number of records that were summarized that normally would have been excluded if the exclude decision had not been overridden by the user-written routine. The EXCLUDED field shows the number of records that were excluded from summarization that normally would have been included if the include decision had not been overridden by the user-written routine.

Figure 15 on page 67 shows the JCL and control statements to produce a Performance Reporting Language (PRL) report from the summary workloads created by the summarization utility. The JCL is from BBSAMP member CMRL, which is generic JCL to run PRL jobs. The JCL must be edited to your sites conditions before running the job. The PRL control statements added beneath the SYSIN DD statement are from member CMRPRL28 located in the BBSAMP data set. These control statements produce a PRL report from summarized 6E and 6D records. The seven control statements shown between the USING and REPORT statements specify the data extracted from the summary records that appears in the PRL report. Each of these seven control statements produces a separate, tabular column in the PRL report.
66
Figure 15
PRL report JCL and control statements
//USR1PRL JOB (NUMBER),'USER NAME', // CLASS=F, // MSGCLASS=R, // MSGLEVEL=(1,1), // NOTIFY=USR1 /*JOBPARM L=99,COPIES=1,R=10,S=SYSB //PRL EXEC PGM=CMRPRL,REGION=2048K,TIME=180 //* //* THIS IS SAMPLE JCL TO EXECUTE THE MAINVIEW for CICS PERFORMANCE //* REPORTING LANGUAGE (PRL). CHANGE THE STATEMENTS INDICATED //* TO REFLECT YOUR INSTALLATION. //* //STEPLIB DD DISP=SHR,DSN=CMR34.CMR1.LOAD //SYSPRINT DD SYSOUT=* //SYSUDUMP DD SYSOUT=* //SYSOUT DD SYSOUT=* //ISYPROG DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //ISYSOUT DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //CMRDETL DD DISP=SHR,DSN=CMR1.TEMP.CMRDETL1 //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,30) //SORTWK02 DD UNIT=SYSDA,SPACE=(CYL,30) //SORTWK03 DD UNIT=SYSDA,SPACE=(CYL,30) //SYSIN DD * * * THIS IS THE SAMPLE REPORT THAT USES THE SUMMARIZED DATA CREATED BY * CMRSUMD. * * (YOU MAY NEED TO MODIFY THE REPORT OUTPUT CLASS ON THE NEXT STATEMENT) SET CLASS = 'R' SET ZHEADING = 'SINGLE' * SET TITLE = 'SUMMARIZED DATA ANALYSIS' SELECT TYPE 6E AND 6D RECORDS FROM CMRDETL USING CMRDATE CMRTIME SUMWKLD SUMCICS T6ETASK T6ERESP T6ECPUR REPORT END
The online PRL report shown in Figure 16 on page 68 is produced by the job shown in Figure 15. The PRL report shows the task count, average response time, and CPU usage of the summarized workloads over four daily periods. Notice that the seven tabular columns of the report appear in the same order as the PRL control statements in the job to produce the report.
Chapter 3
67
Figure 16
RECORDED DATE ---------09/12/1998 09/12/1998 09/12/1998 09/12/1998 09/13/1998 09/13/1998 09/13/1998 09/13/1998 09/14/1998 09/14/1998 09/14/1998 09/14/1998 09/15/1998 09/15/1998 09/15/1998 09/15/1998 RECORDED TIME -------24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00 24:00:00
PRL report using summary data

CICS GROUP -------CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD CICSPROD COUNT OF TASKS IN SUMMARY --------------265 207831 9017 215355 866 614318 140551 739681 226 16229 38577 46405 989 96876 457762 469937 RESPONSE TIME ----------0.171 0.067 0.280 0.075 17.090 0.206 1.042 0.295 0.135 0.035 1.192 0.434 0.156 0.026 1.017 0.750 CPU REAL TIME ---------0.057 0.005 0.015 0.006 0.067 0.005 0.022 0.007 0.064 0.002 0.018 0.014 0.064 0.002 0.016 0.015
WORKLOAD NAME -------CMRTRANS IBMTRANS ALLTRANS ALLTERM CMRTRANS IBMTRANS ALLTRANS ALLTERM CMRTRANS IBMTRANS ALLTRANS ALLTERM CMRTRANS IBMTRANS ALLTRANS ALLTERM
Description of the report fields

The fields of the PRL report shown in Figure 16 are as follows:
Field Name RECORDED DATE RECORDED TIME WORKLOAD NAME CICS GROUP COUNT OF TASKS IN SUMMARY RESPONSE TIME CPU REAL TIME Description ending date of the summarization interval in mm/dd/yyyy format ending time of the summarization interval in hh:mm:ss format name of the workloads with summarized data CICS region from which summary records were created number of tasks summarized for this program average internal response time for the summarized tasks average real CPU time for the summarized tasks
68
Notes on using the summarization utility

The remainder of this chapter gives several hints about using the summarization utility.
Selecting input records

The EXCLxxxx and INCLxxxx control statements are often used together to select CMRDETL records for summarization. As their names suggest, EXCLxxxx excludes records and INCLxxxx includes records used by the summarization utility. The parameters of both control statements define the CICS resources whose records are selected for summarization. Both control statements can be specified more than once for each workload defined in the job. The control statements are interpreted in the following manner during summarization processing:
s
If more than one INCLxxxx control statement is specified for a single workload, the parameters are logically ORed together. Records are included if the selected resources meet any parameter condition. If more than one EXCLxxxx control statement is specified for a single workload, the parameters are logically ORed together. Records are excluded if the selected resources meet any parameter condition. INCLxxxx processing occurs first. If a resource meets a condition specified by the INCLxxxx control statement, the included records are then tested against the conditions set by the EXCLxxxx control statement. Records are summarized if they meet any condition set by the INCLxxxx statement and do not meet any condition set by the EXCLxxxx statement.
Figure 17 on page 70 shows the logic of selecting records for summarization when the INCLxxxx and EXCLxxxx control statements are included in the summarization job.
Chapter 3
69
Figure 17
Selecting records for summarization

No Include ? Yes Exclude ? Skip Records Yes No
Process
Records
User Exit Skip Exclude Summarization
Process
from
Include in Summarization
Notice the USEREXIT control statement at the bottom of Figure 17. You can set additional selection criteria with a user-written routine that tests each record. This routine is given control when a record is processed during the summarization job. In the example shown in Figure 17, the exit to the user-written routine is passed all records that have been tested previously by the INCLxxxx and EXCLxxxx control statements. The user-exit routine is the final arbiter of including or excluding records from the summarization job. Records that have been previously excluded can be reinserted into the job to be summarized. Likewise, records that have been selected for summarization can be overridden and excluded from the job.
Detail file switch

The summarization utilitys detail file switch provides an opportunity to summarize data. By adding a step after the archive is completed, you can summarize the data from the detail data set as needed. You can summarize each detail record into multiple workloads in a single execution of the program. Each execution can produce as many workloads as required. When the output file is VSAM, the summarization program combines data from multiple executions; it is not necessary to summarize only at daily intervals, such as midnight.
70
Replacing summary data

The default summarization program can combine data from the current execution with summary data created earlier if you use a VSAM output data set. Occasionally, you may need to replace an existing workload with new data; for example, you could have omitted a transaction when creating the workload definitions. To replace a summary workload, specify REPLACE=YES when you run the summarization utility. Only workloads and summarization intervals in the current execution are affected by the REPLACE=YES option.
WARNING
The REPLACE=YES control statement replaces all workload data with duplicate time frames and also deletes any current intervals that no longer contain data.
Viewing summary records

If summary records are stored on a VSAM data set, they can be viewed with the MAINVIEW for CICS HISTORY service. Intervals can be selected from the data set and expanded to get varying levels of detail about the records that compose the interval. To view summary records with the HISTORY service:
1 Create summary records and store them on a VSAM data set. SUMMDATA is the
default target name to view summary records. Otherwise, use the NEWAPPL control statement to set the key of the records to a name other than the CICS region.
2 Add SUMMDATA to BBPARM member BBIJNT00 (do not add BBIISP00 block
requests).
3 Add SUMMDATA to the BBPARM member CMRDTL00 and point DSN1 to the
VSAM data set created in Step 1. An example follows:
TARGET=SUMMDATA,DSN1=SUMMARY.VSAM
4 Select the HISTORY service from the MAINVIEW for CICS Primary Option Menu. 5 Access the summary records by setting the CICS target to SUMMDATA in the
upper right corner of the MAINVIEW for CICS History Selection panel. CICS ===> SUMMDATA
Chapter 3 Managing report data 71
6 Complete the remaining fields of the History Selection panel to set any filters and
the time and date intervals of the records that you want to view. Figure 18 shows the four workloads of daily summary records created earlier with the summarization utility. You can get more detail about each workload by expanding the HISTORY views. Figure 18 HISTORY Service Display of summary records
BMC Software -------MAINVIEW FOR CICS HISTORY SUMMARY------ TIME: 08:29:24 COMMAND ===> CICS ===> SUMMDATA -- ENDING -TASK WORKLOAD SUMM CICS RESPONSE CPU FILE STOR DATE TIME COUNT NAME INTV GROUP TIME TIME CALLS (K) A 24:00:00 989 CMRTRANS 1440 CICSPROD 0.156 0.064 8 10 24:00:00 96,876 IBMTRANS 1440 CICSPROD 0.026 0.002 1 8 24:00:00 457,762 ALLTRANS 1440 CICSPROD 1.017 0.016 6 29 * 24:00:00 469,937 ALLTERM 1440 CICSPROD 0.750 0.015 6 28 *
The HISTORY display of summary records varies from the standard detail data. The ending time shown in the HISTORY display is the end of the summarization interval, not the ending time of the detail records associated with the interval.
Storing Summary Data on VSAM Data Sets

A VSAM data set provides the best repository of summarized data:
s
The space required to maintain a month of summarized data is small. The CMRPURG utility of MAINVIEW for CICS can be used to maintain the data set, so you can keep as little or as much data available as you require. Archiving data (CMRPURG) on page 34 gives instructions for removing unwanted records from the VSAM data set with the CMRPURG utility. Summary records can be viewed with the HISTORY service. The summarization program combines new data with existing data only when the output is VSAM. This provides a complete interval of data even when different jobs summarized the data. For example, a daily summary record can comprise data collected from several CMRDETL data sets when switching is active. You can create combined workloads from detail files within multiple CICS regions without first combining the data. Although you can create combined workloads using a sequential data set, data from multiple executions is not combined and the CICS input data must be sorted into date/time sequence. The requirement to sort input data is removed. However, performance improves when input data is stored in sorted order.
72
See BBSAMP member CMRDFSUM for guidelines about allocating a VSAM or sequential data set.
CMRDETL fields altered by the summarization utility

The summarization utility alters the values of some CMRDETL record fields during summarization. Table 4 describes the changes to these fields after they have been summarized: Table 4
Field name TYPE CMR$SYS CMRTIME CMRCLOCK CMRRECID
Changes to CMRDETL record fields after summarization (part 1 of 2)

Contents of a summarized record contains 6D, if any input record during the interval is a 6D abend record (see T6EABCD) assigned the value from the NEWAPPL control statement contains the time at the end of the interval contains the name of the region or SGROUP (if multiple CICS regions are combined) contains a value to make each record unique This record comprises a minute value of one byte followed by a sequential number assigned to each workload. This number is in packed format
T6ETRTY T6EUSER
contains an X indicating a summary record contains the name of the CICS region or SGROUP, if multiple CICS regions are combined The alias, SUMCICS, can be used by PRL reports to specify the data held in the T6EUSER field. Note: CMRCLOCK also sets T6EUSER to the name of the CICS region or SGROUP when multiple CICS regions are combined
T6ETMID
contains the number of minutes in the summarization interval The alias, SUMINTV, can be used by PRL reports to specify the data held in the T6ETMID field.
T6EPGNM
contains the WORKLOAD=name The alias, SUMWKLD, can be used by PRL reports to specify the data held in the T6EPGNM field.
T6ESTIME T6ETASK T6EABCD
contains the start time of the interval or the time from the start of data when it is the first interval for a given execution contains the number of transactions in the summary contains the number of transactions in the summary that were from type 6D records (abend)
Chapter 3
73
Table 4
Field name T6EFCNT T6EFNnnn T6EFTnnn
Changes to CMRDETL record fields after summarization (part 2 of 2)

Contents of a summarized record contains the number of files in the summary record, based on the options you specify contains the file type if you specify DETAIL=NO or the actual file name if you specify DETAIL=YES contains the average file time for all requests to the file in T6EFNnnn Note: T6EFTnnn does not contain the average time of all transactions. It contains the average time of only those transactions that access the file.
T6EFCnnn
contains the average file calls for all requests to the file in T6EFNnnn Note: T6EFCnnn does not contain the average count for all transactions. It contains the average count for those transactions that accessed the file.
Binary, time, and character fields

When you process CMRDETL records with the summarization utility, several data types of the resulting summary records vary from the original record as a result of summarization. Summary record fields that contain binary, time, or character data are processed during summarization in the following manner: Table 5
Data type Binary Time Character
Processing summary record fields during summarization

Summary records contains the average value for all tasks during the interval contains the average time of all tasks during the interval contains asterisks (*), unless all tasks in the interval are the same When all tasks in the interval are the same, the original characters are used.
Combining summarized records

The summarization utility can combine summarized records. When you process summarized records, you can combine records from multiple workloads to create a single, new workload. You can also combine summarized records together to create a longer interval than the original records. For example, you can combine data summarized in intervals of 10 minutes into hourly summary records.
74
NOTE
When you are processing summarized records, you must select records by time period; for example, if you are combining summarized 10-minute records into hourly records, you must specify INCLTERM=10. After executing the summarization utility, the T6ETMID field contains the number of minutes in the summarization interval. This prevents records that overlap the summarization interval from being combined and creating invalid data.
The INCLTERM control statement is explained on page 57.
Chapter 3
75
76
Part
Part 2
MAINVIEW for CICS reports

This part describes miscellaneous MAINVIEW for CICS performance reports that you can produce by using PERFORMANCE REPORTER: Chapter 4, Resource Tracking and Analysis report Chapter 5, Graphic histogram reports Chapter 6, Service Level Analysis report
Part 2 MAINVIEW for CICS reports
77
78
Chapter
Resource Tracking and Analysis report

4
The MAINVIEW for CICS resource analysis program (CMRRAPR) creates the Resource Tracking and Analysis report. This report provides information about resource usage by CICS transactions and programs. A report may show summary, detail, or service data for either an individual transaction or program or a combination of transactions or programs. You select the transactions or programs that appear in the Resource Tracking and Analysis report. You declare how transactions or programs are combined by specifying parameters with the CMRRAPM macro in member CMRURAPM in BBSAMP. After you have created the macro statements, they must be assembled and and linked using member CMRASM in BBSAMP to create table CMRRAPM. You can change the name of the table by modifying the MODNAME value in CMRASM, otherwise the default name of CMRRAPM is used.
NOTE
If you change the name of the CMRRAPM table, you can use the TABLE control statement to associate the name CMRRAPR to generate the report. Refer to Report control statements on page 83 for a description of the TABLE control statement.
The remaining sections of this chapter describe the macro assembly statements, JCL, and control statements used to produce a Resource Tracking and Analysis report.
Creating a Resource Tracking and Analysis report

To create a Resource Tracking and Analysis report, submit the JCL in member CMRRAP in BBSAMP to execute the CMRRAPR report program and any associated control statements as a batch job.
Chapter 4 Resource Tracking and Analysis report 79
Creating a Resource Tracking and Analysis report
The sample job stream shown in Figure 19 produces the Resource Tracking and Analysis report, shown in Figure 21 on page 85. Figure 19 Resource Tracking and Analysis reportinput JCL
//jobname JOB user parameters /*JOBPARM user parameters //JOBLIB DD DISP=SHR,DSN=CMR.CMRV5.BBLINK //CMRRAPR EXEC PGM=CMRRAPR //SYSOUT DD SYSOUT=* //SYSPRINT DD SYSOUT=* //REPORT DD SYSOUT=* //CMRDETL DD DISP=SHR,DSN=CMR.CMRV5.CMRDETL //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,(1,1)) //SYSIN DD * (control statements)
The required JCL to produce this report consists of the following statements: Table 6
JCL JOB EXEC //JOBLIB //CMRDETL //REPORT //SORTWKnn
Required JCL for Resource Tracking and Analysis report

Use initiates the job specifies the name of the program that generates the report (PGM=CMRRAPR) defines the program library containing the MAINVIEW for CICS load modules (BBLINK) defines the VSAM data set containing records for each CICS transaction (CMRDETL) defines the report output for the predefined reports identifies one to nine work data sets that can be defined for CMRRAPR data sorting nn is a numeric value (01-09).
//SYSPRINT
defines the output class for a sort utility The name of the DD statement is determined at CICS system generation.
//TAPEIN
defines the archive data set produced by CMRPURG, described in Archiving data (CMRPURG) on page 34, as input to the report program defines the output class defines PERFORMANCE REPORTER control statements as input to the batch program
//SYSOUT //SYSIN
80
Sample CMRRAPM macro assembly statements

Figure 20 shows an example of BBSAMP (CMRRAPM) macro assembly statements that are needed to create the Resource Tracking and Analysis report. Figure 20 CMRRAPM macro assembly statements
CMRRAPM TYPE=INITIAL CMRRAPM TYPE=ENTRY,GROUP=xxxxxxxxxxxx, MEMBERS=(xxxxxxxx,...,xxxxxxxx), SERVICE=nnn (Additional TYPE=ENTRY statements, as needed) CMRRAPM TYPE=FINAL END
The following table describes each CMRRAPM macro assembly statement: Table 7
Statement TYPE=INITIAL
Description of the CMRRAPM macro assembly statements (part 1 of 2)

Description generates a group of MAINVIEW for CICS modules This statement must be entered first and placed on a separate line by itself. Specify CMR=YES to create a group that includes all MAINVIEW for CICS transaction IDs (the default); otherwise, specify CMR=NO.
Chapter 4
81
Table 7
Statement
Description of the CMRRAPM macro assembly statements (part 2 of 2)

Description defines one group of transactions or programs and the response time to be compared with the specified service level One TYPE=ENTRY statement must be specified for each group to be reported, where:
s
TYPE=ENTRY
GROUP= is any 1- to 12-character symbolic identifier for this group. Note: MAINVIEW for CICS transactions are automatically assigned the group identifier, MAINVIEW for CICS.
MEMBERS= is the list of transaction IDs or program names (PANL, MAPS, or PSBS, if they have been monitored with the transaction monitor) to be included in this group. A skip character is a plus sign (+), which can be used in IDs or names to indicate generic subgroups of transactions or programs to be included. There is no limit on the number of IDs or names specified. However, a given ID or name can exist in only one group. SERVICE= is a service level value to which the response time for each transaction or program is to be compared, expressed in tenths of seconds. The default is 10 (1.0 seconds).
TYPE=FINAL
ends the generation parameters It must be the last statement and must be placed on a line by itself. Note: BBSAMP member CMRASM contains sample code to assemble and link-edit the CMRRAPM module. CMRRAPM must be linked non-reentrant.
82
Report control statements
Report control statements

All control statements are optional. If specified, each statement must begin in column 1. Table 8 describes each control statement: Table 8 Report control statements (part 1 of 2)
Description starting date of the data to be included in the report, in mm/dd/yyyy format If the statement is omitted, no data is rejected as too old. Note: If you specify a date in the old format (mm/dd/yy), yy is interpreted as 19yy. TO DATE ending date of the data to be included in the report, in mm/dd/yyyy format If the statement is omitted, no data is rejected as too new. Note: If you specify a date in the old format (mm/dd/yy), yy is interpreted as 19yy. FROM TIME starting time of the data to be included in the report, in hh:mm:ss format If the statement is omitted, no data is rejected as too early TO TIME ending time of the data to be included in the report, in hh:mm:ss format If the statement is omitted, no data is rejected as too late. SORTWK number of sort work areas to be used Valid values are 1 to 8. If a value greater than 1 is specified, a corresponding number of DD or DLBL statements must appear in the JCL. INPUT type of input data set to be used and can be specified as one of the following values:
s s
Control Statement FROM DATE
VSAM (VSAM data setthe default) TAPE (sequential tape or disk file)
Chapter 4
83
Sample report
Table 8
TYPE
Report control statements (part 2 of 2)

Description type of report to be generated, which can be any one of the following values:
s
Control Statement
s s
SUMMARY reports three lines of transaction execution data for each transaction or program by calendar date. This value is the default. A sample of this type of report is in Figure 21 on page 85. DETAIL reports three lines of transaction execution data for each transaction or program. SERVICE specifies DETAIL information, but only for transactions or programs that did not meet their service levels.
TABLE
module name of the customized CMRRAPM table The default is CMRRAPM.
Sample report
An example of the Transaction Resource Tracking and Analysis report is shown in Figure 21 on page 85. The report is created from the JCL shown in Figure 19 on page 80. A description of the report fields and columns follows Figure 21 on page 85.
84
Report structure and conventions
Figure 21
Transaction Resource Tracking and Analysis report

MAINVIEW for CICS TRANSACTION RESOURCE TRACKING AND ANALYSIS
RUN DATE: 09/24/1999 RUN TIME: 17:17 RUN TYPE: SUMMARY
DATA DATE: 09/24/1999 TO 09/24/1999 DATA TIME: 08:30:00 TO 24:00:00
CMRRAPR PAGE
AVG. AVG. AVG. AVG. AVG. CICS/ TASK/PGM TOTAL NO. AVG. AVG. CPU TASK I/O NO. OF FC/DL1 RESP. ID. DATE OF TASKS RESP. CPU REAL WAIT TIME I/O CALLS LOAD -------- -------- --------- ------- ------- ------- ------- ------- ------ ------ -----DRGI 09241999 7 :01.226 :00.290 :00.009 :00.936 :00.000 0 0 18.0% 0 30.2%
SERVICE LEVEL/ FCP TSP DL1 JCP ECB PCT/CNT WAIT WAIT WAIT WAIT WAIT ------- ------ ------ ------ ------ -----1.0 0.0% 0.0% 0.0% 0.0% 0.0% 42.8% :00.00 :00.00 :00.00 :00.00 :00.00 3 0 0 0 0 0
DRGI
1.0 0.0% 0.0% 0.0% 0.0% 0.0% 42.8% :00.00 :00.00 :00.00 :00.00 :00.00 3 0 0 0 0 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - EMNU 09241999 3 :00.054 :00.005 :00.003 :00.049 :00.000 0 0 0.1% 1.0 0.0% 0.0% 0.0% 0.0% 0.0% 0 0.5% 100.0% :00.00 :00.00 :00.00 :00.00 :00.00 3 0 0 0 0 0 EMNU 1.0 0.0% 0.0% 0.0% 0.0% 0.0% 100.0% :00.00 :00.00 :00.00 :00.00 :00.00 3 0 0 0 0 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - FINC 09241999 12 :00.502 :00.347 :00.005 :00.155 :00.000 0 0 37.0% 1.0 0.0% 0.0% 0.0% 0.0% 0.0% 0 21.2% 92.6% :00.00 :00.00 :00.00 :00.00 :00.00 11 0 0 0 0 0 FINC 1.0 0.0% 0.0% 0.0% 0.0% 0.0% 92.6% :00.00 :00.00 :00.00 :00.00 :00.00 11 0 0 0 0 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - PROV 09241999 11 :00.173 :00.018 :00.004 :00.155 :00.169 0 0 1.7% 1.0 26.6% 0.0% 0.0% 0.0% 0.0% 0 6.7% 90.9% :00.04 :00.00 :00.00 :00.00 :00.00 10 3 0 0 0 0 PROV 1.0 26.6% 0.0% 0.0% 0.0% 0.0% 90.9% :00.04 :00.00 :00.00 :00.00 :00.00 16 12 0 0 0 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - RELG 09241999 2 :00.094 :00.093 :00.003 :00.001 :00.000 0 0 1.6% 1.0 0.0% 0.0% 0.0% 0.0% 0.0% 0 0.6% 100.0% :00.00 :00.00 :00.00 :00.00 :00.00 2 0 0 0 0 0 RELG ****** 2 :00.094 :00.093 :00.003 :00.001 :00.000 0 0 0 1.6% 0.6% 1.0 0.0% 100.0% :00.00 2 0 - - - - - - 4.7% 89.7% :00.01 70 18 - - - - - - 4.7% 89.7% :00.01 70 18 - - - - - - 4.7% 89.7% :00.01 70 18 - - - - - - 4.7% 89.7% :00.01 70 18 - - - - - - 4.7% 89.7% :00.01 70 18 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - 0.0% 0.0% :00.00 :00.00 0 0 - - - - - ****** 11 :00.173 :00.018 :00.004 :00.155 :00.169 0 0 0 1.7% 6.7% ****** 12 :00.502 :00.347 :00.005 :00.155 :00.000 0 0 0 37.0% 21.2% ****** 3 :00.054 :00.005 :00.003 :00.049 :00.000 0 0 0 0.1% 0.5%
******
7 :01.226 :00.290 :00.009 :00.936 :00.000
0 0
18.0% 30.2%
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - G:PAYROLL 35 :00.363 :00.144 :00.008 :00.219 :00.075 0 1 99.6% 0 99.8% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ** SELECTED ** 35 :00.363 :00.144 :00.008 :00.219 :00.075 0 1 99.6% 0 99.8% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - G:UNSELECTED 27 :00.363 :00.144 :00.008 :00.219 :00.075 1 1 99.6% 0 99.8% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - * UNSELECTED * 27 :00.363 :00.144 :00.008 :00.219 :00.075 0 1 99.6% 0 99.8% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - GRAND TOTALS 62 :00.363 :00.144 :00.008 :00.219 :00.075 0 1 99.6% 0 99.8% - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - TRANSACTIONS PER SECOND: 0.00

The report shown in Figure 21 on page 85 is produced by the CMRRAPR program. The report run type, SUMMARY, is printed in the top left corner on line 3 below the date and time fields. A summary report shows a summary of transaction
Chapter 4 Resource Tracking and Analysis report 85
performance in chronological order. A subtotal by transaction ID is marked with asterisks (******) in the DATE column before the next transaction begins. Five transaction IDs are included in this report: DRGI, EMNU, FINC, PROV, and RELG. A detail report shows performance data for every transaction. A service report is also a type of detail report, except it is limited to transactions or programs that did not meet their service levels. Otherwise, the format of both reports is the same as the summary report shown in Figure 21 on page 85. G:PAYROLL is the name of the group formed by programs DRGI, EMNU, FINC, PROV, and RELG. Values in this row represent the sum of the subtotals reported for DRGI, EMNU, FINC, PROV, and RELG. The areas marked ** SELECTED ** and * UNSELECTED * show totals for programs whose IDs were either matched (SELECTED) or not matched (UNSELECTED) against the CMRRAPM table. Both program IDs in this report were matched in the table:
s
The SELECTED row shows values equal to the combined data for programs DRGI, EMNU, FINC, PROV, and RELG in the G:PAYROLL row. The UNSELECTED row shows totals for programs other than the selected programs.
The GRAND TOTALS area shows total counts and percentages for all reported programs. In this report, the GRAND TOTALS field shows the sum of all listed values (SELECTED and UNSELECTED). The TRANSACTIONS PER SECOND field at the lower left of the report is the average rate calculated for all transactions that occurred during the period that data was collected for the report. This example represents 15.5 hours; thus, the transactions per second are 62 / (15.5 * 60 * 60), or 0.001.
NOTE
Any count that exceeds the width of its report field is expressed in millions, indicated by an M suffix. A time that is longer than its report field is forced to a value of 999.999 on the report. The actual time is retained, however, and all values derived from it are maintained correctly.
Column headings in this report are standard; they are described in the following section.
86
Field descriptions
Field descriptions
Table 9 describes the applicable report fields: Table 9
Field TASK/PGM ID
Report fields (part 1 of 3)

Description transaction ID or program name being reported This field can contain any of the following entries:
s
a 4-character transaction ID (unless the selection matches the program name) a 1- to 12-character identifier assigned to this group of transactions or programs This ID is used when subtotals for a group are summarized.
**SELECTED**, indicating a group of transaction or program IDs that were matched to the CMRRAPM table This entry is used when a group break is processed.
*UNSELECTED*, indicating a group of transactions or program IDs that were not matched to the CMRRAPM table This entry is used when a group break is processed.
GRAND TOTALS, showing a total for the entire report
DATE
in a summary report, the Gregorian date of the reported data (with a date format of mm/dd/yyyy) in detail or service reports, the following format:
s s s
Line 1 shows the Gregorian date of the reported activity. Line 2 shows the transaction ending time, in hh:mm:ss format. Line 3 shows the operator identifier that invoked the task or program.
TOTAL NO. OF TASKS AVG. RESP. AVG. CPU AVG. CPU REAL AVG. TASK WAIT
total number of tasks that reference this task/program average internal response time for this task/program, in seconds average elapsed time in seconds that the task was dispatched during each use of this task/program average actual CPU time required to execute this task/program, in seconds average wait time of this task/program, in seconds
Chapter 4
87
Field descriptions
Table 9
Field

Description average amount of time required to complete an I/O operation, in seconds (This value is the I/O service time experienced by this task/program.) average number of file, journal, temporary storage, and DL/I I/Os issued during the execution of this task/program on line 1, shows the average number of file control calls issued during each execution of this task/program on line 2, shows the average number of DL/I calls issued during each execution of this task/program
AVG. I/O TIME
AVG. NO. OF I/O AVG. FC/DLI CALLS
CICS/RESP. LOAD
on line 1, shows the percentage of CICS CPU used by this task/program on line 2, shows the percentage of the total CICS response time used by this task/program
SERVICE LEVEL/PCT/CNT
on line 1, shows the service level of this task/program on line 2, shows the percentage of transactions whose response time met this service level time value on line 3, shows the number of transactions whose response time met this service level time value.
FCP WAIT
on line 1, shows the percentage of this task/programs response time that can be attributed to file control access requests on line 2, shows the average amount of time attributed to file control access requests, in seconds on line 3, shows the total number of file control access requests issued
TSP WAIT
on line 1, shows the percentage of this task/programs response time attributed to temporary storage access requests on line 2, shows the average amount of time attributed to temporary storage access requests, in seconds on line 3, shows the total number of temporary storage access requests issued
DLI WAIT
on line 1, shows the percentage of this task/programs response time attributed to DL/I access requests on line 2, shows the average amount of time attributed to DL/I access requests, in seconds on line 3, shows the total number of DL/I access requests issued
88
Field descriptions
Table 9
Field JCP WAIT

Description on line 1, shows the percentage of this task/programs response time attributed to journal control access requests on line 2, shows the average amount of time attributed to journal control access requests, in seconds on line 3, shows the total number of journal control access requests issued
ECB WAIT
on line 1, shows the percentage of this task/programs response time attributed to user event control block (ECB) wait requests on line 2, shows the average amount of time attributed to ECB wait requests, in seconds on line 3, shows the total number of ECB wait requests issued
Chapter 4
89
Field descriptions
90
Chapter
Graphic histogram reports

The CMRCHRT member of your BBSAMP data set contains sample JCL and control statements to create histograms from data collected by MAINVIEW for CICS. These reports graphically represent selected CICS resources as horizontal histograms. Over 260 data elements recorded by MAINVIEW for CICS can be combined to create tabular historical reports and histograms. Using control statements, you select the resources and the type of processing shown in your histograms.
Creating a histogram report

The CMRCHRT batch program creates a histogram report. Both the JCL and control statements to run the CMRCHRT program are described in the following sections. The sample job shown in Figure 22 produces a histogram report for all versions of CICS. Figure 22 Example JCL to create histogram reports
//jobname JOB user parameters /*JOBPARM user parameters //CMRCHRT EXEC PGM=CMRCHRT //JOBLIB DD DISP=SHR,DSN=CMR.Vnn.BBLINK //CMRDETL DD DISP=SHR,DSN=CMR.Vnn.CMRDETL //CMRSTAT DD DISP=SHR,DSN=CMR.Vnn.CMRSTATS //TAPEIN DD DISP=SHR,DSN=CMR.Vnn.CMRSTATS //REPORT DD SYSOUT=* //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,(1,1)) //SYSPRINT DD SYSOUT=* //SYSOUT DD SYSOUT=* //SYSIN DD * (control statements)
Chapter 5 Graphic histogram reports
91
Creating a histogram report
The JCL to execute the CMRCHRT program consists of the following statements:
JOB EXEC //JOBLIB //CMRDETL //CMRSTAT //TAPEIN //REPORT //SORTWKnn initiates the job specifies the name of the program (PGM=CMRCHRT) that generates the report defines the program library containing the MAINVIEW for CICS load modules (BBLINK) defines the VSAM data set containing T6E detail transaction records defines the VSAM data set or sequential file containing records of CICS operation statistics (CMRSTATS) defines the sequential disk or tape file containing records of CICS operation statistics (CMRSTATS) defines the report output for the predefined reports identifies one to nine work data sets that can be defined for CMRCHRT data sorting Define nn as a numeric value from 01 to 09. //SYSPRINT defines the output class for a sort utility The name of the DD statement is determined at CICS system generation. //SYSOUT //SYSIN defines the output class defines PERFORMANCE REPORTER control statements as input to the batch program
CMRCHRT runs with the following restrictions:

s
Data must be created from the SMF records selected by using CMRSTATS options as follows: GAPPLID or SAPPLID must be used to segregate SMF data by CICS regions into separate files. CMRCHRT cannot select by APPLID.
Only three class types of CMRCHRT can be run: CLASS=CICS CLASS=APPLICATION CLASS=SERVICE
INPUT=TAPE and a TAPEIN DD statement pointing to the file are required.
92
Control statements
Control statements
Control statements identify the CICS resources and associated performance statistics that appear in histogram reports. A control statement set is a group of control statement keywords and parameters that specify one histogram report. A set can contain valid keywords in any sequence, but each keyword can be specified only once in each set. A histogram control statement set is structured as follows: Table 10
Structure keyword=value
Histogram control statement set structure

Description (optional) sets up date and time ranges, titles, graph scales, and so on for this report Defaults are used if these keywords are not specified.
class data=type END
(one required) specifies the CICS resource appearing in the histogram (one required) specifies the type of data appearing in the histogram (required) ends this report request
Control statements follow these rules:

s
Each control statement must begin in column 1. If a control statement keyword appears more than once in a set, only its last specification is used. If a control statement keyword is not specified in a set, its default is used. Multiple control statement sets can be specified in a single execution of the CMRCHRT program to produce multiple histograms. However, separate jobs must be submitted to produce multiple histograms if report data is on tape (see Archiving data (CMRPURG) on page 34). Comments can be added to control statements by leaving at least one blank after the statement and then starting the comment. Any statement with an asterisk (*) in column 1 is considered to be a comment and is ignored.
93
Keyword=value statements
The following keywords specify the range of data, titles, calculation methods, and scales of each histogram report. If any keyword is omitted, the default is used.
TITLE
TITLE specifies the title of the report; up to 60 characters are allowed, including blanks. The default is 60 blanks.
FROM DATE and TO DATE

FROM DATE and TO DATE specify the date range (in mm/dd/yyyy format) to select input data. If used, these keywords must be respecified in every control statement set. The defaults are FROM DATE=09/09/0000 and TO DATE=12/31/9999; data from all available dates appears in the histograms.
NOTE
FROM TIME and TO TIME

FROM TIME and TO TIME specify the daily time range (in hh:mm:ss format) to select input data. If used, these keywords must be respecified in every control statement set. The defaults are FROM TIME=00:00:00 and TO TIME=24:00:00; data from all available times appears in the histograms.
MINUTES
MINUTES represents the length of time the requested file data is to be accumulated, in minutes. The accumulated data is used to create a line of report data. Any value from 1 through 99999 is valid; the default is 1 (minute). The keywords MINUTES and APPL=(ALL) are mutually exclusive. For more information about APPL=(ALL), see Applications class on page 97.
BREAK={DAILY | HOURLY | MONTHLY}

BREAK divides the date and time ranges into fixed report intervals.
94
The keywords BREAK= and APPL=(ALL) are mutually exclusive. For more information on APPL=(ALL), see Applications class on page 97. Valid values for BREAK are as follows:
s s s
DAILY, the default, specifies that each days data starts on a new report page. HOURLY specifies that each hours data starts on a new report page. MONTHLY specifies that each months data starts on a new report page.
CALC={PERCENT | AVERAGE | TOTAL}

CALC specifies the calculation method that is used to present the data. Valid values are as follows:
s s s
PERCENT, the default, reports percentages. AVERAGE reports averages. TOTAL reports total counts.
NOTE
The CALC keyword is ignored if you specify DATA=PAGING.
SCALE={100 | nnnnn | FLOAT}

SCALE defines histogram scale values. Valid values are as follows:
nnnnn specifies the highest scale value Valid scalar numbers include
s s s
from 1 to 99999 for counts up to 100 for percent up to 68700 for time; the default scale is 100
Anything beyond the maximum range is truncated. FLOAT calculates the scale to accommodate the largest reported value
COMPANY=name
COMPANY specifies the company name to be displayed at the top of the histogram. The default is BMC Software. When specified, this parameter remains in effect for the entire program execution or until another COMPANY keyword is encountered.
95
INPUT={VSAM | TAPE}
INPUT identifies the organization of the input data set. Valid values are as follows:
s s
VSAM (VSAM data setthe default) TAPE (sequential tape or disk file)
When specified, this parameter remains in effect either for the entire program execution or until another INPUT keyword is encountered.
NOTE
If INPUT=TAPE is specified, you must specify a TAPEIN DD statement in your JCL.
TABLE=name
TABLE specifies the module name of the modified CMRRAPM table to be used by the Applications Class GROUP parameter. TABLE must be specified before the associated GROUP parameter. The default is CMRRAPM. For more information, see Applications class on page 97. When specified, this parameter remains in effect either for the entire program execution or until another TABLE keyword is encountered.
ENDDATA={DATE | FILE}
ENDDATA specifies whether end of data should correspond with the TO DATE (ENDDATA=DATE) or end of file (ENDDATA=FILE) keyword. The default is DATE. When specified, this parameter remains in effect for the entire execution of the program or until another ENDDATA keyword is encountered. ENDDATA is required if input data is
s
selected from an archive tape that contains data merged from multiple CICS regions (see Archiving data (CMRPURG) on page 34) sorted by criteria other than date and time
96
CLASS and DATA control statements

One CLASS and one DATA statement are required for each histogram report. The CLASS control statement identifies the resource to be reported; the DATA statement identifies the data to be graphed for that class. Each CLASS statement has its own set of data types, and only one DATA type can be selected for each histogram report.
Applications class
The Applications Class control statements identify a specific type of application or group of applications that appear in histogram reports. The format is as follows: APPL={name | (ALL)} where
APPL is any of the following applications:
s s s s s s s
TRAN (transaction name) PROG (program name) TERM (terminal name) PANL (DMS panel name) MAPS=Storage map name PSBS=PSB name GROUP=A group of transactions or programs defined in the CMRRAPM table specified in the TABLE= parameter
name (ALL)
specifies an application name requests all the members in the specified class Note: If you specify ALL, it must be enclosed in parentheses.
The keywords MINUTES, APPL=(ALL), and BREAK are mutually exclusive. For further information about these keywords, refer to Keyword=value statements on page 94. One of the following DATA statements must follow the APPL statement, beginning in column 1:
DATA=RESPONSE DATA=ALL-WAIT DATA=CPU DATA=FC-WAIT DATA=FC-IOCNT DATA=TC-WAIT response time total transaction wait time CPU time file I/O wait time number of file I/Os terminal wait time
97
DATA=TC-IOCNT DATA=JC-WAIT DATA=JC-IOCNT DATA=TS-WAIT DATA=TS-IOCNT DATA=ECB-WAIT DATA=ECB-WCNT DATA=SUSPENDT DATA=SUSPENDC DATA=FCPCALLS DATA=DLICALLS DATA=DL/IWAIT DATA=DL/IOCNT
number of terminal I/Os journal I/O wait time number of journal I/Os temporary storage I/O wait time number of temporary storage I/Os user event wait time number of user event wait requests suspend wait time number of times suspended number of file control (FCP) calls or requests number of DL/I calls DL/I I/O wait time number of DL/I I/Os
Service levels class

The Service Level Class control statements report about service level performance. The format is SERVICE No DATA control statements are used with the Service Level Class control statement.
CICS system class

CICS system information is reported by using system data journaled by CICS. The format is CICS
98
One of the following DATA control statements must follow the CICS statement, beginning in column 1:
s s s s s s s s s s s s s s s s s
DATA=CICSCPU (total CICS CPU time and user CPU time) DATA=KCP-CPU (task control CPU time) DATA=KCP-CPUR (task control CPU real time) DATA=KCP-CPUC (task control dispatch count) DATA=TCP-CPU (terminal control CPU time) DATA=TCP-CPUR (terminal control CPU real time) DATA=TCP-CPUC (terminal control dispatch count) DATA=JCP-CPU (journal control CPU time) DATA=JCP-CPUR (journal control CPU real time) DATA=JCP-CPUC (journal control dispatch count) DATA=USERCPU (user CPU time) DATA=USERCPUR (user CPU real time) DATA=USERCPUC (user dispatch count) DATA=PGIN (CICS page-in operations) DATA=PGOT (CICS page-out operations) DATA=PAGING (CICS page-in and page-out rate per second) DATA=TASKRATE (transaction rate per second)
Processing program table class

The Processing Program Table reports about a specific PPT entry or all PPT entries. The format is PPT={xxxxxxxx | (ALL)} where xxxxxxxx is a valid PPT entry name or (ALL), which reports on all the entries in the PPT. One of the following DATA statements must follow the PPT statement, beginning in column 1:
s s
DATA=USECNT (number of times a program was used) DATA=FETCH (number of times a program was fetched from the OS/MVS load library)
NOTE
If the CALC keyword is specified as CALC=PERCENT and DATA=FETCH is specified here, a FETCH-PERCENT report is produced.
99
Program Control Table (PCT) class

The Program Control Table Class control statement reports about a specific PCT entry or all PCT entries. The format is
PCT={xxxx | (ALL)}
where xxxx is a valid PCT entry name or (ALL), which reports on all the entries in the PCT. One of the following DATA statements must follow the PCT statement, beginning in column 1:
s s s
DATA=USECNT (transaction usage count) DATA=STGVIOL (number of storage violations) DATA=ADDSTG (number of additional storage requests)
NOTE
If the CALC keyword is specified as CALC=PERCENT and DATA=ADDSTG is specified here, the report shows the percentage of time additional storage was needed.
File Control Table class

The File Control Table Class control statement reports about a specific FCT entry or all entries. The format is
FCT={xxxxxxxx | (ALL)}
where xxxxxxxx is a valid FCT entry name. ALL reports on all the entries in the FCT.
100
One of the following DATA statements must follow the FCT statement, beginning in column 1:
s s s s s s s s s s
DATA=READ (file READ requests) DATA=READNEXT (file READNEXT requests (BROWSE)) DATA=READ-UPD (file READ FOR UPDATE requests) DATA=UPDATE (file UPDATE requests) DATA=INSERT (file INSERT (add) requests) DATA=DELETE (file DELETE requests) DATA=DATAEXCP (data component EXCP count) DATA=INDXEXCP (index component EXCP count) DATA=CI-SPLIT (count of control interval (CI) splits) DATA=CA-SPLIT (count of control area (CA) splits)
NOTE
The CALC=PERCENT statement uses file control requests as a divisor. For example, using DATA=DATAEXCP produces a histogram charting the percentage of physical I/O requests in relation to logical I/O requests.
Common System Area class

The Common System Area Class control statement reports about CSA data. The format is CSA One of the following DATA statements must follow the CSA statement, beginning in column 1:
s s s s s s s s s s
DATA=GETMAINS (count of GETMAIN requests) DATA=FREEMAIN (count of FREEMAIN requests) DATA=SOSCNT (times CICS went short-on-storage (SOS)) DATA=DUMPS (number of transaction dumps) DATA=ASRA (number of program checks) DATA=STGWAIT (number of times transactions waited for storage) DATA=MAXCNT (number of times CICS maximum task condition was reached) DATA=STGVIOL (number of storage violations) DATA=DTBLOGS (number of dynamic transaction log records written) DATA=DTBSPILL (number of DTB log records that spilled to temporary auxiliary storage (AUX))
101
NOTE
The CALC=PERCENT statement uses the total CICS transaction count as a divisor.
Transaction Class Table class

The Transaction Class Table Class control statements report about a transaction class or all transaction classes. The format is
CSO={xx | (ALL)}
where xx is any transaction class ID from 1 through 10 or (ALL), which reports on all transaction classes. One of the following DATA statements must follow the CSO statement, beginning in column 1:
s s
DATA=MAXCNT (maximum setting for the class) DATA=HWMCNT (high-water mark reached)
Destination Control Table class

The Destination Control Table Class control statements report about one destination ID or all destination IDs. The format is
DCT={xxxx | (ALL)}
where xxxx is any destination ID or (ALL), which reports on all destination IDs. The only valid DATA statement allowed following the DCT statement is as follows: DATA=USECNT (number of times the destination was used)
Journal Control Table class

The Journal Control Table Class control statement reports about one journal number or all journal numbers. The format is
JCT={xx | (ALL)}
where xx is any journal number from 1 through 99 or (ALL), which reports on all journal numbers.
One of the following DATA statements must follow the JCT statement, beginning in column 1:
s s s s s s
DATA=RECCNT (number of records written) DATA=BLKCNT (number of blocks written) DATA=FULLCNT (number of times buffer was full) DATA=SHIFTCNT (number of buffer shift-ups) DATA=AVEBUF (average block size written) DATA=MAXBUF (maximum block size allowed)
NOTE
If the CALC keyword is set to CALC=PERCENT and DATA=RECCNT is specified here, RECCNT is used as the divisor. If CALC=PERCENT and AVEBUF or MAXBUF are specified here, MAXBUF is used as the divisor.
DL/I class
DL/I Class control statements report about DL/I data. The format is DLZ One of the following DATA statements must follow the DLZ statement, beginning in column 1:
s s s s s s s s
DATA=PSBCNT (number of PSBs in the system) DATA=PSTCNT (number of PSTs in the system) DATA=DBDCNT (number of DBDs in the system) DATA=SUSPENDS (number of suspended tasks) DATA=SCHEDULE (number of scheduling requests) DATA=MAXCNT (maximum task count) DATA=DEADLOCK (number of deadlock occurrences) DATA=DUPPSBS (number of duplicate PSBs created)
103
NOTE
These fields use total DL/I transaction count as a divisor when CALC=PERCENT is specified.
s s s s s s
DATA=READS (number of READ requests) DATA=READBUF (number of READs satisfied by data in the buffer pools) DATA=EXCPIN (number of input EXCP commands) DATA=EXCPOUT (number of output EXCP commands) DATA=ALTERS (number of buffer alterations) DATA=IOERRORS (number of I/O errors)
NOTE
These fields use total buffer requests as a divisor when CALC=PERCENT is specified.
s s
DATA=GETUNIQ (number of GET UNIQUE type requests) DATA=GETNEXT (number of GET NEXT type requests)
NOTE
These fields use total user DL/I calls as a divisor when CALC=PERCENT is specified.
Terminal Control Table class

The Terminal Control Table Class control statement reports about one terminal or all terminals. The format is
TCT={xxxx | (ALL)}
where xxxx is any terminal ID or (ALL), which reports on all terminal IDs. One of the following DATA statements must follow the TCT statement, beginning in column 1:
s s s s s
DATA=INPUTS (number of times the terminal communicates with CICS) DATA=OUTPUTS (number of times CICS communicates with the terminal) DATA=TRANSCNT (number of transactions) DATA=TERMERRS (number of terminal failures, such as a PROG470 abend) DATA=OPERERRS (number of application program failures)
104
END statement
NOTE
These fields use TRANSCCNT as a divisor when CALC=PERCENT is specified.
Temporary Storage Table class

The Temporary Storage Table Class control statement specifies a histogram report that shows the usage of temporary storage data. The format is TST One of the following DATA statements must follow the TST statement, beginning in column 1:
s s s s s s s s s s
DATA=PUTSS (number of PUT requests (MAIN or AUX)) DATA=PUTSQ (number of PUTQ requests (MAIN or AUX)) DATA=PUTSA (number of PUT or PUTQ requests (AUX)) DATA=IOERROR (number of I/O errors) DATA=EXTENDS (number of TSGID extensions) DATA=SUSPENDS (number of suspensions) DATA=COMPRESS (number of compressions) DATA=AVAILCI (number of available control intervals) DATA=TOTALCI (total control intervals in data set) DATA=VIRTHWM (virtual storage use high-water mark)
NOTE
All of these fields except AVAILCI use the total temporary storage requests as a divisor when CALC=PERCENT is specified; AVAILCI uses TOTALCI as the divisor.
END statement
The END statement completes a control statement set for a requested histogram report. Control statements for another histogram report can follow an END statement. An END statement is required for every requested histogram report specified with a control statement set.
105
Sample reports
Sample reports
Table 11 lists some sample histogram reports that can be produced with the CMRCHRT program. These reports offer a broad overview of CICS performance and are recommended for tuning purposes. The remainder of this section contains descriptions and illustrations of the sample reports. Table 11 Sample histogram reports
Sample control statements TRAN=(ALL) DATA=RESPONSE CALC=AVERAGE SCALE=10 END CICS INPUT=TAPE DATA=CICSCPU CALC=PERCENT MINUTES=60 FROM TIME=08:30:00 TO TIME=17:00:00 END CICS INPUT=TAPE DATA=PAGING MINUTES=60 FROM DATE=3/24/2000 TO DATE=3/24/2000 SCALE=10 END CICS INPUT=TAPE DATA=TASKRATE CALC=AVERAGE MINUTES=60 SCALE=10 END Section reference Response time for all transactions on page 106
Sample report title Response Time for All Transactions
CPU Used by CICS Execution
CPU used by CICS execution on page 108
CICS Paging Rate/Second
CICS paging rate/second on page 109
CICS Transaction Rate/Second
CICS transaction rate/second on page 110
Response time for all transactions

This report shows histograms representing the average response time over the selected interval for all transactions.
106
Response time for all transactions
Data
By default, all records stored on the CMRDETL data set are used as input to produce histogram reports. The following control statements are specified:
s s s s s
TRAN=(ALL) reports on all transactions. DATA=RESPONSE reports transaction response times. CALC=AVERAGE calculates averages. SCALE=10 sets the histogram scale to 10 seconds. END ends the report request.
The report produced by these control statements is shown in Figure 23. Figure 23 Response time for all transactions
MAINVIEW for CICS LICENSED PROPERTY OF BMC SOFTWARE CMRCHRT RESPONSE TIME FOR (ALL) TRANSACTIONS PAGE NUMBER 1 FROM DATE: 08/01/1999 TO: 08/31/1999 MINUTES.: N/A CALCULATION TYPE.: AVERAGE RUN DATE.....: 09/25/1999 FROM TIME: 00:00:00 TO: 24:00:00 BREAK...: N/A AVERAGE VALUE....: 7599.57 RUN TIME.....: 17:39:03 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 ==NAME== ====TIME=== |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ CCMF 0.40 |**** + + + + + + + + + + FCD2 5.15 |*************************************************** + + + + + CATD 654.82 |**************************************************************************************************** JNL2 0.00 | + + + + + + + + + + CRSQ 0.69 |****** + + + + + + + + + + CXRE 0.63 |****** + + + + + + + + + + CSFU 2.26 |********************** + + + + + + + + CSGM 29.14 |**************************************************************************************************** CEDA 0.00 | + + + + + + + + + + D8CS 4.46 |******************************************** + + + + + + SMN2 0.46 |**** + + + + + + + + + + CSAC 0.00 | + + + + + + + + + + CSSF 0.00 | + + + + + + + + + + CSPG 0.14 |* + + + + + + + + + + CSSN 0.58 |***** + + + + + + + + + + CEMT 69.49 |**************************************************************************************************** CEOT 0.30 |*** + + + + + + + + + + CSTA 0.00 | + + + + + + + + + + DISP 2.54 |************************* + + + + + + + + DLZZ 0.00 | + + + + + + + + + + DB2 0.00 | + + + + + + + + + + TSM4 72.64 |**************************************************************************************************** TSM5 160.56 |**************************************************************************************************** TSM2 0.00 | + + + + + + + + + + TSM8 0.00 | + + + + + + + + + + CECI 0.41 |**** + + + + + + + + + + HST2 0.00 | + + + + + + + + + + CSKP 1.35 |************* + + + + + + + + + FST2 1.27 |************ + + + + + + + + + CECS 0.07 | + + + + + + + + + + JPH1 10.79 |**************************************************************************************************** CQRY 0.47 |**** + + + + + + + + + + STAR 308.43 |**************************************************************************************************** CSTT 1.82 |****************** + + + + + + + + + CSST 0.00 | + + + + + + + + + + ==NAME== ====TIME=== |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
107
CPU used by CICS execution

This report shows histograms representing the hourly average percent CPU usage by CICS.
Data
By default, all records on the CMRSTATS data set are used to produce histogram reports. In this example, the histogram is limited to data collected during work hours 8:30 A.M. to 4:30 P.M. The following control statements are specified:
s s s s s s s
CICS reports on CICS system data. INPUT=TAPE uses sequential CMRSTATS data set. DATA=CICSCPU reports CICS and user CPU time usage. MINUTES=60 accumulates data for 60-minute periods. FROM TIME=08:30:00 uses data starting at 8:30 A.M. TO TIME=17:30:00 uses data ending at 5:30 P.M. END ends the report request.
The report produced by these statements is shown in Figure 24. Figure 24

:litdata. MAINVIEW FOR CICS LICENSED PROPERTY OF BMC SOFTWARE CMRCHRT CPU USED BY CICS EXECUTION PAGE NUMBER 1 FROM DATE: 09/17/1999 TO: 09/17/1999 MINUTES.: 00060 CALCULATION TYPE.: PERCENT RUN DATE.....: 09/28/1999 FROM TIME: 08:30:00 TO: 17:30:00 BREAK...: DAILY AVERAGE VALUE....: 1.99 RUN TIME.....: 18:09:04 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 MM/DD/YYYY HH:MM ==PERCENT==|---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 09/17/1999 08:34 2.04 |** + + + + + + + + + + 09/17/1999 09:34 2.77 |U* + + + + + + + + + + 09/17/1999 10:34 5.22 |UUU** + + + + + + + + + + 09/17/1999 11:34 7.97 |UUUUUKT + + + + + + + + + + 09/17/1999 12:24 3.04 |U** + + + + + + + + + + 09/17/1999 13:24 2.41 |U* + + + + + + + + + + 09/17/1999 14:24 2.00 |** + + + + + + + + + + 09/17/1999 15:35 44.06 |UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU* + + + + + + 09/17/1999 16:34 10.90 |UUUUUUKKT* + + + + + + + + + 09/17/1999 17:30 6.93 |UUUKTT + + + + + + + + + + MM/DD/YYYY HH:MM ==PERCENT==|---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 LEGEND: K = KCP-CPU T = TCP-CPU J = JCP-CPU U = USER-CPU * = CICS-CPU :elitdata.
An asterisk (*) represents the combined total when it exceeds the graphed total due to truncation of stacked values.
108
CICS paging rate/second
CICS paging rate/second

This report shows histograms representing the paging activity caused by CICS execution.
Data
One day's worth of input is used from the CMRSTATS data set. The following control statements are specified:
s s s s s s s s
CICS reports on CICS system data. INPUT=TAPE uses sequential CMRSTATS data set. DATA=PAGING uses CICS paging data. MINUTES=60 accumulates data for 60-minute periods. FROM DATE=09/17/1999 selects data starting on September 17, 1999. TO DATE=09/17/1999 does not select data after September 17, 1999. SCALE=10 sets the histogram scale to 10 seconds. END ends the report request.
The report produced by these statements is shown in Figure 25. Figure 25

:litdata. MAINVIEW FOR CICS LICENSED PROPERTY OF BMC SOFTWARE CMRCHRT CICS PAGING RATE/SEC. (PAGE-IN + PAGE-OUT) PAGE NUMBER 1 FROM DATE: 09/27/1999 TO: 09/27/1999 MINUTES.: 00060 CALCULATION TYPE.: RATE RUN DATE.....: 09/17/1999 FROM TIME: 00:00:00 TO: 24:00:00 BREAK...: DAILY AVERAGE VALUE....: 0.00 RUN TIME.....: 18:53:18 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 MM/DD/YYYY HH:MM =RATE/SEC.= |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 09/17/1999 07:39 0.00 | + + + + + + + + + + 09/17/1999 08:39 0.00 | + + + + + + + + + + 09/17/1999 09:39 0.02 |I* + + + + + + + + + + 09/17/1999 10:39 0.09 |IIIIIIOOO+ + + + + + + + + + 09/17/1999 12:04 0.30 |IIIIIIIIIIIIIIIIIIOOOOOOOOOOO* + + + + + + + 09/17/1999 13:04 0.16 |IIIIIIIIIIOOOOO* + + + + + + + + + 09/17/1999 14:05 0.32 |IIIIIIIIIIIIIIIIIIIIOOOOOOOOOOOO + + + + + + + 09/17/1999 15:05 0.05 |IIIOO + + + + + + + + + + 09/17/1999 16:05 0.11 |IIIIIIOOOOO + + + + + + + + + 09/17/1999 17:05 0.00 | + + + + + + + + + + 09/17/1999 18:05 0.00 | + + + + + + + + + + MM/DD/YYYY HH:MM =RATE/SEC.= |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 LEGEND: I = CICS-PAGE-IN O = CICS-PAGE-OUT * = CICS-PAGE-IN + CICS-PAGE-OUT :elitdata.
CICS paging rate/second (page-in + page-out)
An asterisk (*) represents the combined total when it exceeds the graphed total due to truncation of stacked values.
109
CICS transaction rate/second

This report shows histograms representing the average transaction rate per second over the selected interval.
Data
By default, all records on the CMRSTATS data set are used as input to produce histogram reports. The following control statements are specified:
s s s s s s s
CICS reports on CICS system data. INPUT=TAPE uses sequential CMRSTATS data set. DATA=TASKRATE reports transaction rate per second. CALC=AVERAGE calculates averages. MINUTES=60 accumulates data for 60-minute periods. SCALE=10 sets the histogram scale to 10 seconds. END ends the report request.
The report produced by these control statements is shown in Figure 26. Figure 26
:litdata. MAINVIEW FOR CICS LICENSED PROPERTY OF BMC SOFTWARE CMRCHRT CICS TRANSACTION RATE/SECOND FROM DATE: 08/01/1999 TO: 08/31/1999 MINUTES.: 00060 CALCULATION TYPE.: RATE FROM TIME: 00:00:00 TO: 24:00:00 BREAK...: DAILY AVERAGE VALUE....: 0.02 PAGE NUMBER 1 RUN DATE.....: 09/03/1999 RUN TIME.....: 07:13:49
MM/DD/YYYY 08/25/1999 08/25/1999 08/25/1999 08/25/1999 08/25/1999 08/25/1999 08/25/1999 08/25/1999 08/25/1999 MM/DD/YYYY
HH:MM 09:28 10:28 11:28 12:28 13:28 14:28 15:28 16:28 17:28 HH:MM
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 =RATE/SEC.= |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 0.3 |TTT + + + + + + + + + + 0.2 |TT + + + + + + + + + + 0.2 |TT + + + + + + + + + + 0.5 |TTTTT + + + + + + + + + + 0.3 |TTT + + + + + + + + + + 0.2 |TT + + + + + + + + + + 0.2 |TT + + + + + + + + + + 0.4 |TTTT + + + + + + + + + + 0.2 |TT + + + + + + + + + + =RATE/SEC.= |---------+---------+---------+---------+---------+---------+---------+---------+---------+---------+ 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 * = TOTAL-TRANSACTION-RATE
LEGEND: T = TERMINAL-ATTACHED-TRANSACTION-RATE :elitdata.
110
Chapter
Service Level Analysis report

The Service Level Analysis report shows the distribution of transaction response times over a range of service levels (specified by the SERVICE statement in the CMRSOPT table). This report can be used to monitor the hourly distribution of transactions response times. JCL and a sample Service Level Analysis report are described in the following sections.
Creating a Service Level Analysis report

Figure 27 shows the JCL to execute the CMRLTXR program as a batch job and create the Service Level Analysis report. Figure 27 Service Level Analysis reportinput JCL
//jobname JOB user parameters /*JOBPARM user parameters //CMRLTXR EXEC PGM=CMRLTXR //JOBLIB DD DISP=SHR,DSN=CMR.CMRV5.BBLINK //CMRSTAT DD DISP=SHR,DSN=CMR.CMRV5.CMRSTATS //REPORT DD SYSOUT=* //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,(1,1)) //SYSPRINT DD SYSOUT=* //SYSOUT DD SYSOUT=* //SYSIN DD *
The JCL statements are as follows:

s
JOB initiates the job. EXEC specifies the name of the program (PGM=CMRLTXR) that generates the Service Level Analysis report. //JOBLIB defines the program library containing the MAINVIEW for CICS load modules (BBLINK).
Chapter 6
111
Control statements
//CMRSTAT defines the VSAM or sequential data set containing records of CICS operational statistics (CMRSTATS). //REPORT defines the report output for the predefined reports. //SORTWKnn identifies one to nine work data sets that can be defined for data sorting. Define nn as a numeric value from 01 to 09. //SYSPRINT defines the output class for a sort utility. The name of the DD statement is determined at CICS system generation. //SYSOUT defines the output class. //SYSIN defines optional control statements.
Control statements
Control statements identify the CICS regions and date of the input records that create the Service Level Analysis report. Valid control statements are as follows:
TARGET=cicsRegion DATE=mm/dd/yyyy name of the CICS region whose input records are used to create the report. date the records that are used in the report were created You can specify the starting date by using a relative format that is an offset from the current date. For example, to start the selected records with yesterdays date, specify DATE=*-1
NOTE
Sample report
Figure 28 on page 113 shows an example of the Service Level Analysis report. Transactions are shown in hourly intervals by counts and percentage for each service level range.
112
Field descriptions
Figure 28

MAINVIEW for CICS SERVICE LEVEL ANALYSIS
RUN DATE: 09/17/1999 RUN TIME: INTVL DATE 18:53:39 0.0 INTVL TIME TOTAL TRANS. 1807 1547 1547 85.6 85.6 29 24 24 82.8 82.8 3 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 0 0 0.0 0.0 1 1 33.3 33.3 0 1 0.0 33.3 1 2 33.3 66.6 0 2 0.0 66.6 0 2 0.0 66.6 0 2 0.0 66.6 0 2 0.0 66.6 0 2 0.0 66.6 3 27 10.3 93.1 0 27 0.0 93.1 0 27 0.0 93.1 0 27 0.0 2 29 6.9 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 0 29 0.0 214 1761 11.8 97.4 28 1789 1.5 98.9 9 1798 0.5 99.4 6 1804 0.3 99.7 2 1806 0.0 99.8 0 1806 0.0 99.8 0 1806 0.0 99.8 1 1807 0.1 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9 THRU 0.5 0.5 THRU 1.0 1.0 THRU 1.5 1.5 THRU 2.0 2.0 THRU 2.5 2.5 THRU 3.0 3.0 THRU 4.0 4.0 THRU 5.0 5.0 THRU 6.0 6.0 THRU 7.0 7.0 THRU 8.0 8.0 THRU 9.0 9.0 THRU 10.0 10.0 THRU 20.0 20.0 THRU 30.0
PAGE: CMRLTXR 30.0 THRU 60.0 60.0 THRU 90.0
1 90.0 AND OVER
-------- ----- ------- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- ----09171999 16:01 SERVICE COUNTS : CUMULATIVE CNTS: SERVICE PERCENT: CUMULATIVE PCTS: 09171999 17:01 SERVICE COUNTS : CUMULATIVE CNTS: SERVICE PERCENT: CUMULATIVE PCTS: 09171999 18:01 SERVICE COUNTS : CUMULATIVE CNTS: SERVICE PERCENT: CUMULATIVE PCTS: 0 1807 0.0 99.9 0 1807 0.0 99.9 0 1807 0.0 99.9
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0 29 0.0 0 29 0.0 0 29 0.0
93.1 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0 2 0.0 66.6 1 3 33.3 99.9 0 3 0.0 99.9
Field descriptions
Table 12 describes the fields of the Service Level Analysis report shown in Figure 28: Table 12
Field INTVL DATE INTVL TIME TOTAL TRANS. 0.0 THROUGH 0.5 90.0 AND OVER
Service Level Analysis report fields

Description gregorian date of the data in mm/dd/yyyy format time stamp of the data in hh:mm format total number of terminal attached transactions executed during this time interval service level ranges in seconds You can adjust service level ranges (see the description of the CMRSOPT macro in the MAINVIEW for CICS Customization Guide for additional information)
SERVICE COUNTS CUMULATIVE CNTS SERVICE PERCENT CUMULATIVE PCTS
number of transactions completed within the service level specified in the column heading cumulative number of transactions completed within the service level specified in the column heading percentage of all transactions completed within the service level specified in the column heading cumulative percentage of transactions completed within the service level specified in the column heading
Chapter 6
113
Field descriptions
114
Part
Part 3

This part describes how to use PERFORMANCE REPORTER and MAINVIEW for CICS detail transaction records to create COBOL and SAS reports. This part presents the following topics: Chapter 7, COBOL reports Chapter 8, SAS reports
Part 3
115
116
Chapter
COBOL reports
The PERFORMANCE REPORTERs CMRFRED and CMRFREDN facilities allow COBOL programs to access transaction records stored on the CMRDETL data set. CMRFREDN and CMRFREDN read CMRDETL transaction records and reformat the fields into a COBOL-compatible format. The reformatted records are then passed back to the COBOL program that originally requested the data. You can produce either standard or custom COBOL reports from the reformatted transaction data.
s
CMRFRED decompresses, processes and creates all records at the F6 level formatted for COBOL by BBSAMP member CMRCOBCP CMRFREDN decompresses, processes and creates all records at the F7 level formatted for COBOL by BBSAMP member CMRCOBCN
See Transaction detail data (CMRDETL) on page 28 for more information. In addition to the CMRFRED and CMRFREDN facilities, the MAINVIEW for CICS BBSAMP data set contains sample COBOL interface programs. These programs create a variety of predefined MAINVIEW for CICS reports after they have been edited to meet your sites conditions. These programs must be compiled and linked before they can be run. You should be able to modify these sample programs if you have some COBOL programming experience. This chapter
s
describes how CMRDETL transaction records are accessed and reformatted provides JCL examples that explain how to compile, link, and run sample COBOL programs explains how to modify sample programs to create customized COBOL reports explains how to take advantage of the expand time fields introduced with CICS/TS 3.2
Chapter 7 COBOL reports 117
Accessing detail transaction records
describes the controls statements used to create COBOL reports provides an example of each predefined report and a description of its fields
NOTE
All COBOL report programs must be compiled again and link-edited again using the current version of the MAINVIEW for CICS libraries. This step ensures that the copy books for the current Type 6E record format are incorporated into the programs.
Accessing detail transaction records

All data fields in detail transaction records are accessible to COBOL programs. CMRDETL record fields are described in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. The following COBOL subprograms access CMRDETL records:
s
CMRFRED converts CMRDETL record fields to a recognizable COBOL format. CMRFRED calls the CMRCMPR program to decompress the records that are compressed in the CMRDETL data set. CMRCOBCP is a BBSAMP member that describes the COBOL format of the converted CMRDETL records.
The following sections describe these COBOL subprograms.
CMRFRED and CMRFREDN programs

The CMRFRED sand CMRFREDN programs
s
reads CMRDETL records from a VSAM or sequential data set The maximum input record block size is 32K. The input data set can contain any record types produced by MAINVIEW for CICS, but only CMRDETL records are passed to the calling COBOL program.
reformats CMRDETL record fields into a COBOL-compatible form
118
CMRFRED and CMRFREDN programs
When reading records from the CMRDETL data set, CMRFRED and CMRFREDN
s s
converts record fields to a COBOL-compatible format passes a return code back to the calling COBOL program
Parameters
Two parameters are passed from a COBOL program to the CMRFRED and CMRFREDN programs:
s s
The first parameter defines the function to be performed. The second parameter defines the CMRDETL record to be passed back to the calling program.
CMRFRED and CMRFREDN FT messages

The CMRFRED and CMRFREDN programs generate FT messages. You can display descriptions of the messages by using the General Service, MESSAGES service, or by browsing the MAINVIEW for CICS BBMLIB data set. CMRFRED and CMRFREDN generate any of the following FT messages:
s
FT256 cancels the job if a VSAM open, close, or read operation receives a non-zero return code. FT257 cancels the job when the COBOL function passed to CMRFRED and CMRFREDN is invalid a read or close operation is requested and the file is closed the function is open and the file is open
FT258 cancels the job because of an end-of-file error. FT260 cancels the job if program CMRFRED and CMRFREDN cannot load program CMRCMPR. FT261 cancels the job if program CMRCMPR passes a nonzero return code to CMRFRED and CMRFREDN.
119
CMRCOBCP and CMRCOBCPN (COBOL copybook) programs
CMRCOBCP and CMRCOBCPN (COBOL copybook) programs

The COBOL copybook contains the formats of the records held in the CMRDETL data set. CMRFRED and CMRFREDN passes the CMRDETL records to the calling COBOL program. The COBOL copybook is in member CMRCOBCP and CMRCOBCPN of the BBSAMP data set. CMRCOBCP and CMRCOBCPN define three COBOL 01 data structures:
s s s
CMR-CONTROL CMRDETL COPYRIGHT
The purpose of these data structures is explained in the following sections.
CMR-CONTROL
The CMR-CONTROL data structure passes control fields between CMRFRED/CMRFREDN and the executing COBOL program. CMR-CONTROL contains the following fields:
s s
The Status field is defined with a COBOL 88-level name to indicate end of file. The Function field defines the operation to be performed in CMRFRED and CMRFREDN.
CMRDETL
CMRDETL contains the COBOL-compatible definitions of the record fields. The COBOL versions of CMRDETL records use the same field names described in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference.
NOTE
Because of COBOLs processing requirements, the definitions of the data types associated with the record fields vary from the descriptions given in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference.
COBOL CMRDETL field formats are as follows:

Field Numeric Time Format packed decimal seconds
120
Customizing sample COBOL reports
Field TYPE T6EQUAL
Format two bytes: 6D or 6E two bytes, packed decimal
COPYRIGHT
The COPYRIGHT data structure contains a copyright statement. The copyright also appears in the compiled COBOL object program.
Customizing sample COBOL reports

You can fulfill virtually any need your site has for specialized reports by using the COBOL interface. The sample COBOL programs use batch processing to create CICS performance reports. You can use the sample programs to
s s
generate the sample reports by using your data develop your own COBOL programs
When you understand how these sample COBOL programs create reports, you can write your own programs to create CICS performance reports from CMRDETL records.
NOTE
Use of these sample programs requires an understanding of COBOL programming.
The following sections explain how to modify a sample COBOL program, including
s s s
copying and modifying the sample program compiling and linking the program executing the program
121
Copying and modifying the program
Copying and modifying the program

To modify a sample COBOL program:
1 Copy the sample program. (Sample COBOL programs are listed in Table 13 on
page 124.)
2 Copy the standard compile and link JCL from your library.
If you do not have standard compile and link JCL, member CMRCOBJ1 in your BBSAMP data set contains samples of compile and link JCL that can be used with the sample COBOL programs.
3 Modify the compile and link JCL according to your site standards.
Compiling and linking the program

To compile and link the program that you modified:
1 Submit the JCL to compile and link the sample program. 2 Verify successful compilation and linking. 3 If you have any JCL errors, make the corrections and resubmit the job. NOTE
Be sure to compile and link the program using the current version of the MAINVIEW for CICS libraries. This step ensures that the copy books for the current Type 6E record format are incorporated into the programs.
Running the program

To run the program:
1 Copy the sample JCL from member CMRCOBJ2. (The JCL is explained in
CMRCOBJ2sample execution JCL on page 125.)
2 Modify the JCL according to your site standards. 3 Enter the control statements you need to limit the input to the program (on the
CARDIN DD statement), using your site standards.
122
Sample programs
4 Submit the job to run the sample program. 5 Verify successful completion. 6 If you have any JCL errors, make the corrections and resubmit the job. NOTE
Be sure to run the program using the current version of the MAINVIEW for CICS libraries.
Sample programs
Two types of sample COBOL programs exist: transaction and file analysis. The transaction programs create reports that list performance statistics by the transactions recorded in a CICS region:
s
Performance Analysis reports contain CICS performance statistics that are summarized with control breaks by transaction ID program name terminal ID node name user ID
Rogues Report lists transactions with the longest response times. Abended Transaction Report lists all transactions that abended in a CICS region.
The file analysis program has control breaks based on the file name field of transaction records. This program creates the File Analysis for All Files report, which lists wait time for CICS files by CICS region. Table 13 identifies the sample COBOL programs and their associated reports. The table also includes a reference to a section in this book that shows an example of the COBOL report.
123
Sample programs
Table 13
Sample COBOL programs

Sample library member/program name CMRCOB11 CMRCOB11N Description Reference
Sample report Performance Analysis for All Transactions
produces a report that Performance Analysis for evaluates task performance All Transactions report by transaction ID (CMRCOB11 and CMRCOB11N) on page 128 Performance Analysis by produces a report with control breaks based on the Program report (CMRCOB12 and T6EPGNM field CMRCOB12N) on page 129 It is useful for evaluating task performance when a Fourth Generation Language (4GL) is used. Performance Analysis by produces a report that evaluates task performance Terminal report by terminal ID (CMRCOB13 and CMRCOB13N) on page 130 produces a report that evaluates task performance by VTAM symbolic node name Performance Analysis by Netname report (CMRCOB14 and CMRCOB14N) on page 131
Performance Analysis for All Programs
CMRCOB12 CMRCOB12N
Performance Analysis by Terminal
CMRCOB13 CMRCOB13N
Performance Analysis by Netname
CMRCOB14 CMRCOB14N
Performance Analysis for All Users
CMRCOB15 CMRCOB15N
produces a report that Performance Analysis for evaluates task performance All Users report by user ID (CMRCOB15 and CMRCOB15N) on page 132 produces a report that evaluates file performance by file name produces a report that lists the 100 tasks that have the longest response times File Analysis for All Files report (CMRCOB21 and CMRCOB21N) on page 133 Rogues report (CMRCOB23 and CMRCOB23N) on page 135
File Analysis for All Files
CMRCOB21 CMRCOB21N
Rogues Report
CMRCOB23 CMRCOB23N
Abended CMRCOB25 Transaction Report CMRCOB25N
produces a report that lists Abended Transaction all transactions that suffered report (CMRCOB25 and an abend CMRCOB25N) on page 136
124
CMRCOBJ2sample execution JCL

The JCL shown in Figure 29 is in member CMRCOBJ2 of your BBSAMP data set. CMRCOBJ2 runs the sample COBOL programs described in Table 13 on page 124. You must include the program name with the EXEC statement to run a specific sample report program or to run your own COBOL report program. Figure 29 JCL Execution statements for all COBOL reports
//* //* CMRCOBJ2 //* //* THIS IS SAMPLE EXECUTION JCL. //* THIS JCL MUST BE MODIFIED BEFORE USE. //* THE LOAD LIBRARY MUST BE DEFINED IN THE STEPLIB DD STATEMENT. //* THE CMRDETL FILE MUST BE DEFINED. //* THE CMRDETL FILE MAY BE A VSAM OR SEQUENTIAL DATA SET. //* THE CMRDETL FILE HAS A MAXIMUM BLOCK SIZE OF 32K. //* //* EXEC //CMRCOB00 EXEC PGM=CMRCOB00,REGION=3096K //STEPLIB DD DSN=SYS1.VSCOBII.SYS.COB2LIB,DISP=SHR // DD DSN=.....BBLINK,DISP=SHR //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,(5,5)) //SORTWK02 DD UNIT=SYSDA,SPACE=(CYL,(5,5)) //SORTWK03 DD UNIT=SYSDA,SPACE=(CYL,(5,5)) //CMRDETL DD DISP=SHR,DSN=CMR.CMRV5.CMRDETL //CARDIN DD * //REPORT DD SYSOUT=* //SYSABOUT DD SYSOUT=* //SYSOUT DD SYSOUT=* //SYSPRINT DD SYSOUT=* //
125
The sample JCL contains the following statements:

s
JOB initiates the job. EXEC specifies the name of the program that creates the report (PGM=).
Generated Report Transaction Analysis for All Transactions Transaction Analysis by Program Transaction Analysis by Terminal Transaction Analysis by Node Name Transaction Analysis by User ID File Analysis for All Files Rogues Report Abended Transaction
Program Name CMRCOB11 CMRCOB12 CMRCOB13 CMRCOB14 CMRCOB15 CMRCOB21 CMRCOB23 CMRCOB25
//STEPLIB defines your program library and the MAINVIEW for CICS load modules. //SORTWK01 defines the first sort area data set. //SORTWK02 defines the second sort area data set. //SORTWK03 defines the third sort area data set. //CMRDETL defines VSAM or sequential data sets containing T6E detail transaction records. //CARDIN defines PERFORMANCE REPORTER control statements to use as input to the report program. Some examples of control statements are as follows:
Description provides the CICS region name (for example, CICS=cicsprod) provides the start date for processing in the format yyyymmdd (for example, DATE=19990917) provides the start time for processing in the format hhmmss (for example, TIME=081133)
Control Statement CICS= DATE= TIME=
//REPORT defines the report output for the predefined reports. //SYSABOUT defines the abbreviated dump data set. //SYSOUT defines the output class.
126
COBOL report fields
//SYSPRINT defines the output class for the sort utility. The name of the DD statement is determined at sort system generation.
COBOL report fields

The following column headings appear in all sample COBOL reports described in this chapter:
s
CICS REGION (name of the CICS region associated with the data) RUN DATE (gregorian date associated with the data in yyyy/mm/dd format) TRAN CODE (transaction identification of the T6ECICS task) PROGRAM NAME (name of the program associated with the data) TERM ID (identification number of the terminal) NETNAME (name of the net node) USER ID (identification of the CICS user) FILE NAME (name of the file being accessed) RUN TIME (time associated with the data in hh:mm:ss format) ABEND CODE (code associated with this abended transaction) TRAN COUNT (number of transactions executed from this terminal) AVERAGE CPU REAL (average seconds of CPU real time used by each transaction) AVG FILE WAIT TIME (average seconds that a transaction waits for called files) AVG FILE CALLS (average number of times files were called for each transaction) AVG DB WAIT TIME (average time that a transaction waited for access to a
database)
s
AVG DB CALLS (average number of calls to the database for each transaction) AVERAGE STRG HWM (average storage high-water mark in bytes during this time
interval)
s
STORAGE OCCUPANCY (usage of storage over time)
127
Performance Analysis for All Transactions report (CMRCOB11 and CMRCOB11N)
A value of 1 is equivalent to 250 bytes held for .25 seconds. This value can be used to evaluate the impact on available storage caused by a transaction.
s
AVG PROG STRG HWM (average program storage high-water mark reached, in
bytes, for all transactions)
Performance Analysis for All Transactions report (CMRCOB11 and CMRCOB11N)

The CMRCOB11 and CMRCOB11N sample programs produce the Performance Analysis for All Transactions report. This report has control breaks based on the T6ETRID (transaction ID) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate task performance subdivided by transaction.
128
Sample report
Sample report
Figure 30
_ PROGRAM: CMRCOB11
Performance Analysis for All Transactions report

PERFORMANCE ANALYSIS FOR ALL TRANSACTIONS DATE: 1999/09/17 TIME: 16:06:33 PAGE: 1 AVERAGE STORAGE AVG PROG STRG HWM OCCUPANCY STRG HWM -------- --------- -------8,360 4,913 5,744 6,000 6,568 5,976 6,924 5,744 6,416 10,314 6,856 81,704 79,683 9,717 8,360 4,905 5,332 6,000 6,689 5,793 8,481 80,192 6,040 6,446 10,312 6,340 82,549 90,923 13,513 12,815 12,815 31 0 78 131 85 5 265 29 1 3 0 3,579 3,612 137 31 1 42 132 10 2 18 4,004 34 12,337 5 0 1,356 1,624 5,239 4,301 4,301 16,232 752 784 544 1,632 3,016 7,384 3,664 7,752 8,337 4,616 152,952 190,087 12,872 16,232 752 784 544 1,632 3,096 7,408 150,936 3,664 7,746 8,330 3,856 152,962 197,186 20,604 19,183 19,183
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS RUN
TRAN TRAN RUN DATE CODE COUNT ---------- ------- ------1999/09/17 MATD 1999/09/17 PC00 1999/09/17 RXD2 1999/09/17 TSFU 1999/09/17 TSGM 1999/09/17 MC01 1999/09/17 MC02 1999/09/17 MC00 1999/09/17 DSNC 1999/09/17 FITX 1999/09/17 NN02 1999/09/17 MCP2 1999/09/17 MC00 1999/09/17 *TOTAL* 1999/09/23 MATD 1999/09/23 PC00 1999/09/23 RXD2 1999/09/23 TSFU 1999/09/23 TSGM 1999/09/23 MC01 1999/09/23 MC02 1999/09/23 CUA 1999/09/23 TERE 1999/09/23 DSNC 1999/09/23 FITX 1999/09/23 NN01 1999/09/23 MCP2 1999/09/23 MC00 1999/09/23 *TOTAL* *TOTAL* *TOTAL* 1 161 1 1 1 6 2 1 311 161 1 1 24 672 1 738 2 1 7 22 5 1 1 1,236 738 2 6 223 2,983 3,655 3,655
AVERAGE RESPONSE -------.369 .017 .937 1.570 .986 .056 .323 .354 .015 .041 .002 5.066 3.177 .150 .373 .025 .512 1.587 .100 .034 .189 5.448 .401 3.159 .055 .002 1.937 1,336 1.436 1.200 1.200
AVERAGE CPU REAL --------.018 .001 .007 .008 .007 .006 .007 .038 .001 .001 .001 .013 .048 .003 .019 .001 .004 .008 .002 .001 .004 .032 .084 .001 .002 .000 .027 .036 .004 .004 .004
AVG FILE WAIT TIME --------.000 .000 .000 .000 .000 .000 .000 .000 .000 .014 .000 .018 .004 .003 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .019 .000 .004 .005 .005 .004 .004
AVG FILE CALLS -------.000 .000 .000 .000 .000 .000 .000 .000 .000 2.180 .000 1.000 1.125 .563 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 3.699 .000 1.000 1.206 1.007 .925 .925
AVG DB WAIT TIME --------.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 1.581 .056 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .676 .050 .051 .051
AVG DB CALLS -------.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 12.250 .437 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 7.156 .535 .517 .517
Performance Analysis by Program report (CMRCOB12 and CMRCOB12N)

The CMRCOB12 and CMRCOB12N sample programs produce the Performance Analysis by Program report. This report has control breaks based on the T6EPGNM (program name) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate task performance subdivided by program name.
NOTE
This report is most useful when a fourth-generation language (4GL) is used.
129
Sample report
Sample report
Figure 31
PROGRAM: CMRCOB12
Performance Analysis by Program report

PERFORMANCE ANALYSIS FOR ALL PROGRAMS DATE: 1999/09/24 TIME: 16:19:46 PAGE: 1 AVERAGE STORAGE AVG PROG STRG HWM OCCUPANCY STRG HWM -------- --------- -------6,856 10,314 4,913 5,744 6,000 6,568 6,924 5,976 8,360 5,744 6,416 79,764 9,717 6,340 10,312 4,905 5,332 6,000 6,689 8,481 5,793 8,360 6,040 6,446 90,658 13,513 12,815 12,815 0 3 0 78 131 85 265 5 31 29 1 3,610 137 0 5 1 42 132 10 18 2 31 34 12,337 1,628 5,239 4,301 4,301 4,616 8,337 752 784 544 1,632 7,384 3,016 16,232 3,664 7,752 188,601 12,872 3,856 8,330 752 784 544 1,632 7,408 3,016 16,232 3,664 7,746 195,831 20,604 19,183 19,183
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS RUN
RUN DATE ---------1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 *TOTAL* *TOTAL*
PROGRAM NAME -------CMRJRNL CMRROLL DFHCCMF DFHCRQ DFHFCU DFHGMM DFHSNP DFHTPQ DFHZATD DFHZXRE DSNCCOM1 T2CINIT *TOTAL* CMRJRNL CMRROLL DFHCCMF DFHCRQ DFHFCU DFHGMM DFHSNP DFHTPQ DFHZATD DFHZXRE DSNCCOM1 T2CINIT *TOTAL*
TRAN COUNT ------1 161 161 1 1 1 2 6 1 1 311 25 672 2 738 738 2 1 7 5 22 1 1 1,236 230 2,983 3,655 3,655
AVERAGE RESPONSE -------.002 .041 .017 .937 1.570 .986 .323 .056 .369 .354 .015 3.252 .150 .002 .055 .025 .512 1.587 .100 .189 .034 .373 .401 3.159 1.369 1.436 1.200 1.200
AVERAGE CPU REAL --------.001 .001 .001 .007 .008 .007 .007 .006 .018 .038 .001 .050 .003 .000 .002 .001 .004 .008 .002 .004 .001 .019 .084 .001 .035 .004 .004 .004
AVG FILE WAIT TIME --------.000 .014 .000 .000 .000 .000 .000 .000 .000 .000 .000 .005 .003 .000 .019 .000 .000 .000 .000 .000 .000 .000 .000 .000 .005 .005 .004 .004
AVG FILE CALLS -------.000 2.180 .000 .000 .000 .000 .000 .000 .000 .000 .000 1.120 .563 .000 3.699 .000 .000 .000 .000 .000 .000 .000 .000 .000 1.196 1.007 .925 .925
AVG DB WAIT TIME --------.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 1.517 .056 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .656 .050 .051 .051
AVG DB CALLS -------.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 11.760 .437 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 6.939 .535 .517 .517
Performance Analysis by Terminal report (CMRCOB13 and CMRCOB13N)

The CMRCOB13 and CMRCOB13N sample programs produce the Performance Analysis by Terminal report. This report has control breaks based on the T6ETMID (terminal ID) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate task performance subdivided by terminal.
130
Sample report
Sample report
Figure 32
PROGRAM: CMRCOB13 CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS
Performance Analysis by Terminal report

PERFORMANCE ANALYSIS BY TERMINAL 10:27 FRIDAY, SEPTEMBER 20, 1999 AVG DB WAIT TIME --------.000 .334 .785 .147 .000 .924 .079 .000 .172 .000 .003 .000 1.734 .115 .595 .186 .037 .000 1.355 .056 .000 1.034 .077 .705 .538 1.198 .000 .786 .050 AVG DB CALLS -------.000 9.391 7.629 9.285 .000 10.000 2.224 .000 1.367 .000 .030 .000 6.640 .818 3.000 4.558 .463 .000 10.500 .437 .000 9.150 .777 5.296 6.802 10.615 .000 3.777 .535 AVERAGE STORAGE STRG HWM OCCUPANCY -------- --------7,384 7,151 98,364 920 74,227 2,072 92,671 598 23,130 4,281 98,348 6,003 28,825 5,667 7,061 6,831 66,515 872 45,788 44 8,417 6,693 7,086 4,234 70,617 3,301 35,663 3,181 52,203 1,623 82,966 598 14,322 3,911 7,012 2 71,947 3,245 9,717 137 7,066 5,565 85,640 2,278 54,978 814 82,023 2,079 93,258 1,355 59,868 2,210 47,016 2,182 70,284 1,797 13,513 5,239 AVG PROG STRG HWM -------6,899 198,420 167,457 183,859 19,422 96,360 52,040 6,010 173,636 77,292 9,802 5,915 171,156 82,848 89,860 175,692 22,368 6,084 168,980 12,872 5,973 178,929 137,496 172,496 198,442 150,362 77,292 160,321 20,604 1
RUN DATE ---------1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/18 1999/09/18 1999/09/18 1999/09/18 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/17 1999/09/17 1999/09/17 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23
TERM TRAN AVERAGE ID COUNT RESPONSE ------- ------- --------3,582 4.387 E202 1,008 .743 E210 27 1.811 S211 84 .678 S214 4 1.706 TERC 2 6.312 *TOTAL* 4,707 3.524 3,442 13.937 X002 79 1.123 X003 2 .051 *TOTAL* 3,523 13.642 1,986 13.340 S201 25 3.457 S202 11 .333 TN03 8 1.427 X003 181 .620 *TOTAL* 2,211 12.079 644 .027 E209 28 2.962 *TOTAL* 672 .150 2,741 1.447 CZE3 20 2.112 E206 9 1.168 E207 17 1.924 S204 163 .996 S210 13 2.649 X002 2 3.125 X003 18 1.582 *TOTAL* 2,983 1.436
AVERAGE CPU REAL -------.001 .042 .036 .054 .470 .087 .012 .001 .012 .004 .001 .001 .040 .007 .014 .019 .004 .001 .046 .003 .001 .036 .018 .032 .035 .028 .063 .026 .004
AVG FILE WAIT TIME --------.008 .000 .013 .013 .000 .000 .007 .004 .004 .000 .004 .006 .010 .014 .000 .016 .007 .003 .004 .003 .005 .016 .019 .008 .001 .016 .014 .013 .005
AVG FILE CALLS -------1.770 1.083 1.518 1.511 .000 .000 1.615 .535 1.101 .500 .548 .898 1.200 1.272 .375 1.718 .969 .545 1.000 .563 .996 1.650 1.777 1.411 1.036 .615 .500 1.333 1.007
Performance Analysis by Netname report (CMRCOB14 and CMRCOB14N)

The CMRCOB14 and CMRCOB14N sample programs produce the Performance Analysis by Netname report. This sample report has control breaks based on the T6ENETNM (network node name) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate task performance subdivided by VTAM symbolic node name.
131
Sample report
Sample report
Figure 33
PROGRAM: CMRCOB14
Performance Analysis by Netname report

PERFORMANCE ANALYSIS BY NETNAME DATE: 1999/09/24 TIME: 16:27:44 PAGE: 1 AVERAGE STORAGE AVG PROG STRG HWM OCCUPANCY STRG HWM -------- --------- -------7,061 66,515 45,788 8,417 7,086 52,203 82,966 70,617 35,663 14,322 7,012 71,947 9,717 7,066 47,016 70,284 85,640 54,978 82,023 93,258 59,868 13,513 11,519 11,519 6,831 872 44 6,693 4,234 1,623 598 3,301 3,181 3,911 2 3,245 137 5,565 2,182 1,797 2,278 814 2,079 1,355 2,210 5,239 5,107 5,107 6,010 173,636 77,292 9,809 5,915 89,860 175,692 171,156 82,848 22,368 6,084 168,980 12,872 5,973 77,292 160,321 178,929 137,496 172,496 198,442 150,362 20,604 16,416 16,416
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS RUN
NETNAME RUN DATE ---------1999/09/18 1999/09/18 1999/09/18 1999/09/18 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/17 1999/09/17 1999/09/17 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 *TOTAL* *TOTAL*
TRAN AVERAGE COUNT RESPONSE -------- ------- --------3,442 79 2 3,523 1,986 8 181 25 11 2,211 644 28 672 2,741 2 18 20 9 17 163 13 2,983 9,389 9,389 13.937 1.123 .051 13.642 13.340 1.427 .620 3.457 .333 12.079 .027 2.962 .150 1.447 3.125 1.582 2.112 1.168 1.964 .994 2.649 1.436 8.430 8.430
AVERAGE CPU REAL -------.001 .022 .004 .001 .001 .014 .029 .040 .007 .004 .001 .046 .003 .001 .063 .026 .036 .018 .032 .035 .028 .004 .003 .003
AVG FILE WAIT TIME --------.004 .004 .000 .004 .006 .000 .016 .010 .014 .007 .003 .004 .003 .005 .014 .013 .016 .019 .008 .001 .016 .005 .005 .005
AVG FILE CALLS -------.535 1.101 .500 .548 .898 .375 1.718 1.200 1.272 .969 .545 1.000 .563 .996 .500 1.333 1.650 1.777 1.411 1.036 .615 1.007 .794 .794
AVG DB WAIT TIME --------.000 .172 .000 .003 .000 .596 .186 1.734 .115 .037 .000 1.355 .056 .000 .000 .786 1.034 .077 .705 .538 1.198 .050 .030 .030
AVG DB CALLS -------.000 1.367 .000 .030 .000 3.000 4.558 6.640 .818 .463 .000 10.500 .437 .000 .000 3.777 9.150 .777 5.294 6.801 10.615 .535 .321 .321
A100A02 A100A03 *TOTAL*
A051A03 A100A03 SWTCA201 SWTCA202 *TOTAL*
SWTAE209 *TOTAL*
A100A02 A100A03 LCZHA1E4 SWTAE206 SWTAE207 SWTCA204 SWTCA210 *TOTAL*
Performance Analysis for All Users report (CMRCOB15 and CMRCOB15N)

The CMRCOB15 and CMRCOB15N sample programs produce the Performance Analysis for All Users report. This sample report has control breaks based on the T6EUSER (user ID) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate task performance subdivided by user ID.
132
Sample report
Sample report
Figure 34
PROGRAM: CMRCOB15
Performance Analysis for All Users report

PERFORMANCE ANALYSIS FOR ALL USERS DATE: 1999/09/24 TIME: 12:00:46 PAGE: 1 AVERAGE STORAGE AVG PROG STRG HWM OCCUPANCY STRG HWM -------- --------- -------68,281 7,076 78,969 12,112 68,075 7,069 88,636 22,403 92,132 7,066 28,825 55,559 7,068 8,417 72,762 7,099 68,080 14,322 6,805 7,021 65,482 9,717 79,803 7,076 13,513 17,617 17,617 275 5,520 1,477 5,216 580 6,825 1,191 5,677 1,709 7,028 5,667 704 6,864 6,693 965 4,275 1,280 3,911 205 2 2,912 137 1,419 5,610 5,239 5,360 5,360 142,680 5,999 164,303 17,110 165,676 6,020 187,063 41,115 185,907 6,030 52,040 142,087 6,024 9,809 155,302 5,942 143,877 22,368 5,466 6,113 152,697 12,872 171,052 5,997 20,604 29,356 29,356
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS RUN
USER ID RUN DATE ---------1999/09/15 1999/09/15 1999/09/15 1999/09/15 1999/09/16 1999/09/16 1999/09/16 1999/09/16 1999/09/17 1999/09/17 1999/09/17 1999/09/18 1999/09/18 1999/09/18 1999/09/21 1999/09/21 1999/09/21 1999/09/21 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/23 1999/09/23 1999/09/23 *TOTAL* *TOTAL*
TRAN COUNT -------- ------________ CPS19 CMR4 *TOTAL* ________ JEN1 CMR4 *TOTAL* ________ CMR4 *TOTAL* ________ CMRG *TOTAL* ________ CMR7 CMR8 *TOTAL* ________ CMR5 CMR1 *TOTAL* ________ CMR6 *TOTAL* 33 2,678 174 2,885 186 3,402 661 4,256 1,204 3,503 4,707 98 3,425 3,523 233 1,967 11 2,211 3 638 31 672 264 2,719 2,983 21,237 21,237
AVERAGE RESPONSE --------.298 1.253 1.240 1.242 .613 13.994 .970 11.387 .733 4.483 3.524 .910 14.006 13.642 .858 13.468 1.313 12.079 .544 .027 2.634 .150 1.202 1.459 1.436 6.959 6.959
AVERAGE CPU REAL -------.030 .001 .038 .004 .021 .001 .040 .008 .042 .001 .012 .018 .001 .001 .027 .001 .016 .004 .007 .001 .042 .003 .031 .001 .004 .006 .006
AVG FILE WAIT TIME --------.002 .004 .003 .004 .005 .006 .000 .005 .005 .007 .007 .003 .004 .004 .013 .006 .014 .007 .000 .003 .004 .003 .005 .005 .005 .005 .005
AVG FILE CALLS -------1.393 1.147 1.040 1.144 1.198 2.126 .989 1.909 1.151 1.774 1.615 .897 .538 .548 1.463 .907 1.545 .969 .000 .550 .903 .563 1.041 1.004 1.007 1.247 1.247
AVG DB WAIT TIME --------.103 .000 .618 .038 .177 .000 .468 .080 .309 .000 .079 .139 .000 .003 .331 .000 .549 .037 .000 .000 1.224 .056 .571 .000 .050 .052 .052
AVG DB CALLS -------4.787 .000 9.264 .613 3.112 .000 10.028 1.693 8.698 .000 2.224 1.102 .000 .030 4.253 .000 3.000 .463 .000 .000 9.483 .437 6.045 .000 .535 1.058 1.058
File Analysis for All Files report (CMRCOB21 and CMRCOB21N)

The CMRCOB21 and CMRCOB21N sample programs produce the File Analysis for All Files report. This report has control breaks based on the T6EFN (file name) field. See Transaction detail data (CMRDETL) on page 28 for more information. Generate this report to evaluate file performance.
NOTE
A CMRDETL data set may contain two forms of T6E records, original and expanded. Any program that you run against T6E file data needs to know which type of record it is processing. CMRCOB21 and CMRCOB21N use the expanded record flag (field T6EQUAL) to determine whether a given record is expanded or not. This program illustrates how to process both original and expanded format records in the same file.
133
Sample report
Sample report
Figure 35
PROGRAM: CMRCOB21 DATE: 1999/09/24 TIME: PAGE: CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS RUN 16:36:20 1 RUN DATE ---------1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 1999/09/23 *TOTAL* *TOTAL* FILE NAME -------CMRSTAT CUDCNLOG CUISCNTL CUPFR01 CUPINIT CUPSSON DFHRPL DFHUECB M2DCNLOG M2DPCT M2ISCNTL M2PACHP M2PIB21 M2PIG21 M2PINIT M2PMC01 M2PMC06 M2PMC21 M2PMC43 M2PMD21 M2PMI03 M2PMI04 M2PRA01 M2PRA21 M2PRB01 M2PRB21 M2PRC21 M2PRD21 M2PRK50 M2PSSON *TOTAL* FILE TOTAL TOTAL TYPE TRANS WAITS -------- --------- ---------CICS FCT CICS FCT CICS FCT DB2 DB2 DB2 PSEUDO PSEUDO CICS FCT CICS FCT CICS FCT DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 148 1 2 1 3 2 241 1 1 3 12 4 2 5 65 1 4 2 2 15 3 1 1 4 2 6 1 2 1 5 541 541 541 198 2 7 50 6 82 1,829 1 2 4 27 42 34 112 167 67 61 5 8 507 59 6 18 50 14 104 17 84 4 99 3,666 3,666 3,666 AVG. WAITS PER TRAN ----------1.337 2.000 3.500 50.000 2.000 41.000 7.589 1.000 2.000 1.333 2.250 10.500 17.000 22.400 2.569 67.000 15.250 2.500 4.000 33.800 19.666 6.000 18.000 12.500 7.000 17.333 17.000 42.000 4.000 19.800 6.776 6.776 6.776 AVERAGE WAIT TIME ----------.096 .035 .088 7.246 .862 2.875 .275 .346 .052 .046 .078 .811 2.517 3.621 .172 3.647 .354 1.031 .336 2.096 1.624 1.951 .422 2.396 .071 2.566 5.219 5.057 2.383 1.682 .430 .430 .430 AVG. TIME PER WAIT ----------.071 .017 .025 .144 .431 .070 .036 .346 .026 .035 .034 .077 .148 .161 .067 .054 .023 .412 .084 .061 .082 .325 .023 .191 .010 .148 .307 .120 .596 .084 .063 .063 .063
File Analysis for All Files report

FILE ANALYSIS FOR ALL FILES
134
Rogues report (CMRCOB23 and CMRCOB23N)
Rogues report (CMRCOB23 and CMRCOB23N)

The CMRCOB23 and CMRCOB23N sample programs produce the Highest Response Times All Transactions Report. This report is also known as the Rogues Report. The report is produced by
s
reading the CMRDETL file calculating response times of tasks
NOTE
Response times under 1.000 seconds are bypassed.
producing a list of the 100 tasks with the longest response times
Generate this report to evaluate tasks with the greatest response times. Modify the report control statements in the JCL stream to limit the input for your report. See Transaction detail data (CMRDETL) on page 28 for more information.
Sample report
Figure 36
PROGRAM: CMRCOB23
COBOL rogues report

HIGHEST RESPONSE TIMES ALL TRANSACTIONS DATE: 1999/09/24 TIME: 17:03:32 PAGE: 1 STORAGE PROGRAM STRG HWM OCCUPANCY STRG HWM -------- --------- -------165,360 126,128 147,104 121,168 168,104 126,944 122,080 123,592 128,656 148,280 74,488 121,288 67,344 74,488 126,544 74,648 74,312 138,864 74,648 74,488 136,120 148,080 131,104 146,816 126,992 160,616 74,488 29,066 30,117 27,343 23,743 29,506 30,053 29,601 29,806 31,397 31,336 16,249 29,911 14,934 16,795 35,143 17,916 18,052 44,330 19,387 19,002 35,632 36,061 40,124 39,550 51,905 63,766 31,827 198,856 194,440 192,448 189,632 214,288 194,440 211,400 200,120 198,872 215,496 189,744 211,696 181,784 189,744 194,440 189,744 189,744 217,496 189,744 189,744 193,528 192,680 194,440 212,960 196,424 178,656 189,744
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS
RUN DATE ---------1999/09/17 1999/09/10 1999/09/05 1999/09/23 1999/09/17 1999/09/06 1999/09/12 1999/09/15 1999/09/17 1999/09/15 1999/09/07 1999/09/10 1999/09/08 1999/09/03 1999/09/05 1999/09/06 1999/09/03 1999/09/14 1999/09/06 1999/09/23 1999/09/14 1999/09/17 1999/09/16 1999/09/08 1999/09/23 1999/09/14 1999/09/14
TRAN RUN TIME CODE -------- -----09:06:50 12:41:21 11:48:39 10:03:28 11:58:22 10:16:29 10:33:21 08:16:53 12:39:15 14:22:13 10:15:13 12:41:50 12:59:26 15:00:35 11:18:39 10:15:15 14:53:07 13:58:44 11:00:53 10:02:22 12:29:10 08:00:34 10:29:20 02:38:47 13:40:50 11:21:51 10:59:21 MC00 MC00 MC01 MC00 MC00 MC01 MC00 MC00 JC02 JC02 MC00 MC00 MC00 JC00 JC00 MC00 MC00 MC00 MC00 MC00 PC01 PC01 PC02 PC02 MC00 MC00 MC00
RESPONSE --------17.045 17.278 17.522 17.538 17.686 17.784 17.912 17.996 18.554 18.682 19.178 19.477 19.861 20.027 20.820 21.177 21.232 22.130 22.449 22.509 22.661 24.136 24.159 25.308 30.297 33.303 37.881
CPU REAL --------.157 .105 .126 .121 .156 .123 .084 .121 .127 .118 .136 .082 .029 .133 .123 .167 .112 .112 .161 .139 .077 .105 .124 .086 .112 .173 .136
FILE WAIT TIME --------.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .076 .000 .000 .177 .000 .123 .178 .000 .305 .118 .000 .000 .000 .000 .000 .000 .136
FILE CALLS -------1.000 1.000 1.000 1.000 2.000 1.000 1.000 1.000 1.000 1.000 4.000 1.000 1.000 4.000 1.000 4.000 4.000 2.000 4.000 4.000 1.000 1.000 1.000 1.000 1.000 1.000 4.000
DB WAIT TIME --------10.664 14.631 13.545 13.143 12.921 14.172 14.355 15.335 14.967 15.834 11.469 15.437 .000 10.800 16.772 12.261 12.337 18.572 11.313 10.282 19.436 18.302 20.919 22.968 25.396 27.457 23.105
DB CALLS -------33.000 38.000 48.000 24.000 47.000 41.000 10.000 55.000 46.000 48.000 82.000 18.000 .000 82.000 42.000 82.000 81.000 40.000 82.000 81.000 19.000 49.000 54.000 19.000 38.000 60.000 81.000
135
Abended Transaction report (CMRCOB25 and CMRCOB25N)
Abended Transaction report (CMRCOB25 and CMRCOB25N)

The CMRCOB25 and CMRCOB25N sample programs produce the Abended Transaction report by
s s
reading the CMRDETL file listing all transactions with an abend code
Generate this report to list all transactions that suffered an abend during the recorded interval. Modify the report control statements in the JCL stream to limit the input for your report. See Transaction detail data (CMRDETL) on page 28 for more information.
Sample report
Figure 37
PROGRAM: CMRCOB25
Abended Transaction report

ABENDED TRANSACTION REPORT DATE: 1999/09/24 TIME: 17:04:54 PAGE: 1 STORAGE STRG HWM OCCUPANCY ---------------67,344 67,344 74,176 66,928 87,208 87,208 87,208 30,962 67,344 116,080 64,544 112,176 112,176 128,848 111,056 124,944 110,816 96,208 103,248 31,048 67,344 2,456 2,098 80 222 3,038 758 142 204,839 2,670 7,269 618 3,692 243 20,458 2,574 698 1,404 1,029,337 2,461,196 28,320 2,567
CICS REGION -------PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS PRODCICS
RUN DATE ---------1999/09/05 1999/09/15 1999/09/15 1999/09/15 1999/09/08 1999/09/08 1999/09/08 1999/09/08 1999/09/09 1999/09/15 1999/09/15 1999/09/15 1999/09/15 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/17 1999/09/18
RUN TIME -------18:09:07 13:33:20 13:41:31 15:52:03 13:02:52 14:03:19 14:53:43 15:02:38 18:19:39 08:11:48 08:19:29 08:21:03 08:37:47 13:12:22 13:17:54 13:25:39 13:34:35 14:00:19 14:34:24 14:57:22 18:46:08
TRAN ABEND CODE CODE RESPONSE ------ ------- -------AEY9 AEY9 AEY9 AEY9 QUIT QUIT QUIT ATNI AEY9 DB2E DB2E DB2E DB2E ASRA ASRA ASRA ASRA LLL LLL ATNI AEY9 MC00 3.016 MC00 2.714 MC00 .110 MC00 .321 MC00 3.171 MC00 .850 MC00 .165 FITX 1.946 MC00 3.382 MC00 4.760 MC00 .802 MC00 2.316 MC00 .209 MC00 12.296 MC00 1.699 MC00 .483 MC00 1.069 MC00 717.335 MC00 1,712.149 FITX .145 MC00 3.243
CPU REAL -------.057 .063 .014 .014 .067 .028 .027 .023 .059 .087 .023 .035 .034 .133 .089 .033 .033 .014 .075 .003 .060
FILE WAIT TIME ---------.059 .035 .000 .000 .068 .077 .052 .000 .020 .090 .087 .140 .045 .080 .052 .081 .329 .000 .650 .000 .020
FILE CALLS -----1.000 1.000 1.000 1.000 6.000 6.000 6.000 .000 1.000 6.000 6.000 6.000 6.000 6.000 6.000 6.000 6.000 .000 24.000 .000 1.000
DB WAIT TIME --------.000 .000 .000 .000 .049 .064 .006 .000 .000 1.120 .230 .064 .004 3.665 .013 .042 .033 .000 .000 .000 .000
DB CALLS -----.000 .000 .000 .000 2.000 2.000 2.000 .000 .000 1.000 3.000 1.000 1.000 12.000 5.000 12.000 10.000 .000 .000 .000 .000
136
Chapter
SAS reports
Sample programs are included in your BBSAMP data set to create SAS reports from data collected by MAINVIEW for CICS. Copies of these sample programs can be modified to create unique SAS reports. This chapter contains
s
control statements to specify the input data for SAS programs explanations about how to take advantage of the expand time fields introduced with CICS/TS 3.2 descriptions of SAS report fields examples of SAS reports
Detail transaction records are compressed when they are stored on the CMRDETL data set. You have the choice of creating SAS reports from compressed or decompressed 6E and 6D records. You must run a separate job if you want to create SAS reports with decompressed CMRDETL records. Member CMRSAS10 or CMRSAS10N of your BBSAMP data set decompresses 6E and 6F records. The decompressed output records are compatible with SAS programs. Edit a copy of CMRSAS10 or CMRSAS10Nand run the program to create decompressed detail transaction records before running your SAS report programs.
NOTE
Be sure to run your SAS report programs using the current version of the MAINVIEW for CICS libraries.
The MAINVIEW for CICS SAS reporting facility supports the F6 and F7 formats of the T6E history record. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels.
Chapter 8 SAS reports
137
Sample SAS programs
The following information is available for the F6 level history records:

s
CMRSIFUE is the SAS in-file user exit that reads and decompresses T6E records SASIN6Ex members in the BBSAMP data set are SAS formats for the various sections of the T6E record CMRSASxx members in the BBSAMP data set are sample programs CMRCMPW reads the CMRDETL records and decompresses them to the F6 format
The following information is available for the F7 level history records:

s
CMRNIFUE is the SAS in-file user exit that reads and decompresses T6E records (counterpart to CMRSIFUE) SASI6ExN members in the BBSAMP data set are SAS formats for the various sections of the T6E record (counterpart to the SASIN6Ex members) CMRSAxxN members in the BBSAMP data set are sample programs (counterpart to the CMRSASxx members) CMRCMPWN reads the CMRDETL records and decompresses them to the F7 format (counterpart to CMRCMPW)
Sample SAS programs

The interface routines in MAINVIEW for CICS give SAS programs access to records held in the CMRDETL and CMRSTATS data sets. SAS reports are created from these records. See the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. for the format of CMRDETL and CMRSTATS records. Table 14 on page 139 identifies the sample SAS programs and their associated reports. The table also includes a reference to a section in this book that shows an example of the SAS report.
138
Sample SAS programs
Table 14
Sample SAS programs (part 1 of 2)

Sample library member/program name CMRSAS11 CMRSAS11N
Sample report title Performance Analysis for All Transactions
Description produces a report with control breaks based on the T6ETRID field This report gives an evaluation of task performance subdivided by transaction.
Section reference Performance Analysis for All Transactions report (CMRSAS11 and CMRSAS11N) on page 141
Performance Analysis by Program
CMRSAS12 CMRSAS12N
Performance Analysis by Program report (CMRSAS12 and CMRSAS12N) This report is useful for evaluating task performance subdivided by T6EPGNM. This on page 142 is most helpful when a Fourth Generation Language (4GL) is used. The report is almost identical to the report produced by CMRSAS11 or CMRSAS11N produces a report with control breaks based on the T6EPGNM field produces a report with control breaks based on the T6ETMID field This report is useful for evaluating task performance subdivided by terminal. The report is almost identical to the report produced by CMRSAS11 or CMRSAS11N Performance Analysis by Terminal report (CMRSAS13 and CMRSAS13N) on page 143
Performance Analysis by Terminal
CMRSAS13 CMRSAS13N
Performance Analysis by VTAM Symbolic Node Name
CMRSAS14 CMRSAS14N
Performance Analysis by Node Name report (CMRSAS14 and This report is useful for evaluating task performance subdivided by VTAM symbolic CMRSAS14N) on node name. The report is almost identical to page 144 the report produced by CMRSAS11 or CMRSAS11N produces a report with control breaks based on the T6ENETNM field produces a report with control breaks based on the T6EUSER field This report is useful for evaluating task performance subdivided by user ID. The report is almost identical to the report produced by CMRSAS11 or CMRSAS11N Performance Analysis by User ID report (CMRSAS15 and CMRSAS15N) on page 145
Performance CMRSAS15 Analysis by User CMRSAS15N ID
File Analysis for All Files
CMRSAS21 CMRSAS21
produces a report with control breaks based on the T6EFN field This report is useful for evaluating file performance by file name.
File Analysis for All Files report (CMRSAS21 and CMRSAS21N) on page 146
Rogues Report
CMRSAS23 CMRSAS23N
produces a report that lists the tasks with the Rogues report longest response times (CMRSAS23 and CMRSAS23N) on page 147
139
Control statements
Table 14
Sample SAS programs (part 2 of 2)

Sample library member/program name CMRSAS25 CMRSAS25N
Sample report title Abended Transaction Roster
Description
Section reference
produces a report of all transactions that had Abended an abend during execution Transaction Roster report (CMRSAS25 and CMRSAS25N) on page 148 produces a report that lists overall Operational transaction performance across CICS regions statistics report (CMRSAS6F and CMRSAS6FN) on page 149
Operational Statistics Report
CMRSAS6F CMRSAS6FN
Control statements
Control statements are processed by the SAS programs to select data from specific CICS regions. You also can provide a starting date and a starting time. Use the following report control statements to select the data that appears in the SAS reports:
Control Statement CICS DATE TIME Description name of the CICS region providing data for the reports (for example, CICS=cicsprod) processing start date in yyyymmdd format (for example, DATE=19990924) processing start time in hhmmss format (for example, TIME=082314)
SAS report fields

The following table describes the columnar data shown in the SAS sample reports in this chapter: Table 15
Field CICS REGION RUN DATE
SAS report fields (part 1 of 2)

Description name of the CICS region associated with the data gregorian date associated with the data, in yyyy/mm/dd format
140
Performance Analysis for All Transactions report (CMRSAS11 and CMRSAS11N)
Table 15
Field
SAS report fields (part 2 of 2)

Description transaction identification of the CICS task name of the program associated with the data terminal identifier symbolic node name identification of the CICS user name of the file task termination time associated with the data, in hh:mm:ss format abend code of an abended transaction number of executed transactions average amount of real CPU time used by each transaction (in seconds) average time waiting for files during this transaction (in seconds) average number of file calls for each transaction average amount of wait time to access a database for each transaction average number of database calls per transaction average storage high-water mark during this time interval average storage usage in kilobytes A value of 1 is equivalent to 250 bytes held for .25 seconds. This value can be used to evaluate the impact on available storage caused by a transaction.
TRAN CODE PROGRAM NAME TERM ID NETNAME VTAM USER ID FILE NAME RUN TIME ABEND CODE TRAN COUNT AVERAGE CPU REAL AVG FILE WAIT TIME AVG FILE CALLS AVG DB WAIT TIME AVG DB CALLS AVERAGE STRG HWM STORAGE OCCUPANCY
AVG PROG STRG HWM
average program storage high-water mark for all transactions
Performance Analysis for All Transactions report (CMRSAS11 and CMRSAS11N)

The CMRSAS11 and CMRSAS11N sample programs produce the Performance Analysis for All Transactions report. This report has control breaks based on the T6ETRID (transaction ID) field. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate task performance subdivided by transaction.
141
Sample report
Sample report
Figure 38 Performance Analysis for All Transactions report
PROGRAM: CMRSAS11 PERFORMANCE ANALYSIS FOR ALL TRANSACTIONS 10:27 WEDNESDAY, SEPTEMBER 18, 1999 1 CICS TRAN TRAN AVERAGE AVERAGE AVG FILE AVG FILE AVG DB AVG DB AVERAGE STORAGE AVG PROG REGION RUN DATE CODE COUNT RESPONSE CPU REAL WAIT TIME CALLS WAIT TIME CALLS STRG HWM OCCUPANCY STRG HWM PRODCICS 1999/09/22 CATD 1 .369 .018 .000 .000 .000 .000 8,360 31 16,232 PRODCICS 1999/09/22 CCMF 161 .017 .001 .000 .000 .000 .000 4,913 0 752 PRODCICS 1999/09/22 CRSQ 1 .937 .007 .000 .000 .000 .000 5,744 78 784 PRODCICS 1999/09/22 CSFU 1 1.570 .008 .000 .000 .000 .000 6,000 131 544 PRODCICS 1999/09/22 CSGM 1 .986 .007 .000 .000 .000 .000 6,568 85 1,632 PRODCICS 1999/09/22 CSPQ 6 .056 .006 .000 .000 .000 .000 5,976 5 3,096 PRODCICS 1999/09/22 CSSN 2 .323 .007 .000 .000 .000 .000 6,924 265 7,384 PRODCICS 1999/09/22 CXRE 1 .354 .038 .000 .000 .000 .000 5,744 29 3,664 PRODCICS 1999/09/22 DSNC 311 .015 .001 .000 .000 .000 .000 6,416 1 7,752 PRODCICS 1999/09/22 FIC2 161 .041 .001 .014 2.180 .000 .000 10,314 3 8,337 PRODCICS 1999/09/22 JNL2 1 .002 .001 .000 .000 .000 .000 6,856 0 4,616 PRODCICS 1999/09/22 MCP2 1 5.066 .093 .028 1.000 .000 .000 81,704 3,579 152,952 PRODCICS 1999/09/22 MC00 24 3.177 .048 .004 1.125 1.581 12.250 79,683 3,612 190,087 PRODCICS 1999/09/22 *TOTAL* 672 .150 .003 .003 .563 .056 .437 9,717 137 12,872 PRODCICS 1999/09/23 CATD 1 .373 .019 .000 .000 .000 .000 8,360 31 16,232 PRODCICS 1999/09/23 CCMF 738 .025 .001 .000 .000 .000 .000 4,905 1 752 PRODCICS 1999/09/23 CRSQ 2 .512 .004 .000 .000 .000 .000 5,332 42 784 PRODCICS 1999/09/23 CSFU 1 1.587 .008 .000 .000 .000 .000 6,000 132 544 PRODCICS 1999/09/23 CSGM 7 .100 .002 .000 .000 .000 .000 6,689 10 1,632 PRODCICS 1999/09/23 CSPQ 22 .034 .001 .000 .000 .000 .000 5,793 2 3,096 PRODCICS 1999/09/23 CSSF 5 .189 .004 .000 .000 .000 .000 8,481 18 7,408 PRODCICS 1999/09/23 CUA 1 5.448 .032 .000 .000 .000 .000 80,192 4,004 150,936 PRODCICS 1999/09/23 CXRE 1 .401 .084 .000 .000 .000 .000 6,040 34 3,664 PRODCICS 1999/09/23 DSNC 1,236 3.159 .001 .000 .000 .000 .000 6,446 12,337 7,746 PRODCICS 1999/09/23 FIC2 738 .055 .002 .019 3.699 .000 .000 10,312 5 8,330 PRODCICS 1999/09/23 JNL2 2 .002 .000 .000 .000 .000 .000 6,340 0 3,856 PRODCICS 1999/09/23 MCP2 6 1.937 .027 .004 1.000 .000 .000 82,549 1,356 152,952 PRODCICS 1999/09/23 MC00 223 1.336 .036 .005 1.206 .676 7.156 90,923 1,624 197,186 PRODCICS 1999/09/23 *TOTAL* 2,983 1.436 .004 .005 1.007 .050 .535 13,513 5,239 20,604
Performance Analysis by Program report (CMRSAS12 and CMRSAS12N)

The CMRSAS12 and CMRSAS12N sample programs produce the Performance Analysis by Program report. This report has control breaks based on the T6EPGNM (program name) field. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate task performance subdivided by program name.
NOTE
This report is most useful when a fourth-generation language (4GL) is used.
142
Sample report
Sample report
Figure 39
PROGRAM: CMRSAS12 CICS REGION RUN DATE PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PROGRAM NAME CMRJRNL CMRROLL DFHCCMF DFHCRQ DFHFCU DFHGMM DFHSNP DFHTPQ DFHZATD DFHZXRE DSNCCOM1 T2CINIT *TOTAL* CMRJRNL CMRROLL DFHCCMF DFHCRQ DFHFCU DFHGMM DFHSNP DFHTPQ DFHZATD DFHZXRE DSNCCOM1 T2CINIT *TOTAL*
Performance Analysis by Program report

PERFORMANCE ANALYSIS BY PROGRAM AVERAGE AVERAGE AVG FILE AVG FILE RESPONSE CPU REAL WAIT TIME CALLS .002 .001 .000 .000 .041 .001 .014 2.180 .017 .001 .000 .000 .937 .007 .000 .000 1.570 .008 .000 .000 .986 .007 .000 .000 .323 .007 .000 .000 .056 .006 .000 .000 .369 .018 .000 .000 .354 .038 .000 .000 .015 .001 .000 .000 3.252 .050 .005 1.120 .150 .003 .003 .563 .002 .000 .000 .000 .055 .009 .019 3.699 .025 .001 .000 .000 .512 .004 .000 .000 1.587 .008 .000 .000 .100 .002 .000 .000 .189 .004 .000 .000 .034 .001 .000 .000 .373 .019 .000 .000 .401 .084 .000 .000 3.159 .001 .000 .000 1.369 .035 .005 1.195 1.436 .004 .005 1.007 AVG DB WAIT TIME .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 1.517 .056 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .656 .050 10:27 WEDNESDAY, SEPTEMBER 25, 1999 1 AVG DB AVERAGE STORAGE AVG PROG CALLS STRG HWM OCCUPANCY STRG HWM .000 6,856 0 4,616 .000 10,314 3 8,337 .000 4,913 0 752 .000 5,744 78 784 .000 6,000 131 544 .000 6,568 85 1,632 .000 6,924 265 7,384 .000 5,976 5 3,096 .000 8,360 31 16,232 .000 5,744 29 3,664 .000 6,416 1 7,752 11.760 79,764 3,610 188,601 .437 9,717 137 12,872 .000 6,340 0 3,856 .000 10,312 5 8,330 .000 4,905 1 752 .000 5,332 42 784 .000 6,000 132 544 .000 6,689 10 1,632 .000 8,481 18 7,408 .000 5,793 2 3,096 .000 8,360 31 16,232 .000 6,040 34 3,664 .000 6,446 12,337 7,746 6.939 90,658 1,628 195,831 .535 13,513 5,239 20,604
TRAN COUNT 1 161 161 1 1 1 2 6 1 1 311 25 672 2 738 738 2 1 7 5 22 1 1 1,236 230 2,983
Performance Analysis by Terminal report (CMRSAS13 and CMRSAS13N)

The CMRSAS13 and CMRSAS13N sample programs produce the Performance Analysis by Terminal report. This report has control breaks based on the T6ETMID (terminal ID) field. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate task performance subdivided by terminal.
143
Sample report
Sample report
Figure 40
PROGRAM: CMRSAS13 CICS REGION RUN DATE PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 TERM ID E202 E210 S211 S214 TERC *TOTAL* X002 X003 *TOTAL* S201 S202 TN03 X003 *TOTAL* E209 *TOTAL* CZE3 E206 E207 S204 S210 X002 X003 *TOTAL*
Performance Analysis by Terminal report

PERFORMANCE ANALYSIS BY TERMINAL AVERAGE AVERAGE AVG FILE AVG FILE RESPONSE CPU REAL WAIT TIME CALLS 4.387 .001 .008 1.770 .743 .042 .000 1.083 1.811 .036 .013 1.518 .678 .054 .013 1.511 1.706 .470 .000 .000 6.312 .087 .000 .000 3.524 .012 .007 1.615 13.937 .001 .004 .535 1.123 .022 .004 1.101 .051 .004 .000 .500 13.642 .001 .004 .548 13.340 .001 .006 .898 3.457 .040 .010 1.200 .333 .007 .014 1.272 1.427 .014 .000 .375 .620 .029 .016 1.718 12.079 .004 .007 .969 .027 .001 .003 .545 2.962 .046 .004 1.000 .150 .003 .003 .563 1.447 .001 .005 .995 2.112 .036 .016 1.650 1.168 .018 .019 1.777 1.924 .032 .008 1.411 .994 .035 .001 1.036 2.649 .028 .016 .615 3.125 .063 .014 .500 1.582 .026 .013 1.333 1.436 .004 .005 1.007 AVG DB WAIT TIME .000 .334 .785 .147 .000 .924 .079 .000 .172 .000 .003 .000 1.734 .115 .595 .186 .037 .000 1.355 .056 .000 1.034 .077 .705 .538 1.198 .000 .786 .050 10:27 FRIDAY, SEPTEMBER 27, 1999 1 AVG DB AVERAGE STORAGE AVG PROG CALLS STRG HWM OCCUPANCY STRG HWM .000 7,384 7,151 6,899 9.391 98,364 940 198,420 7.629 74,227 2,072 167,457 9.285 94,671 598 183,859 .000 23,130 4,281 19,422 10.000 98,348 6,003 95,360 2.224 28,825 5,667 52,040 .000 7,061 6,831 6,010 1.367 66,515 872 173,636 .000 45,788 44 77,292 .030 8,417 6,693 9,809 .000 7,086 4,234 5,915 6.640 70,617 3,301 171,156 .818 35,663 3,181 82,848 3.000 52,203 1,623 89,860 4.558 82,966 598 175,692 .463 14,322 3,911 22,368 .000 7,012 2 6,084 10.500 71,947 3,245 168,980 .437 9,717 137 12,872 .000 7,066 5,565 5,973 9.150 85,640 2,278 178,929 .777 54,978 814 137,496 5.294 82,023 2,079 172,495 6.809 93,258 1,355 198,442 10.615 59,868 2,210 150,362 .000 47,016 2,182 77,292 3.777 70,284 1,797 160,321 .535 13,513 5,239 20,604
TRAN COUNT 3,582 1,008 27 84 4 2 4,707 3,442 79 2 3,523 1,986 25 11 8 181 2,211 644 28 672 2,741 20 9 17 163 13 2 18 2,983
Performance Analysis by Node Name report (CMRSAS14 and CMRSAS14N)

The CMRSAS14 and CMRSAS14Nsample programs produce the Performance Analysis by Node Name report. This sample report has control breaks based on the T6NETNM VTAM symbolic node name. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate task performance subdivided by VTAM symbolic node name.
144
Sample report
Sample report
Figure 41
PROGRAM: CMRSAS14 CICS REGION RUN DATE PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 NETNAME
Performance Analysis by VTAM Symbolic Node Name report

PERFORMANCE ANALYSIS BY AVERAGE AVERAGE AVG RESPONSE CPU REAL WAIT 13.937 .001 1.123 .022 .051 .004 13.642 .001 13.340 .001 1.427 .014 .620 .029 3.457 .040 .333 .007 12.079 .004 .027 .001 2.962 .046 .150 .003 1.447 .001 3.125 .063 1.582 .026 2.112 .036 1.168 .018 1.924 .032 .994 .035 2.649 .028 1.436 .004 VTAM SYMBOLIC NODE NAME FILE AVG FILE AVG DB TIME CALLS WAIT TIME .004 .535 .000 .004 1.101 .172 .000 .500 .000 .004 .548 .003 .006 .898 .000 .000 .375 .595 .016 1.718 .186 .010 1.200 1.734 .014 1.272 .115 .007 .969 .037 .003 .545 .000 .004 1.000 1.355 .003 .563 .056 .005 .995 .000 .014 .500 .000 .013 1.333 .786 .016 1.650 1.034 .019 1.777 .077 .008 1.411 .705 .001 1.036 .538 .016 .615 1.198 .005 1.007 .050 10:27 FRIDAY, SEPTEMBER 27, 1999 1 AVG DB AVERAGE STORAGE AVG PROG CALLS STRG HWM OCCUPANCY STRG HWM .000 7,061 6,831 6,010 1.367 66,515 872 173,636 .000 45,788 44 77,292 .030 8,417 6,693 9,809 .000 7,086 4,234 5,915 3.000 52,203 1,623 89,860 4.558 82,966 598 175,692 6.640 70,617 3,301 171,156 .818 35,663 3,181 82,848 .463 14,322 3,911 22,368 .000 7,012 2 6,084 10.500 71,947 3,245 168,980 .437 9,717 137 12,872 .000 7,066 5,565 5,973 .000 47,016 2,182 77,292 3.777 70,284 1,797 160,321 9.150 85,640 2,278 178,929 .777 54,978 814 137,496 5.294 82,093 2,079 172,495 6.809 93,258 1,355 198,442 10.615 59,868 2,210 150,362 .535 13,513 5,239 20,604
A100A02 A100A03 *TOTAL* A051A03 A100A03 SWTCA201 SWTCA202 *TOTAL* SWTAE209 *TOTAL* A100A02 A100A03 LCZHA1E4 SWTAE206 SWTAE207 SWTCA204 SWTCA210 *TOTAL*
TRAN COUNT 3,442 79 2 3,523 1,986 8 181 25 11 2,211 644 28 672 2,741 2 18 20 9 17 163 13 2,983
Performance Analysis by User ID report (CMRSAS15 and CMRSAS15N)

The CMRSAS15 and CMRSAS15 N sample programs produce the Performance Analysis by User ID report. This sample report has control breaks based on the T6EUSER (user ID) field. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate task performance subdivided by user ID.
145
Sample report
Sample report
Figure 42
PROGRAM: CMRSAS15 CICS REGION RUN DATE PRODCICS 1999/02/15 PRODCICS 1999/09/15 PRODCICS 1999/09/15 PRODCICS 1999/09/15 PRODCICS 1999/09/16 PRODCICS 1999/09/16 PRODCICS 1999/09/16 PRODCICS 1999/09/16 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/21 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/22 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 USER ID
Performance Analysis by User ID report

TRAN COUNT 33 2,678 174 2,885 186 3,409 661 4,256 1,204 3,503 4,707 98 3,425 3,523 233 1,967 11 2,211 3 638 31 672 264 2,719 2,983 PERFORMANCE ANALYSIS BY USER ID AVERAGE AVERAGE AVG FILE AVG FILE RESPONSE CPU REAL WAIT TIME CALLS .298 .030 .009 1.393 1.253 .001 .004 1.147 1.240 .038 .003 1.040 1.242 .004 .004 1.144 .613 .021 .005 1.198 13.994 .001 .006 2.126 .970 .040 .000 .989 11.387 .008 .005 1.909 .733 .042 .005 1.151 4.483 .001 .007 1.774 3.524 .012 .007 1.615 .910 .018 .003 .897 14.006 .001 .004 .538 13.642 .001 .004 .548 .858 .027 .013 1.463 13.468 .001 .006 .907 1.313 .016 .014 1.545 12.079 .004 .007 .969 .544 .007 .000 .000 .027 .001 .003 .550 2.634 .042 .004 .903 .150 .003 .003 .563 1.202 .031 .005 1.041 1.459 .001 .005 1.004 1.436 .004 .005 1.007 AVG DB WAIT TIME .103 .000 .618 .038 .177 .000 .468 .080 .309 .000 .079 .139 .000 .003 .331 .000 .549 .037 .000 .000 1.224 .056 .571 .000 .050 10:27 FRIDAY, SEPTEMBER 27, 1999 1 AVG DB AVERAGE STORAGE AVG PROG CALLS STRG HWM OCCUPANCY STRG HWM 4.787 68,281 275 142,680 .000 7,076 5,520 5,999 9.264 78,969 1,477 164,303 .613 12,112 5,216 17,110 3.112 68,075 580 165,676 .000 7,069 6,825 6,020 10.028 88,636 1,191 187,063 1.693 22,403 5,677 41,115 8.698 94,132 1,709 185,907 .000 7,066 7,028 6,030 2.224 28,825 5,667 52,040 1.102 55,559 704 142,087 .000 7,068 6,864 6,024 .030 8,417 6,693 9,809 4.253 72,762 965 155,302 .000 7,099 4,275 5,942 3.000 68,080 1,280 143,877 .463 14,322 3,911 22,368 .000 6,805 205 5,466 .000 7,021 2 6,113 9.483 65,482 2,912 152,697 .437 9,717 137 12,872 6.045 79,803 1,419 171,052 .000 7,076 5,610 5,997 .535 13,513 5,239 20,604
CICS OH@01 *TOTAL* CICS OH@01 *TOTAL* CICS *TOTAL* CICS *TOTAL* CICS OH@01 *TOTAL* CICS OH@01 *TOTAL* CICS *TOTAL*
File Analysis for All Files report (CMRSAS21 and CMRSAS21N)

The CMRSAS21 and CMRSAS21N sample program produce the File Analysis for All Files report. This report has control breaks based on the T6EFN (file name) field. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels. Generate this report to evaluate file performance.
NOTE
A CMRDETL data set may contain two forms of T6E records, original and expanded. Any program that you run against T6E file data needs to know which type of record it is processing. CMRSAS21 and CMRSAS21N use the expanded record flag (field T6EQUAL) to determine whether a given record is expanded or not. This program also uses field T6EOFFFL as a pointer to the file information area. CMRSAS21/CMRSAS21N illustrate how to process both original and expanded format records in the same file.
146
Sample report
Sample report
Figure 43
PROGRAM: CMRSAS21 CICS REGION RUN DATE PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 FILE NAME CMRSTAT CUDCNLOG CUISCNTL CUPFR01 CUPINIT CUPSSON DFHRPL DFHUECB M2DCNLOG M2DPCT M2ISCNTL M2PACHP M2PIB21 M2PIG21 M2PINIT M2PMC01 M2PMC06 M2PMC21 M2PMC43 M2PMD21 M2PMI03 M2PMI04 M2PRA01 M2PRA21 M2PRB01 M2PRB21 M2PRC21 M2PRD21 M2PRK50 M2PSSON *TOTAL*
File Analysis for All Files report

TOTAL TRANS 148 1 2 1 3 2 241 1 1 3 12 4 2 5 65 1 4 2 2 15 3 1 1 4 2 6 1 2 1 5 541 FILE ANALYSIS FOR ALL FILES TOTAL AVG. WAITS AVERAGE WAITS PER TRAN WAIT TIME 198 1.337 .095 2 2.000 .035 7 3.500 .088 50 50.000 7.246 6 2.000 .862 82 41.000 2.875 1,829 7.589 .275 1 1.000 .346 2 2.000 .052 4 1.333 .046 27 2.250 .078 42 10.500 .811 34 17.000 2.517 112 22.400 3.621 167 2.569 .172 67 67.000 3.647 61 15.250 .354 5 2.500 1.031 8 4.000 .336 507 33.800 2.095 59 19.666 1.624 6 6.000 1.951 18 18.000 .422 50 12.500 2.396 14 7.000 .071 104 17.333 2.566 17 17.000 5.219 84 42.000 5.057 4 4.000 2.383 99 19.800 1.682 3,666 6.776 .430 10:27 FRIDAY, SEPTEMBER 27, 1999 AVG. TIME PER WAIT .071 .017 .025 .144 .431 .070 .036 .346 .026 .035 .034 .077 .148 .161 .067 .054 .023 .412 .084 .061 .082 .325 .023 .191 .010 .148 .307 .120 .595 .084 .063 1
FILE TYPE CICS FCT CICS FCT CICS FCT DB2 DB2 DB2 PSEUDO PSEUDO CICS FCT CICS FCT CICS FCT DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2 DB2
Rogues report (CMRSAS23 and CMRSAS23N)

The CMRSAS23 and CMRSAS23Nsample programs create the Performance Analysis for All Transactions report. This report is also known as the Rogues Report. CMRSAS23 produces the report by
s
reading records from the CMRDETL file calculating task response times
NOTE
Response times under 1.000 second are omitted from the report.
producing a list of the tasks with the longest response times
Generate this report to evaluate tasks with the longest response times. Modify the report control statements in the JCL stream to limit the range of input data for the report.
147
Sample report
See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels.
Sample report
Figure 44
PROGRAM: CMRSAS23 CICS REGION RUN DATE PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/23 PRODCICS 1999/09/22 PRODCICS 1999/09/21 PRODCICS 1999/09/18 PRODCICS 1999/09/18 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/16 PRODCICS 1999/09/16 PRODCICS 1999/09/16 PRODCICS 1999/09/15 PRODCICS 1999/09/15 PRODCICS 1999/09/14 PRODCICS 1999/09/14 PRODCICS 1999/09/14 PRODCICS 1999/09/10 PRODCICS 1999/09/08 PRODCICS 1999/09/08 PRODCICS 1999/09/04 PRODCICS 1999/09/03 PRODCICS 1999/09/17 PRODCICS 1999/09/17 PRODCICS 1999/09/15 PRODCICS 1999/09/14 RUN TIME 13:40:50 10:02:22 10:03:28 13:48:21 08:00:34 10:59:21 11:59:31 11:59:06 11:58:22 09:06:50 12:23:54 10:29:20 12:34:02 09:54:31 14:22:13 08:16:53 11:21:51 12:29:10 13:58:44 07:27:13 02:38:47 12:59:26 10:12:08 14:53:07 12:39:15 12:05:36 19:01:45 06:21:25
SAS rogues report

TRAN CODE MC00 MC00 MC00 MC01 PC01 MC00 MC00 MCP2 MC00 MC00 MC00 PC02 MC00 MC00 JC02 MC00 MC00 PC01 MC00 MC19 PC02 MC00 MC01 MC00 JC02 MC00 MC00 MC00 RESPONSE 30.297 22.509 17.538 15.534 24.136 37.881 16.295 14.855 17.686 17.045 14.631 24.159 15.062 14.531 18.682 17.995 33.303 22.661 22.130 15.323 25.308 19.861 14.605 21.232 18.554 14.800 15.225 15.374 PERFORMANCE ANALYSIS FOR ALL TRANSACTIONS FILE FILE DB CPU REAL WAIT TIME CALLS WAIT TIME .112 .000 1.000 25.395 .139 .118 4.000 10.282 .121 .000 1.000 13.143 .136 .000 1.000 13.165 .105 .000 1.000 18.302 .136 .136 4.000 23.105 .120 .091 4.000 6.760 .125 .070 1.000 .000 .156 .000 2.000 12.921 .157 .000 1.000 10.664 .120 .000 2.000 13.057 .124 .000 1.000 20.919 .151 .000 1.000 8.983 .121 .000 1.000 11.422 .118 .000 1.000 15.834 .121 .000 1.000 15.335 .173 .000 1.000 27.457 .077 .000 1.000 19.436 .112 .000 2.000 18.572 .131 .076 4.000 6.956 .086 .000 1.000 22.958 .029 .000 1.000 .000 .110 .000 1.000 11.859 .112 .178 4.000 12.337 .127 .000 1.000 14.967 .137 .159 4.000 7.566 .138 .000 1.000 11.878 .116 .000 1.000 11.970 10:27 THURSDAY, SEPTEMBER 26, 1999 1 DB STORAGE PROGRAM CALLS STRG HWM OCCUPANCY STRG HWM 38.000 126,992 51,905 195,424 81.000 74,488 19,002 189,744 24.000 121,168 23,743 189,632 56.000 167,504 33,529 189,512 49.000 148,080 36,061 192,680 81.000 74,488 31,827 189,744 81.000 74,312 13,598 189,744 .000 84,664 11,905 152,952 47.000 168,104 29,506 214,288 33.000 165,360 29,066 198,856 49.000 154,792 29,272 217,496 54.000 131,104 40,124 195,440 50.000 153,464 22,786 196,832 42.000 127,408 24,691 195,512 48.000 148,280 31,336 215,496 55.000 123,592 29,806 200,120 60.000 160,616 63,766 178,656 19.000 136,120 35,632 193,528 40.000 138,864 44,330 217,496 81.000 74,488 12,909 189,744 19.000 146,816 39,550 212,960 .000 67,344 14,934 181,784 37.000 126,896 25,073 196,664 81.000 74,312 18,052 189,744 46.000 128,656 31,397 198,872 82.000 74,488 12,512 189,744 41.000 127,600 25,607 195,424 39.000 126,608 25,630 194,440
Abended Transaction Roster report (CMRSAS25 and CMRSAS25N)

The CMRSAS25 and CMRSAS25N sample programs produce the Abended Transaction Roster report by
s s
reading records from the CMRDETL data set listing all transactions with an abend code
The CMRSAS25 program fails and no report is created when no abended transactions are in the selected data. See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels.
148
Sample report
Sample report
Figure 45 Abended Transaction Roster report
10:27 FRIDAY, SEPTEMBER 06, 1999 1 DB STORAGE PROGRAM CALLS STRG HWM OCCUPANCY STRG HWM .000 67,344 2,456 298,480 .000 74,176 89 242,072 .000 67,344 2,098 298,480 .000 74,176 80 242,072 .000 66,928 222 242,072 2.000 87,208 3,038 193,312 2.000 87,208 758 193,312 2.000 87,208 458 193,312 2.000 87,208 142 193,312 .000 30,952 204,839 8,896 .000 67,344 2,670 298,480 1.000 116,080 7,269 217,496 3.000 64,544 618 189,808 1.000 112,176 3,692 217,496 5.000 64,544 136 189,808 1.000 112,176 338 217,496 1.000 112,176 243 217,496 5.000 64,544 217 189,808 12.000 128,848 20,458 214,760 5.000 111,056 2,574 214,760 12.000 124,944 698 214,760 10.000 110,816 1,404 214,760 2.000 108,440 2,901 214,760 .000 96,208 1,029,337 8,896 .000 103,248 2,461,194 54,072 .000 31,048 28,320 8,896 .000 67,344 2,567 298,480
PROGRAM: CMRSAS25 ABENDED TRANSACTION ROSTER CICS TRAN FILE FILE DB REGION RUN DATE RUN TIME CODE RESPONSE CPU REAL WAIT TIME CALLS WAIT TIME PRODCICS 1999/09/05 18:09:07 MC00 3.095 .057 .059 1.000 .000 PRODCICS 1999/09/05 18:11:52 MC00 .120 .014 .000 1.000 .000 PRODCICS 1999/09/15 13:33:20 MC00 2.714 .063 .035 1.000 .000 PRODCICS 1999/09/15 13:41:31 MC00 .110 .014 .000 1.000 .000 PRODCICS 1999/09/15 15:52:03 MC00 .321 .014 .000 1.000 .000 PRODCICS 1999/09/08 13:02:52 MC00 3.171 .067 .068 6.000 .049 PRODCICS 1999/09/08 14:03:19 MC00 .850 .028 .077 6.000 .064 PRODCICS 1999/09/08 14:15:09 MC00 .431 .028 .246 6.000 .037 PRODCICS 1999/09/08 14:53:43 MC00 .165 .027 .052 6.000 .006 PRODCICS 1999/09/08 15:02:38 CEMT 1.926 .023 .000 .000 .000 PRODCICS 1999/09/09 18:19:39 MC00 3.382 .059 .020 1.000 .000 PRODCICS 1999/09/15 08:11:48 MC00 4.760 .087 .090 6.000 1.120 PRODCICS 1999/09/15 08:19:29 MC00 .802 .023 .087 6.000 .230 PRODCICS 1999/09/15 08:21:03 MC00 2.316 .035 .140 6.000 .064 PRODCICS 1999/09/15 08:27:57 MC00 .198 .024 .024 6.000 .018 PRODCICS 1999/09/15 08:33:50 MC00 .276 .035 .068 6.000 .006 PRODCICS 1999/09/15 08:37:47 MC00 .209 .034 .045 6.000 .004 PRODCICS 1999/09.15 08:40:13 MC00 .299 .024 .080 6.000 .015 PRODCICS 1999/09/17 13:12:22 MC00 12.296 .133 .080 6.000 3.665 PRODCICS 1999/09/17 13:17:54 MC00 1.699 .089 .052 6.000 .013 PRODCICS 1999/09/17 13:25:39 MC00 .483 .033 .081 6.000 .042 PRODCICS 1999/09/17 13:34:35 MC00 1.069 .033 .329 6.000 .033 PRODCICS 1999/09/17 13:48:21 MC00 2.044 .077 .548 21.000 .413 PRODCICS 1999/09/17 14:00:19 MC00 717.335 .014 .000 .000 .000 PRODCICS 1999/09/17 14:34:24 MC00 1,712.149 .075 .650 24.000 .000 PRODCICS 1999/09/17 14:57:22 CEMT .145 .003 .000 .000 .000 PRODCICS 1999/09/18 18:46:08 MC00 3.243 .060 .020 1.000 .000
Operational statistics report (CMRSAS6F and CMRSAS6FN)

The CMRSAS6F and CMRSAS6FN sample programs generate a report that lists overall transaction performance across CICS regions. The program produces the Operational Statistics report by
s s
reading records from the CMRSTATS data set reporting the CPU resource usage and response times of transactions by CICS regions
The program variable PERIOD controls the processing of CMRSTATS records and the appearance of resulting performance statistics shown in the CMRSAS6F report. Based upon the value of the PERIOD variable, performance statistics produced by each transaction record can be listed as individual report lines or they can be summarized over several fixed intervals. Permissible values for the PERIOD variable are as follows:
s s
PERIOD=1 produces a report with one print line for each input record. PERIOD=xx produces a report with one print line for each xx minutes of CICS execution time, where xx may be 10, 15, 20, 30, or 60 minutes. The default is 15 minutes.
149
Sample report
See Transaction detail data (CMRDETL) on page 28 for information about the F7 and F6 record levels.
Sample report
Figure 46
PROGRAM: CMRSAS6F CICS REGION RUN DATE CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09 CICSPROD 1999/09/09
Operational statistics report

ALL TRANSACTIONS 118 32 37 31 31 32 32 24 32 31 31 32 24 32 31 31 32 30 31 32 31 31 32 24 32 CICS OPERATIONAL STATISTICS REPORT TOTAL AVERAGE TERMINAL CPU REAL CPU REAL TRANSACTIONS 4.884 0.041 83 0.844 0.026 1 1.205 0.033 5 0.901 0.029 5 0.896 0.029 0 0.872 0.027 2 0.877 0.027 0 0.865 0.036 0 0.903 0.028 0 0.862 0.028 0 0.860 0.028 0 0.841 0.026 0 0.898 0.037 0 0.862 0.027 0 0.860 0.028 0 0.829 0.027 0 0.890 0.028 0 1.444 0.048 12 0.692 0.022 0 0.827 0.026 0 0.861 0.028 4 0.698 0.023 0 0.595 0.019 0 0.803 0.033 0 0.641 0.020 0 12:07 WEDNESDAY, SEPTEMBER 25, 1999 AVG. TERM MAXIMUM RESPONSE ACTIVE TASKS 0.355 3 0.008 3 1.165 3 0.254 3 0.000 3 0.010 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.000 3 0.515 3 0.000 3 0.000 3 1.572 3 0.000 3 0.000 3 0.000 3 0.000 3 1
RUN TIME 08/00/00 08/15/00 08/30/00 08/45/00 06/00/00 06/15/00 06/30/00 06/45/00 10/00/00 10/15/00 10/30/00 10/45/00 11/00/00 11/15/00 11/30/00 11/45/00 12/00/00 12/15/00 12/30/00 12/45/00 13/00/00 13/15/00 13/30/00 13/45/00 14/00/00
Column descriptions
The following table describes the columnar data specific to Figure 46. Table 16
Field ALL TRANSACTIONS TOTAL CPU REAL AVERAGE CPU REAL TERMINAL TRANSACTIONS AVG. TERM RESPONSE MAXIMUM ACTIVE TASKS
Operational statistics report fields

Description total number of transactions total amount of CPU real time used during this transaction average amount of CPU real time used during this transaction number of transactions executed from this terminal during this time interval average response time per transactions executed from this terminal during this time interval AMXT value from the CICS region at the time this data is collected
150
Part

Part 4
This part explains how to create custom reports using the MAINVIEW for CICS Performance Reporting Language (PRL). This part presents the following topics: Chapter 9, User-defined reports (CMRPRL and CMRPRLN) Chapter 10, PRL control statement verbs Chapter 11, PRL programming techniques
Part 4
151
152
Chapter
User-defined reports (CMRPRL and CMRPRLN)

9
You can create custom reports by writing a program with the MAINVIEW for CICS Performance Reporter Language (PRL). With PRL, you select records from the CMRDETL and CMRSTATS data sets that hold CICS transaction and statistical data. Your program determines how these records are processed and formatted in a printed report. You create PRL reports by executing a batch job that runs the CMRPRL or CMRPRLN program and its associated control statements. CMRPRL and CMRPRLN process these user-written PRL control statements to select the input VSAM records and to define the format of the report. The Performance Reporting Language supports both formats of the T6E record:
s
CMRPRL processes the F6 format of the T6E history record and uses the CMRDETL dictionary format CMRPRLN processes the F7 format of the T6E history record and uses the CMRDETLN dictionary format
The program specification on the PGM= job control statement dictates the program that is executed. Since the dictionary is used based on the program being executed, the supplied sample programs work for both types of T6E history records. Most user-written programs will also work just by changing the name of the program specified on the PGM= statement. User modifications are necessary in user-written PRL programs that use the timing fields. See Transaction detail data (CMRDETL) on page 28 for information about the F6 and F7 formats of the CMRDETL data set.
Chapter 9
153
Performance Reporting Language (PRL)
The most efficient way to execute PRL programs is by running them against small volumes of data during CICS execution or at daily CICS shutdown. Long-term trends can be reported for analysis or debugging by archiving the MAINVIEW for CICScollected data to tape (see Archiving data (CMRPURG) on page 34), and then specifying the tape as input to the PRL program (see Selecting archive data on page 207).Sample PRL programs are located in your BBSAMP library. They are named CMRPRLnn, where nn is a two-digit suffix that identifies an individual report. The JCL for submitting the programs is in the CMRL member of your BBSAMP data set. This JCL is distributed as an in-stream procedure. The JCL can be changed and installed in your sites procedure library. You should execute the sample programs to learn PRL and to obtain copies of the default reports.
Performance Reporting Language (PRL)

A typical PRL program consists of the following statements: Table 17
Statement Verbs Operators Substrings Variables
PRL statements
Description perform an action symbolically indicate the action to be performed in a mathematical expression exist for character string evaluation exist for value substitution Variables can be symbolic for user-defined data or predefined for system-provided data.
You write these statements in a PRL program to generate a user-defined batch report. The statements can assign a value to a variable, establish a condition, or control PRL execution. A PRL program is created with an interactive editor as a member of a data set. The member can be part of a PRL procedure library. BMC Software distributes sample PRL programs as members in the BBSAMP data set. The program is executed as a batch job that defines the input data and type of report produced. Sample JCL is provided in the CMRL member of the BBSAMP data set.
154
Samples
Samples
Table 18 shows a list of PRL sample programs distributed with MAINVIEW for CICS. Table 18
CMRPRL0 CMRPRL3 CMRPRL4 CMRPRL5 CMRPRL6 CMRPRL7 CMRPRL8 CMRPRL10 CMRPRL11 CMRPRL12 CMRPRL13 CMRPRL14 CMRPRL15 CMRPRL16
PRL samples (part 1 of 2)

resource usage summary for all transactions transaction analysis for all transactions highest 20 users of CPU time for a given date abended transactions list Service Level Analysis report Service Level Analysis graph summary of terminal attached transactions at 10-minute intervals for any given date Service Level Analysis Report for terminal attached transactions file control and suspend time analysis for terminal attached transactions Transaction Activity Graph terminal errors response time analysis for terminal attached transactions summary of activity for a specific program
s s s s
BBSAMP member Report type
transaction file control wait analysis file wait analysis by field transaction to file summary dataset activity and placement analysis transaction response time analysis hourly report of transactions processed summary of transaction processing summary of transactions within response time
CMRPRL17 CMRPRL18
transaction response time analysis for MRO data

s s s s
CMRPRL19 CMRPRL20 CMRPRL21 CMRPRL24 CMRPRL25 CMRPRL27 CMRPRL28 CMRPRL30 CMRPRL31 CMRPRL32 CMRPRL33 CMRPRL34 CMRPX0BD
total counts by record types from CMRDETL and CMRSTATS files transaction response analysis by response range I/O Activity report in descending I/O response by transaction DB2 Plan Usage Report SQL Activity Report by DB2 Plan Task Exception Analysis report summarized data from CMRSUMD DB2 data from variable file entries MQ data from variable file entries DBCTL data from detail file entries when file expansion is turned on Session data from variable resource entries Program Timing data from variable resource entries Program Control Detail report (CMRSTATS)
Chapter 9
155
Expressions and operators
Table 18
PRL samples (part 2 of 2)

program control summary report (CMRSTATS) transaction class report (CMRSTATS) terminal control detail report (CMRSTATS) terminal control summary report (CMRSTATS) temporary storage detail report (CMRSTATS) temporary storage summary report (CMRSTATS) file control detail report (CMRSTATS) file control summary report (CMRSTATS)
BBSAMP member Report type CMRPX0BS CMRPX0C CMRPX22D CMRPX22S CMRPX30D CMRPX30S CMRPX43D CMRPX43S
Expressions and operators

Use any arithmetic, comparative, or logical operator in a statement expression. An expression combines variables, whole numbers, and character strings with operators. Operators are as follows:
= > < + * / (start-n:length-n) equal to greater than less than not plus minus multiply divided by substring
Information about specifying substrings is provided in the following section.
Substrings
Substrings are portions of a character string. They are selected by specifying a numeric range enclosed in parentheses such as (start-n:length-n) where
156
Substrings
(start-n:length-n) is an expression that specifies a count of alphanumeric characters from the beginning of a string and a length: start-n specifies the start of the substring as indicated by n, which is the number of alphanumeric characters from the beginning of a string or referenced field. length-n specifies the number of alphanumeric characters for the substring as indicated by n. This is the length of the substring.
Examples
The following example selects the fourth character of a string and specifies a substring length of three characters. The substring, FEB, is assigned to the M2 variable.
SET Q1 = 'JANFEBMAR' SET M2 = Q1(4:3)
The next example shows how substrings can be used to determine a value, in this case the name of the month:
SET SET SET SET SET SET MONS MM V1 V2 V3 MNTH = = = = = = 'JANFEBMARAPRMAYJUNJULAUGSEPOCTNOVDEC' ZDATE(1:2) MM - 1 V1 * 3 V2 + 1 MONS(&V3:3)
The substring expression ZDATE(1:2) specifies the first character, for a length of two characters, of the mm/dd/yyyy value passed to ZDATE (03 for example). The substring expression, MONS(&V3:3), selects the portion of the character string determined by the value of the V3 variable and limits the substring to 3 characters. If 03/15/2000 is passed to the ZDATE predefined variable, the variables are assigned the following values:
Variable V1 V2 V3 MNTH Assigned value 2 6 7 MAR
Chapter 9
157
Symbolic variables
Symbolic variables
PERFORMANCE REPORTER supports symbolic variables. Variables are symbolic characters used as a substitute for a real value. They are predefined or user-defined. User-defined variables can be any alphanumeric character string that is eight bytes long and starts with an alphabetic character. A symbolic variable name can be preceded by an ampersand (&) to distinguish it from a literal, as described in PRL control statement syntax on page 170.
Predefined variables
PRL provides predefined, reserved variables. Use these variables to obtain information about the current CICS environment. The following descriptions of predefined variables show variable characters in uppercase, to indicate that these exact characters must be specified for that variable.
CDATE (Day Date Month, Year)

This variable passes the current date in the form: Thursday 12th September, 2006 It can be used to look for certain conditions on a specific day of the week or month, or to print a more descriptive title. An example follows:
SET TITLE = 'TRANSACTION DAILY REPORT' SET TITLE1 = &CDATE
ZDATE (mm/dd/yyyy)
The ZDATE variable passes the current date, expressed as mm/dd/yyyy. An example follows:
SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = &ZDATE
ZTIME (hh:mm:ss)
The ZTIME variable passes the current time, expressed as hh:mm:ss. An example follows:
PRINT 'REPORT GENERATED AT' &ZTIME
158
Generating PRL reports
The following section describes how to create PRL CICS performance reports.
Generating PRL reports

A JCL batch job executes your PRL program to create a report. An example of the job to run a PRL program can be found in member CMRL of your sites BBSAMP data set. Figure 47 shows an example of the JCL found in CMRL. Figure 47 JCL used for PRL batch execution
NEEDED NEEDED NEEDED
//jobname JOB user parameters <-- CHANGE AS /*JOBPARM user parameters <-- CHANGE AS //CMRL PROC //PRL EXEC PGM=CMRPRL //STEPLIB DD DISP=SHR,DSN=CMR.CMRV3.BBLINK <-- CHANGE AS //SYSPRINT DD SYSOUT=* //SYSUDUMP DD SYSOUT=* //SYSOUT DD SYSOUT=* //ISYPROG DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //ISYSOUT DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //PNLLIB DD DISP=SHR,DSN=CMR.CMRV3.BBPLIB <-- CHANGE AS //CMRDETL DD DISP=SHR,DSN=CMR.CMRV3.CMRDETL //CMRSTATS DD DISP=SHR,DSN=CMR.CMRV3.CMRSTATS //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,5) <-- CHANGE AS //SORTWK02 DD UNIT=SYSDA,SPACE=(CYL,5) <-- CHANGE AS //SORTWK03 DD UNIT=SYSDA,SPACE=(CYL,5) <-- CHANGE AS // PEND //RUNSTEP EXEC CMRL //SYSIN DD *
NEEDED NEEDED NEEDED NEEDED
Job control statements

The JCL for PRL batch execution is used as follows:
s
//procname PROC indicates procedural JCL. Defaults can be assigned to parameters in the procedure. //stepname EXEC specifies the name of the program that processes user-written PRL statements, which is PGM=CMRPRL or PGM=CMRPRLN.
NOTE
It might be necessary to specify a region size of 2048K.
//STEPLIB defines the program library containing the MAINVIEW for CICS load modules (BBLINK).
Chapter 9
159
Job control statements
NOTE
Alternatively, you can specify a //JOBLIB statement.
//SYSPRINT defines the output class for a sort utility. The name of the DD statement is determined at CICS system generation. //SYSUDUMP defines the output for a job that abnormally terminates. //SYSOUT defines the output class. //ddname is optional and is the name of the Data Definition (DD) statement. The ddname must start with an alphabetic character, can comprise any alphanumeric characters, and is from one to eight characters in length. The DD statement defines the output for a user-defined report. The report is allocated as defined by the DD statement if the ddname is the same as the value specified by the PRL SET REPORTID statement, as described in SET (assign a value) on page 191. Multiple DD statements can be used to direct the output of multiple reports if the ddname for each output allocation is the same as a REPORTID value, as described in Routing report output on page 203.
NOTE
A report DD statement, if used, must have DCB specifications that correspond to the report generated (the LRECL size must be large enough to accommodate all the variables specified with the USING statement).
A DD statement is not needed for report output allocation. If the DD statement is omitted and SET CLASS or SET FORM is not used, PERFORMANCE REPORTER allocates a default output class and form number. The defaults are A for the output class and STD for the form number.
s
//ISYPROG defines an output class for the Performance Reporting Language program listing. //ISYSOUT defines an output class for the messages. //PNLLIB defines the data set (BBPLIB) containing the data dictionary (CMRDETL, CMRDETLN and CMRSTATS members). PERFORMANCE REPORTER uses the data dictionary to generate report column headings for the CMRDETL and CMRSTATS records that you select, as described in Predefined data dictionary descriptions on page 205.
160
Basic PRL verbs used to create a report
//CMRSTATS defines a VSAM disk or tape data set containing records of CICS operational statistics (CMRSTATS). //CMRDETL defines a VSAM disk or tape data set containing records for each CICS transaction (CMRDETL). //SORTWKnn identifies one to nine work data sets that can be defined for data sorting. Define nn as a numeric value from 01 to 09. A minimum of one work data set is required for each user-written report. //SYSIN defines PRL control statements as input to the batch program.
Basic PRL verbs used to create a report

The following verbs generate a basic PRL report. For a complete description of each PRL verb, refer to Chapter 10, PRL control statement verbs on page 169. Table 19
Verb SET
PRL verbs
Function sets the report structure and defines any variables to be processed and reported with the CMRSTATS or CMRDETL data specifies the CMRSTATS or CMRDETL data to be processed specifies the data to be reported reports the data until end of file terminates the PRL program
SELECT USING REPORT END
You can select and report abend transaction data by combining the basic statements shown in Figure 48 with the JCL described in Job control statements on page 159 Figure 48 Example PRL verbs for creating a report
SET TITLE = 'TRANSACTIONS THAT ABENDED' SELECT TYPE 6D RECORDS FROM CMRDETL USING T6ETRID CMRDATE CMRTIME T6EABCD T6EOPID T6ETMID T6ERESP T6ECPUR REPORT END
Chapter 9
161
PRL processing
PRL processing
As shown in the previous section, some PRL statements are required for proper report execution. The use of other PRL statements depends upon what you want your report to do. This section describes how PERFORMANCE REPORTER processes a PRL program.
Logic flow
The sequence that follows shows the order in which PRL statements are processed. Those steps that are noted as optional are user dependent. If the PRL described in an optional step is used, it should be written as sequenced.
1 (optional) Set the titles and size of your report. 2 (optional) Set any variables that you want as constants.
A value assigned to a variable before a SELECT statement does not change unless the variable is used again after the SELECT statement.
NOTE
If DETAIL, SUMMARY, or USING will be used with a COMPUTE or SET statement, the variable should be defined first (the DEFINE statement).
3 Select the record types as input to your report using the SELECT statement. 4 (optional) Qualify record selection with a WHERE statement. 5 (optional) Evaluate data with a COMPUTE statement. 6 Specify the data to be reported with a USING statement. 7 (optional) Sort the data with ORDER BY.
If DETAIL or SUMMARY is specified, the first variable of ORDER BY must be the same as the first variable of the USING statement. The records are selected and ordered until end of file, and then reported.
8 (optional) Tabulate the data with DETAIL or SUMMARY.

DETAIL or SUMMARY causes a break in the reporting whenever the first variable of the USING statement changes. A summary is computed at each break.
162
Record processing
9 Report each selected record using the REPORT statement.

Any variables defined after a SELECT statement and before a REPORT statement are assigned a specified value every time a record is reported and a new record selected, until end of file.
10 (optional) Print footnotes, calculated variables, or comments after all records are
reported using the PRINT statement.
NOTE
PRINT is processed wherever it is used in the PRL program.
11 End the PRL program (the END statement).
Record processing
PERFORMANCE REPORTER processes records chosen in the SELECT statement according to the PRL verbs used between the SELECT and REPORT statements until the end of file is reached. These records are processed in chronological order unless an ORDER BY statement is used. If an ORDER BY statement is used, the processing takes place according to the PRL verbs used between the SELECT and ORDER BY statements, and the resulting data is processed by the REPORT statement. The END statement specifies the end of file.
Chapter 9
163
Record processing
Reporting selected data

SELECT requires a REPORT statement to show the results of each record processed by the PRL statements between SELECT and REPORT, until there are no more records. Figure 49 shows an example. Figure 49 Detail file control table analysissample PRL
SET TITLE = 'ANALYSIS OF FILE CONTROL TABLE' SET WIDTH = 132 DEFINE FILFNAM CHAR 08 DEFINE FILDSRD EDIT ZZZZ,ZZ9 DEFINE FILDSWRU EDIT ZZZZ,ZZ9 DEFINE FILDSWRA EDIT ZZZZ,ZZ9 DEFINE FILDSDEL EDIT ZZZ,ZZ9 DEFINE FILDSXCP EDIT ZZZ,ZZ9 DEFINE FILDSIXP EDIT ZZZ,ZZ9 * * ADD STARTING AT CLAUSE TO BEGIN AT A SPECIFIC DATE AND TIME * SELECT TYPE 43 RECORDS FROM CMRSTATS WHERE FILDSRD > 0 OR FILDSWRU > 0 OR FILDSWRA > 0 OR FILDSDEL > 0 USING CMRDATE CMRTIME FILFNAM FILDSRD FILDSWRA FILDSWRU FILDSDEL FILDSXCP FILDSIXP ORDER BY FILFNAM DETAIL REPORT END
The PRL statements shown in Figure 49 search the CMRSTATS data set for records identified by hexadecimal 43 in the TYPE field. The statements then report each specified field until end of file.
Sorting and reporting data

Data shown in PRL reports is in chronological order. Use the ORDER BY statement to change the order. ORDER BY sorts the data by the record variable specified with the statement.
164
Example: Reporting average terminal response time
The data is arranged in ascending order (low to high), unless IN DESCENDING SEQUENCE (high to low) is specified. Adding ORDER BY before REPORT in the preceding example, such as ORDER BY FCTNAME, processes each record between SELECT and ORDER BY until end of file, and then sorts each record by file name (FCTNAME) and reports the sorted data. Only one ORDER BY variable can be used per report set.
NOTE
The ORDER BY variable should be the same as the first USING variable if SUMMARY is used after ORDER BY. If it is not, unexpected results can occur.

The example in Figure 50 on page 166 shows the use of PRL to average and report terminal response time. The reported results are shown in Figure 51 on page 167.
Chapter 9
165
Figure 50
Sample PRL using the COMPUTE statement
** REPORT1 ************************************************************* * * * RESPONSE TIME SUMMARY FOR ALL TERMINAL ATTACHED TRANSACTIONS * * BASED ON THE CC RECORD TYPE 5 MINUTE RECORDING INTERVAL * ************************************************************************ ** PRINT CONTROL SECTION *********************************************** * * ************************************************************************ SET SET SET SET SET SET SET SET SET CASE CLASS FORM LINES REPORTID WIDTH TITLE TITLE1 TITLE2 = = = = = = = = = 'UPPER' 'R' 'STD ' 50 'REPORT1' '132' 'TRANSACTION RESPONSE TIME' 'Terminal-Attached Transactions' ' '
** USER DEFINED VARIABLES ********************************************** * * * VARIABLES DEFINED BEFORE THE SELECT STATEMENT WILL ONLY BE * * INITIALIZED ONCE. * * * ************************************************************************ ************************************************************************ * INCLUDE A RANGE OF DAYS. DAYS AFTER 09/11, IE 09/12. DAYS BEFORE * * 09/14, IE 09/13. * ************************************************************************ SET FDATE = '09/19/1999' SET LDATE = '09/22/1999' ************************************************************************ * INCLUDE A RANGE OF TIMES. HOURS AFTER 7:59 AM AND HOURS BEFORE * * 8:00 PM. * ************************************************************************ SET FTIME = '07:59:00' SET LTIME = '20:00:01' ** SELECT CONTROL SECTION ********************************************** * * * BEGIN LOOP: THE SELECT STATEMENT BEGINS A REPORT LOOP FOR EACH * * RECORD UNTIL END OF FILE. * * * ************************************************************************ SELECT TYPE CC RECORDS FROM CMRSTATS WHERE CMRDATE > FDATE AND CMRDATE < LDATE AND CMRTIME > FTIME AND CMRTIME < LTIME ** USER AND OTHER PRL VARIABLES **************************************** * * * VARIABLES DEFINED HERE WILL BE ASSIGNED EVERY TIME A RECORD IS * * SELECTED. * * * ************************************************************************ COMPUTE AVTRMRSP = T6FRSPTM / T6FTRMCT USING CMRTIME AVTRMRSP T6FTRMCT T6FSSHWM T6FSPGIN T6FSPGOU REPORT END -
Figure 51 on page 167 is the result of using COMPUTE as shown in the PRL example in Figure 50.
166
Figure 51
REPORT1
Terminal-attached transaction response time report

VERS: REL5.4.0 TIME: 06:27:33 DATE: 09/22/1999 PAGE: 1 PAGE-IN COUNT -----------165 172 312 215 293 319 174 100 PAGEOUT COUNT ------------26 31 27 67 124 67 18 20
TRANSACTION RESPONSE TIME TERMINAL-ATTACHED TRANSACTIONS
RECORDED TIME -------12:33:00 12:38:01 12:43:00 12:48:00 12:53:00 12:58:00 13:03:00 13:08:00
AVTRMRSP -------------0.00000 0.00000 0.25196 0.07513 0.14303 0.12650 3.05300 0.00000
TERMINAL TASKS ----------0 0 32 118 62 42 1 0
DSA HWM ----------1310 1310 1321 1404 1424 1421 1415 1415
Chapter 9
167
168
Chapter
10
10
PRL control statement verbs

The first word of every PRL statement is a verb or a condition (IF, FOR, WHERE) with a verb in the expression. This chapter describes PRL verbs in alphabetical order. Each verb is discussed in a separate section that includes a description of the verb, its uses, syntax, and a coding example.
Table 20
PRL verb COMPUTE DEFINE DETAIL
PRL control statement verbs (part 1 of 2)

Description calculates a mathematical expression assigns the data characteristics of a variable creates a tabular summary at each control break See this page 173 175 177
DIAGNOSE END FOR and NEXT GOTO IF and THEN LET ORDER BY PRINT REPORT SCAN SELECT SET
creates a system snap dump and diagnostic 178 report terminates the execution of a PRL program 179 repeats a series of instructions in a loop branches to another location of a PRL program identified by a label tests conditions and makes a program branch according to the outcome assigns a length to variable 1 from the length of variable 2 sorts data prints a report 180 182 183 185 185 187
creates a report from the data processed by 188 PRL statements reads a variable and performs an editing function on the data selects records from CMRDETL and CMRSTATS data sets assigns a value to a variable 188 189 191
Chapter 10
169
PRL control statement syntax
Table 20
PRL verb
PRL control statement verbs (part 2 of 2)

Description See this page creates a tabular summary of the first 195 preceding variable in the USING statement designates column headings and the data 197 shown in a report based upon record fields qualifies record selection based on whether 199 a condition is met
SUMMARY USING WHERE
PRL control statement syntax

PRL is a free-form language with minimal restrictions. It allows approximately 1200 user-defined variables. The number of allowed PRL statements is limited by the available memory. The left most word of the statement is the verb.
Conventions
To enter statements and comments
s
Start each statement with the verb. Each of the PRL verbs is described in this section. PRL verbs must be written in uppercase. Write the PRL statements between columns 1 through 72. Characters between columns 73 and 80 are ignored. Use one or more blank spaces to separate the verb and parameters in a statement. Use indentation and blank lines to improve readability. Describe statements with comments. Comments are indicated by an asterisk (*) in column 1. Each comment should be on a separate line preceding the statement it describes. Use the END verb followed by a semicolon (END;) to indicate the end of a set of PRL verbs that specify an IF-THEN-BEGIN branching routine. Place each verb on a separate line for readability.
To terminate execution
Use the END verb to terminate PRL execution.
170
Conventions
To use labels
Use labels to direct program flow to another statement. Labels
s s s s s s s
are used with GOTO statements must be unique within the PRL program should be on a separate line can start in any column can be any length can have special characters but no blank characters must end with a colon (:) sign
To specify a literal
Enclose a literal in single quotation marks to distinguish it from a variable.
To specify a variable
s
Begin each variable with an alphabetic character; remaining characters for the variable can be alphanumeric. All variables must be eight characters or less. Predefined variables must be written as shown. Put an ampersand (&) sign in front of a variable used within a quoted literal. If an & does not precede any characters enclosed in quotation marks, all characters are processed as literals. For example, the following specification sets V3 to ABCDEF: SET V1 = 'ABC' SET V2 = 'DE' SET V3 = '&V1&V2.F' If the last statement in this example used 'V1V2.F', V3 would be set to V1V2.F. (See SET (assign a value) on page 191.)
To concatenate variables and literals

Use a period to concatenate a variable to a literal within single quotes, as shown on the preceding page.
To assign a numeric value to a variable

Use a decimal point, a sign, or commas (no blank characters) with a number used in an expression or assigned as a value to a variable, as described in SET (assign a value) on page 191.
Chapter 10
171
PRL reserved words
The numeric value is not enclosed in single quotation marks. An example follows: SET DAYS = 7
To specify a substring
Use (start-n:length-n) to select a portion of a character string, as described in Substrings on page 156. Start-n specifies the number of characters from the beginning of the string. Length-n is the length of the substring. The numerics must be delimited with a colon and enclosed in parentheses, such as (4:2), which counts to the fourth character of a string and limits the substring to two characters.
To continue the USING or PRINT statement

If necessary, use a minus (-) sign to continue the USING and PRINT statements one line to another line.
PRL reserved words

The following list shows PRL reserved words. None of these reserved words should be used as variable names in PRL jobs to create batch reports.
NOTE
Note that nnnnnnn are numbers, Dnnnnnnn are numbers reserved for titles, and Unnnnnnn are numbers reserved for details.
& AND CDATE CTIME Dnnnnnnn DETAIL END FACT FOR GOTO HELP IF
ALARM BLINK COLOR CURSOR DEFINE DIAGNOSE EXEC FCMD GCMD GRAF HIGHLIGHT LET
ALTERVS CCMD COMPUTE CYCLE DELAY DUMP EXIT FIND GET HEADING HISTORY LIN001
172
COMPUTE (calculate mathematical expression)
LOGOFF OR PANELID PROFMEM RAND REVERSE SCMD SUMMARY TCOL TROW UNDERSCORE ZCARD ZCT ZDATE ZIFY ZLINES ZMSG ZREQUEST ZTIME ZUSER
NEW ORDER PRINT PROFRC REPORT RVIDEO SELECT SYMBOLISE TRACE TUTOR USING ZCICS ZCURSOR ZERRMSG ZLAST ZLOC ZOPTERR ZSCREEN ZTITLE ZWRAP
NEXT PAD PROFCNT PUT RETURN SCAN SET SYMBOLIZE TRIM Unnnnnnn WHERE ZCSA ZCW ZIFX ZLINE ZMEMBER ZPF ZSHOW ZTRAN ZXDT
COMPUTE (calculate mathematical expression)

COMPUTE calculates a mathematical expression and places the results in a variable. The expression specified by the COMPUTE statement uses CMRSTATS or CMRDETL records as variables. The SELECT statement declares the CMRSTATS or CMRDETL records that are used as COMPUTE variables. The SELECT statement usually precedes the COMPUTE statement in a PRL program. The results assigned to the variable by COMPUTE can be reported by naming the variable in the USING statement. The COMPUTE statement usually appears before the USING statement in a PRL program.
Chapter 10
173
Syntax
Syntax
COMPUTE V1 = expression where
s
V1 is a 1- to 8-byte user-specified alphanumeric character string beginning with an alphabetic character. expression sets the user-specified variable to the results of the calculation as expressed by the = sign with any of the following operators:
adds two record variables together subtracts one record variable from another multiplies record variables divides record variables
+ * /
Whole numbers with signs, commas, or decimal points can be used with any of these operators. They also can be used alone to represent real numbers.
Usage
When the variable being computed is used with DETAIL, SUMMARY, or USING statements, a DEFINE statement should be inserted before the COMPUTE statement. The computations on printouts can result in the maximum of a 16-digit mathematical expression.
Example
In the following example, COMPUTE uses a simple addition expression and places the results in TOTALPG. The results of the computation are reported for each selected record until the end of file. The USING statement specifies the user-defined TOTALPG variable as a column heading.
174
DEFINE (define variable attributesoptional)
SET TITLE = 'DAILY PAGING AND TASK ANALYSIS' SELECT TYPE CC RECORDS FROM CMRSTATS COMPUTE TOTALPG = T6FSPGIN + T6FSPGOU USING CMRDATE T6FSPGIN T6FSPGOU TOTALPG T6FTRNCT SUMMARY REPORT END
DEFINE (define variable attributesoptional)

DEFINE fixes the picture of a variable for the duration of the PRL run. This occurs on the first use; it is ignored for subsequent uses. The use of this verb causes the variable attribute length and decimal significance to be unalterable. In addition, it provides you with the means to specify an edited pattern for display of fields. This is particularly useful if the variable is to appear on the report with USING. This function also can be particularly useful when used with the SUMMARY or DETAIL functions. DEFINE ensures that the column width accommodates the totals without decimal truncation or alteration. DEFINE can be used only with alphabetic or numeric fields. It cannot be used with DATE, TIME, or HEX fields.
Syntax
DEFINE variable {CHAR nn | EDIT ZZ,ZZ9.99}
where
s
variable is a symbolic variable. The symbolic variable can be predefined or userdefined. User-defined variables must start with an alphabetic character, can be up to eight characters long, and can comprise any alphanumeric character. For a further discussion about variables, see SET (assign a value) on page 191. CHAR nn identifies the verb as a DEFINE for a character variable; nn specifies the number of characters to be shown.
Chapter 10
175
Usage
EDIT ZZ,ZZ9.99 identifies the verb as a DEFINE for a numeric variable; ZZ,ZZ9.99 are symbols representing the digits in the number. The description can contain up to 25 characters or symbols (but only 15 digits). The following symbols are supported:
decimal placement and representation numerical editing decimal significance decimal significance (zero suppresses digits) blanks sign characters
'.' , 9 'Z' '' -
NOTE
The edit pattern should be at least as long as the default title for that field.
Usage
Use the DEFINE statement only once per variable. Define the variable before using it in another statement. A DEFINE statement should be inserted before a COMPUTE or SET statement when the COMPUTE or SET statements are used with DETAIL, SUMMARY, or USING.
Example
DEFINE T6ERESP EDIT ZZ9.99 SELECT TYPE 6E RECORDS FROM CMRDETL USING T6ETRID T6ERESP SUMMARY REPORT END
176
DETAIL (tabulate and summarize each record)
DETAIL (tabulate and summarize each record)

DETAIL requests a tabular summary at each control break based on the first variable mentioned in the USING statement that precedes it. A count is maintained for each item that has a common value matching the first variable. When the value of the first variable changes, a control break occurs and the totals and averages are printed. If neither SUMMARY nor DETAIL is used, a tabular report with no control breaks is printed.
Syntax
DETAIL
Usage
DETAIL can be coded only after the USING statement but prior to REPORT. If ORDER BY is used, DETAIL must follow this statement.
Example
The report shown in Figure 52 on page 178 is produced when the following PRL is executed:
SET TITLE = 'DETAIL TRANSACTION ROSTER' SELECT TYPE 6E RECORDS FROM CMRDETL USING CMRDATE CMRTIME T6ETRID T6ERESP T6ECPUR T6ETMID T6EOPID DETAIL REPORT END
Chapter 10
177
DIAGNOSE (perform diagnostics)
Figure 52
REPORT1
Detail report
VERS: REL3.5.0 TIME: 06:27:25 DATE: 09/17/1998 PAGE: 1 CPU REAL TIME -----------0.067 0.021 0.088 0.044 0.321 0.078 0.399 0.199 0.174 0.155 0.138 0.467 0.156 L06B L06B R01 R01 TERM ID. ---OPR ID. ---
DETAIL TRANSACTION ROSTER
RECORDED DATE ---------09/17/1998 09/17/1998
RECORDED TIME -------10:21:12 10:31:12
TRAN NAME TOTAL --------------ETI0 ETI0 ETI0
RESPONSE TIME ----------3.511 0.186 3.697 1.848 5.633 0.906 6.539 3.269 5.504 1.353 1.099 7.956 2.652
2 AVG *****
09/17/1998 09/17/1998
10:21:23 10:22:31
ETMO ETMO ETMO
2 AVG *****
09/17/1998 09/17/1998 09/17/1998
10:21:33 10:22:14 10:22:26
ETM2 ETM2 ETM2 ETM2
L06B L06B L06B
R01 R01 R01
3 AVG *****
DIAGNOSE (perform diagnostics)

DIAGNOSE forces entry into the PRL diagnostic environment with a system snap dump to produce a diagnostic report that helps BMC Software isolate any problems that might occur. The report shows
s s s s s
name length assigned value data type (E is numeric; blank indicates characters) significant decimal places (3 is .001)
Syntax
DIAGNOSE
Usage
The DIAGNOSE statement should be used only for PRL program execution analysis.
Example
Example
The following example shows the use of DIAGNOSE to analyze a previously executed PRL program. DIAGNOSE is added before the END statement.
SET TITLE = 'MAINVIEW for CICS COMMON SYSTEM AREA (CSA) ANALYSIS' SELECT TYPE 11 RECORDS FROM CMRSTATS USING CMRDATE CMRTIME CSATSKCT CSAGTMAN CSAFRMAN CSADUMPS CSASOSCT ORDER BY CSATSKCT REPORT ************************************************************** * Insert DIAGNOSE statement before END if PRL does not execute * properly. DIAGNOSE produces system snap dump and list of * variable pool elements showing variable attributes. ************************************************************** DIAGNOSE END
* * *
END (stop execution)

END terminates PRL execution.
Syntax
END
Usage
Use END
s
to indicate PRL report termination in the THEN clause of an IF statement to force termination of PRL execution
Chapter 10
179
Example
For example, to get only the first 50 selected records, use END with IF-THEN as follows:
SET Y = Y + 1 IF Y > 51 THEN END
(See the example for IF and THEN (do statement to test conditions) on page 183.)
s
after DIAGNOSE
Example
SET TITLE = 'TRANSACTIONS THAT ABENDED' SELECT TYPE 6D RECORDS FROM CMRDETL USING T6ETRID CMRDATE CMRTIME T6EABCD T6EOPID T6ETMID T6ERESP T6ECPUR REPORT END
FOR and NEXT (execute a set of statements on a condition)

FOR and NEXT perform a series of instructions a given number of times in a loop and cause a variable to be used repeatedly.
Syntax
FOR V1 = V2 TO V3 statement statement NEXT V1
where
s
V1 is the counter value. The value should be defined using the SET verb.
180
Usage
V2 is the initial value of the counter. The length attribute of V2 must be the same as the V3 length attribute. V3 is the final value of the counter. If a variable name is used, it must have an ampersand (&) prefix. statement can be any PRL statement.
The set of statements following the FOR statement are executed until a NEXT statement is encountered and a specific condition is met. There is at least one execution of the series of statements, even if the final counter value is less than the initial value. After each execution, the V1 value is incremented by 1. Processing loops back to the beginning of the set of statements after the FOR statement until the final value is reached. When the counter value is greater than the final value, the statement following the NEXT statement is executed.
Usage
You can
s
use FOR and NEXT anywhere within the PRL The FOR-NEXT loop usually follows a SELECT statement.
nest FOR and NEXT loops, as shown in the following example The counter within each nested loop must have a unique variable name. The NEXT statement for the inner FOR-NEXT loop must appear before the NEXT statement for the outer FOR-NEXT loop.
use any of the variables that are described in Symbolic variables on page 158 or Predefined variables on page 158
Example
FOR &X = 001 TO 100 FOR &Y = 001 TO 100 SET Z = X + Y NEXT &Y NEXT &X
Chapter 10
181
GOTO (continue processing at label with : suffix)
GOTO (continue processing at label with : suffix)

GOTO branches to the PRL marked by a label ending with a colon (:) sign. Execution continues at the first statement following the label.
Syntax
GOTO label where
s s
GOTO directs program flow to another set of PRL statements. label is the name of a set of PRL statements. The label must have a colon (:) suffix. The label is alphanumeric, can include special characters but no blank characters, and must end with a colon (:) sign. The label should be on a line by itself and must be unique. It can be any length and can start in any column, as follows:
IF V1 = V2 THEN GOTO LABEL statement statement LABEL: statement END
182
Example
Example
SELECT TYPE 6E AND 6D RECORDS FROM CMRDETL WHERE T6ETMID = ' ' IF TYPE = '6D' THEN GOTO ABEND SET CIC = T6ETRID(1:1) IF CIC = 'C' THEN GOTO CICS ABEND: SET REPORTID = 'TRANAB' SET TITLE = 'TRANSACTIONS THAT ABENDED' SET CLASS = 'R' USING CMRDATE CMRTIME T6ERESP T6ECPUR T6ETRID T6EOPID T6ETMID REPORT CICS: SET REPORTID = 'CICSTRAN' SET TITLE = 'CICS-SUPPLIED TRANSACTIONS' SET CLASS = 'R' SET HH = &CMRTIME(1:4) SET TIME = '&HH.0 ' USING TIME T6ETMID T6EOPID T6ECPUR T6ETCWT T6ETRID T6ERESP DETAIL REPORT END
IF and THEN (do statement to test conditions)

The IF and THEN statements test conditions. The results of the test determine the action to be taken by the PRL program.
Syntax
IF expression THEN statement | BEGIN
where
s
IF-THEN statement is all on one line. expression consists of variables, whole numbers, or character strings connected by operators (see Expressions and operators on page 156).
NOTE
The first parameter is used to define the length in an alphabetic compare.
Chapter 10
183
Example
THEN is required and can be followed by a PRL statement or a BEGIN clause terminated by an END statement with a semicolon (;) suffix. statement can be any of the following PRL statements: BEGIN, GOTO, END, SET, or EXEC. An example follows:
IF V1 = V2 THEN SET V3 = V4
s
BEGIN indicates that a series of PRL statements follow the IF-THEN condition. More than one statement can be nested within the IF condition when a true condition results from the test. The end of the statements is signaled by an END statement followed by a semicolon (END;):
IF V1 = V2 THEN BEGIN statement statement END;
The semicolon (;) suffix distinguishes this END statement from an END that terminates PRL execution. END without a semicolon (;) suffix terminates PRL execution as follows:
IF V1 > V2 THEN END
The entire report generation sequence can be coded in an IF statement. For example, to produce a report that outlines, among other things, the 20 busiest CPU users in your system, keep a variable called COUNT and increment it just before REPORT. Follow this with a conditional statement that terminates the job when 20 are reported.
Example
SET TITLE = 'HIGHEST 20 USERS OF CPU TIME - &ZDATE' SELECT TYPE 6E RECORDS FROM CMRDETL WHERE &CMRDATE = &ZDATE USING T6ETRID CMRDATE CMRTIME T6EUSER T6ETMID T6ERESP T6ECPUR ORDER BY T6ECPUR IN DESCENDING SEQUENCE SET &COUNT = &COUNT + 1 IF COUNT > 20 THEN END REPORT END
184
LET (set variable length)
LET (set variable length)

LET, in combination with the LENGTH parameter, sets variable1 in a statement to the length of variable2. As a result, the first variable becomes a 3-digit value between 000 and 256, depending on the length of the second variable.
Syntax
LET variable1 = LENGTH variable2
where
s
variable1 is the name of a variable or array element to receive a length value. variable1 can be a character string, a numeric variable, or an array element. variable2 is the variable whose length will be assigned to variable1.
Example
The following example keeps the length of file and transaction names to 70 characters or less:
LET &LTR = LENGTH &TRNID(&K) IF &LTR > 70 THEN GOTO LAB3
ORDER BY (sort selected records)

ORDER BY sorts variables into a different order from that presented by the SELECT verb. If the report is to be in chronological order, you do not need to sort your data because all of the records are time stamped with CMRDATE and CMRTIME. Occasionally, you might want to arrange the report by transaction code or by response time. This order is useful if you are interested in summaries based on a group of transactions, a range of execution periods, or other similar information. The ORDER BY statement relies upon the IBM Sort Facility to pass user data to a sort exit as the second parameter in the list. Other sort facilities can be used if they are compatible with the IBM Sort Facility (use standard sort exits E15 and E35).
Chapter 10
185
Syntax
The ORDER BY statement can be overridden with your own sort control statement, such as
//SORTCNTL DD * options - /*
Syntax
ORDER BY v1 [IN DESCENDING SEQUENCE]
Usage
s
Only one ORDER BY statement is allowed per report set. Only one variable can be specified with ORDER BY. It can be either a user-defined or predefined variable. If DETAIL or SUMMARY is used, the ORDER BY variable must be the same as the first variable in the USING statement, as shown below. IN DESCENDING SEQUENCE orders the data from high to low values.
Example
SELECT TYPE 6E RECORDS FROM CMRDETL USING T6ETRID T6ERESP T6ECPUR T6EFCWT T6ESHWM T6EPAGCT ORDER BY T6ETRID SUMMARY REPORT END
186
PRINT (print one-line message)
PRINT (print one-line message)

PRINT causes a one-line message, the value of a variable, or a combination of these to be printed in a report whenever the PRINT statement is processed.
Syntax
PRINT [V1|'string'|V1..Vn 'string'|fieldName]
where
s
PRINT (with no parameters) prints a blank line. V1 is a variable that is assigned a value by the SET statement. The value of the variable is printed. 'string' is a literal character string enclosed in single quotation marks. The string can contain alphanumeric, blank, or special characters. fieldName is the name of a field in a CMRDETL or CMRSTATS record. See MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. The field value is printed.
Example
SET DOWN = ' * * * MAINVIEW for CICS NOT ACTIVE * * *' SET TITLE = 'MAINVIEW for CICS GLOBAL PERFORMANCE ANALYSIS' SELECT TYPE CC RECORDS FROM CMRSTATX USING CMRDATE CMRTIME T6FSUCP T6FSTCCP T6FSJCCP T6FSKCCP T6FSOSCP IF T6FSFLAG = 'L' THEN BEGIN PRINT PRINT DOWN PRINT END; REPORT END
Chapter 10
187
REPORT (report selected record)
REPORT (report selected record)

REPORT produces a report that shows the results of each record processed by the PRL statements between SELECT and REPORT. Each field specified by the USING statement is reported until end of file. The default report size is 50 lines x 132 print positions. The report size can be changed by SET LINES = and SET WIDTH = as shown by SET (assign a value) on page 191. The width of the body of the report is determined by the record fields and variables specified in the USING statement (see USING (designate column headings and record fields) on page 197.) The statement following REPORT receives control when all selection criteria presented by the SELECT verb have been met and the file has reached its end.
Syntax
REPORT
Example
SELECT TYPE CC RECORDS FROM CMRSTATS USING CMRDATE T6FSPGIN T6FSPGOU SUMMARY REPORT END
SCAN (scan a variable field)

SCAN scans a variable and edits it.
Syntax
SCAN &Vn {REMOVING|REPLACING} {LEADING|ALL} {BLANK|NULL|c} [WITH|[{BLANK|NULL|c}]]
188
Example
where
s
&Vn is the name of a defined variable that contains the target data to be edited. REMOVING | REPLACING indicates the editing to be performed on the data of the target variable. LEADING | ALL edits the leading or trailing data to one byte or all the data as indicated. WITH must be used if REPLACING is specified. BLANK specifies blank characters. NULL specifies null characters. c represents any alphanumeric characters.
Example
SET BB = '&vbarSELECT TYPE CC RECORDS FROM CMRSTATS WHERE CMRDATE = '09/17/1998' SET Y = T6FTRNCT / 5 SET Y = Y + 1 IF Y > THEN 51 THEN SET Y = 52 SCAN &Y REMOVING LEADING 0 SET &X = BB(1:&Y) USING CMRDATE CMRTIME T6FTRNCT X REPORT END
SELECT (find record type)

SELECT chooses records from CMRDETL or CMRSTATS by the record type. Optionally, the starting point of the SELECT can be qualified with a date and time.
Chapter 10
189
Syntax
Syntax
SELECT SELECT SELECT SELECT SELECT SELECT TYPE xx RECORDS FROM dataSet TYPE xx xx RECORDS FROM dataSet TYPE xx AND xx RECORDS FROM dataSet TYPE xx xx xx ... xx RECORDS FROM dataSet ALL RECORDS FROM dataSet TYPE xx RECORDS FROM dataSet STARTING AT time ON date
where
s
TYPE xx or ALL names the specific records to be used from the MAINVIEW for CICS CMRSTATS or CMRDETL data sets; xx is the hexadecimal identifier of the record as described in Collecting report data on page 27 and shown in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. As shown in the syntax example, more than one record type can be selected from a data set with a single SELECT statement. A maximum of nine record types can be specified on one line. ALL selects all record types from a single data set. The specified records must reside in the same data set. Type 6E and 6D records reside in the CMRDETL data set, while all other record types are in CMRSTATS. A maximum of nine specific record types can be selected.
RECORDS FROM is a required clause. dataSet can be either CMRSTATS or CMRDETL on tape or disk, as defined by the JCL. PERFORMANCE REPORTER searches the data set for a record ID that matches the value specified by SELECT TYPE xx and makes those records available for PRL processing. STARTING AT time ON date specifies when reporting is to begin. MAINVIEW for CICS performance data is arranged in chronological order with date and time stamps. The date is always represented as mm/dd/yyyy and the time as hh:mm:ss. To concentrate on a particular time period, add this parameter on the end of the SELECT statement. The selection process will not start until the first record is found. Todays date is given by using the variable ZDATE. The current time is given by the variable ZTIME.
For example, the following statement reports CICS activity since the start of business today:
SELECT TYPE CC RECORDS FROM CMRSTATS STARTING AT 08:00:00 ON &ZDATE
This job can be run on any day and the start date does not need to be changed.
190
Usage
NOTE
If records are to be selected from an archive tape, STARTING AT has no effect and should not be used. Records can be selected from an archive tape by using WHERE as follows: SELECT TYPE CC RECORDS FROM CMRSTATS WHERE CMRDATE = '09/17/1998'
Usage
Only one SELECT control statement per CMRDETL or CMRSTATS per run can be specified in each PRL program. You could receive unpredictable results if you do use more than one SELECT control statement in your program.
Example
SET TITLE = 'D E T A I L T R A N S A C T I O N R O S T E R' SELECT TYPE 6E RECORDS FROM CMRDETL USING CMRDATE CMRTIME T6ERESP T6EFCWTC T6ETRID T6ENETNM REPORT END
SET (assign a value)

SET, in combination with the = operator, assigns a value (characters or numerics) to a variable. It can be used to specify report headings and the contents of variables and to control the printing of the report.
Syntax
SET variable = expression
where
s
variable is a symbolic variable. The symbolic variable can be predefined or userdefined. A user-defined variable must start with an alphabetic character, can be up to eight characters long, and can comprise any alphanumeric character. It can be preceded by an ampersand (&) to distinguish it from a literal.
Chapter 10
191
Usage
expression can be variables, whole numbers, or character strings alone or in combination with operators (see Expressions and operators on page 156).
A value that was assigned to a variable before a SELECT statement does not change.
Usage
When the variable being set is used with DETAIL, SUMMARY, or USING statements, a DEFINE statement should be inserted before the SET statement. A symbolic variable can be set to
s
the value of another variable, which can be symbolic or predefined (see Predefined variables on page 158)
SET WEEK = &DAYS
NOTE
Ampersand (&) is a reserved word and should not be used as a variable value. In other words, &a is a variable while & or && is not.
a character string Enclose characters in single quotation marks. Begin a variable inside quotation marks with an ampersand. Use a period to delimit a variable from a literal. The characters can be a direct substitute for the variable, a substring, or concatenated with other variables, as shown in the following examples: Example 1: Substitution In this example, ABCDEF is substituted for the symbolic variable, ALPHA:
SET ALPHA = 'ABCDEF'
Example 2: Substring In this example, B is assigned to the symbolic variable, BETA. The substring expression specifies that the second character of the string be substituted for &ALPHA, for a length of one character.
SET BETA = &ALPHA(2:1)
192
Usage
Example 3: Concatenation Character strings can be concatenated with other variables and literals. Identify variables with an ampersand prefix and add a period at the end of the variable to delimit it from a literal, as follows:
SET ALPHA = 'A' SET DELTA = 'DEF' SET AD = '&ALPHA.BC&DELTA'
The AD variable is assigned a value of ABCDEF. If an ampersand prefix or a period suffix is not used to delimit the variables from the literals, the characters are assigned to the AD variable. An example follows:
SET AD = 'ALPHABCDELTA'
In the preceding example, the AD variable is assigned a value of ALPHABCDELTA.

s
a numeric integer to be used as a counter For decimal arithmetic, use the COMPUTE verb.
SET DAYS = 7
A number can be an integer from 0 to 9. The number can have one decimal point, a sign, or commas, but it cannot contain any blank characters or be enclosed in single quotation marks. A numeric value also can be assigned to a variable as a counter. The number of digits is initially set to 00000 as a default. When 99999 is reached, the value resets to 00000. The default digits can be changed with the SET statement. For example, SET number = 01 establishes a two-digit counter that, at 99, resets to 00. An example follows:
SET NUMBER = 01 SET V1 = V2 + NUMBER
NOTE
SET does not keep track of decimals or signs.
Chapter 10
193
Usage
Report output can be set as follows (all specified values must be enclosed in single quotation marks):
s
CASE specifies that the output characters be in uppercase or mixed case. The default is uppercase. SET CASE must precede other PRL statements and is written as
SET CASE = 'MIXED' SET CASE = 'UPPER'

s
CLASS specifies the output class for the report. It is a 1-byte, site-defined character. A is the default.
NOTE
Report output also can be allocated with a DD statement, as described in Routing report output on page 203, that has the same name as the value of SET REPORTID.
FORM specifies four characters defining the type of paper on which the report is to be printed. STD is the default.
NOTE
Instead of specifying SET FORM, a DD allocation that has the same name as the value of SET REPORTID can specify the form type.
LINES specifies the report page length. The default is 50 lines. The minimum value is 001 and the maximum value is 255. REPORTID specifies a 1- to 8-character alphanumeric ID for the report. The ID is in the upper left corner of the report. The execution job name is the default. If a DD output allocation has the same name as the user-specified value for REPORTID, the report is allocated to the DD output, as described in Routing report output on page 203.
TITLE specifies the first of four lines to be generated by PRL on every page. Specify any alphanumeric string up to a maximum of 80 characters. The default is MAINVIEW for CICS. The title is centered automatically at the top of each page. TITLE1 and TITLE2 specify a subtitle for the second line (TITLE1) or third line (TITLE2) of every page. The subtitle length is 80 characters. Each specified subtitle is centered automatically at the top of each page. Blank is the default.
194
Example
WIDTH specifies the width of the report title. The default is 132 print positions. From 32 to 255 print positions can be specified for your report. Report titles are adjusted to the specified width. The USING statement determines the column data to be reported. The reported data is truncated if it does not fit within the specified width. If USING specifies only a few variables and the WIDTH is at a large value, the column data is reported as formatted by the PNLLIB data dictionary. It is not adjusted to the specified WIDTH value. ZHEADING specifies the print characteristics of the column headings of PRL tabular reports. If ZHEADING is not specified, the default prints each column heading three times without a carriage return. The default should be used with line printers to create bold text in the column headings by printing each word three times. Use ZHEADING = 'SINGLE' to create viewable online reports. Use ZHEADING = 'NONE' to create PRL reports without column headings.
Example
SET SET SET SET SET SET SET CLASS = 'R' FORM = 'C012' WIDTH = 70 LINES = 58 REPORTID = 'CICSM16' TITLE = 'MAINVIEW for CICS - Monthly System Status' TITLE1 = ' User Logon Times ' SELECT TYPE 6E RECORDS FROM CMRDETL WHERE T6ETMID NOT EQUAL ' ' SET COUNT = COUNT + 1 USING CMRDATE CMRTIME T6ETRID T6ERESP T6ECPUR T6ETMID T6EOPID REPORT PRINT COUNT END
SUMMARY (total and average selected records)

SUMMARY requests a tabular summary based on the first variable mentioned in the USING statement that precedes it. A count is maintained for each item that has a common value matching the first variable. Only the totals appear on the report, followed by an average, for each and every numeric variable on the print line. Even after all selection criteria have been met, each record generates a separate report line; the report could become quite large if a large volume of data is being processed.
Chapter 10 PRL control statement verbs 195
Syntax
If you provide summaries, your nominated summary fields will gather and condense all detail records into two lines of print for each control break. If neither SUMMARY nor DETAIL is used, a tabular report is printed.
Syntax
SUMMARY
Usage
SUMMARY can be coded only after the USING statement but prior to REPORT. If ORDER BY is used, SUMMARY must follow this statement.
NOTE
Character data can only be included in a summary report if it is specified as the first USING variable.
Example
The following example produces the report shown in Figure 53 on page 197:
SET CLASS = 'R' SET REPORTID = 'RES2RPT2' SET TITLE = 'SUMMARY TRANSACTION ROSTER' SELECT TYPE 6E RECORDS FROM CMRDETL USING CMRDATE T6ERESP T6ECPUR T6EDIST T6EFCWT T6ESUST T6ETSWT SUMMARY REPORT END
196
USING (designate column headings and record fields)
Figure 53
RES2RPT2
Summary report
VERS: REL5.3.0 TIME: 16:21:18 DATE: 09/17/1998 PAGE: 1 CPU REAL TIME -------0.248 0.124 0.297 0.148 TERM ID. ---OPR ID. ---
SUMMARY TRANSACTION ROSTER
RECORDED DATE ---1998/09/17 1998/09/17
TOTAL ------2 AVG ***** 2 AVG *****
RECORDED TIME --------
TRAN NAME ----
RESPONSE TIME -------4.808 2.404 1.495 0.747
USING (designate column headings and record fields)

USING designates column headings and record contents for each named record field specified with USING. Only one USING statement is required. The USING statement selects the column heading from an internal data dictionary. To specify your own column heading that is not in the dictionary, set a variable to a literal (maximum of eight alphanumeric characters). Then specify the variable name in the USING statement, as described in User-defined data on page 206. Column headings are printed on lines 5 and 6, with the corresponding underlines on line 7 of the report. Column headings and data appear on the report line in the order specified by the USING statement, reading from left to right. If an undefined field name or variable is specified, it does not appear on the report. If the column headings do not fit within the page width, the headings are truncated.
Syntax
USING Vn . . . Vn
where Vn represents the record field names shown in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference or user variables defined by SET.
NOTE
Depending on the record field widths, use of more than eight names can cause truncation. More than 11 field names generally cause both truncation and unpredictable results.
Chapter 10
197
Usage
Usage
To continue a list of variables on the next line, use a hyphen (-). A hyphen can be used only to continue variables to another line, not to continue statements.
NOTE
The variables listed on the USING statement have implied DEFINE statements associated with them that are used unless a previous DEFINE statement for the variable has been specified.
USING fixes the meaning of a variable for the duration of the PRL run. This occurs on the first use; subsequent uses are ignored. The effect of the USING statement is to make the variable attribute length and the decimal significance unalterable. If the variables listed in a USING statement require special editing characteristics, consider placing a DEFINE statement before the USING statement. For further information on DEFINE, see DEFINE (define variable attributesoptional) on page 175. The maximum number of variables in a USING statement is limited to the number of variables that can fit on a print line. The default print line is 132 characters long. To change the width of the print line, use
SET WIDTH=xxx
You can increase the size of the print line to up to 255 print positions.
Example
SET TITLE = 'SUMMARY TRANSACTION ROSTER' SET TITLE = 'MAINVIEW for CICS PAGE ALLOCATION MAP ANALYSIS' SELECT TYPE 17 RECORDS FROM CMRSTATS USING CMRDATE CMRTIME PAMCTLP PAMTELP PAMMIXP PAMISOP PAMSHRP PAMRPLP PAMPGMP PAMDSAT PAMTREE PAMGETVT REPORT END
198
WHERE (qualify record selection)
WHERE (qualify record selection)

WHERE qualifies record selections. If the condition is matched, control flows to the next statement.
Syntax
WHERE Vn = value
where
s
Vn is the field name of a CMRSTATS or CMRDETL record. See the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. value can be a literal character string or a defined variable.
WHERE is written after the SELECT statement and is used with any of the following operators to qualify a selected record:
= > < OR = < > equal to greater than less than not equal to greater than or equal to less than or equal to NA NA
AND
Usage
The argument in the WHERE clause can be repeated with other arguments, appropriately using the operators AND and OR. Only one variable can be qualified per line. The following example restricts the selection to activity by user DJD, at terminal SC32, with a response time that is greater than two seconds:
SELECT TYPE 6E RECORDS FROM CMRDETL WHERE T6ETMID = 'SC32' AND T6EOPID = 'DJD' AND T6ERESP > 2
Chapter 10
199
Usage
NOTE
AND must be used if NOT ( ) is specified with another qualifying condition for WHERE. An example follows: SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = '09/27/1998' AND T6ETMID = ' '
Alternative data can be selected for the same variable by adding an OR operator to the right of the WHERE expression. An example follows:
SELECT TYPE 6E RECORDS FROM CMRDETL WHERE T6ETRID = 'CSSN' OR 'CSSF' OR 'CSTT'
This example selects 6E records created by transactions CSSN, CSSF, or CSTT. This same WHERE condition could also be repeated with OR as follows:
WHERE T6ETRID = 'CSSN' OR T6ETRID = 'CSSF' OR T6ETRID = 'CSTT'
An AND operator with WHERE combines specific data from selected records. The following example selects 6E records that had activity on 09/27/1998 between 10:00 A.M. and 11:00 A.M.:
SET DATE = '09/27/1998' SET FTIME = '09:59:59' SET LTIME = '11:00:01' SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = DATE AND CMRTIME > FTIME AND CMRTIME < LTIME
These PRL statements select data from 6E records based upon the date and time specified in the CMRDATE and CMRTIME fields that
s
match the 09/27/1998 date assigned to the DATE variable have a time stamp (recorded in the CMRTIME common header field) greater than the time assigned to the FTIME variable (start at 10:00) have a time stamp less than the time assigned to the LTIME variable (end at 11:00)
Any number of AND or OR conditions can be used with a WHERE statement. However, a condition must be defined with either all ANDs or all ORs because combining them may create a logically ambiguous condition.
200
Example
If = is used with more than one AND or OR clause, the not condition inverts the logic and should be avoided as follows:
s s
Use an all positive condition. Relegate a single = to the last clause.
The following example shows the use of OR statements to select alternative data from the same field (T6ETRID - transaction identifier). An AND statement adds another required condition to the WHERE condition that further restricts data collected from the field. The expression = is used last and qualifies the record selection to identified terminals.
SELECT TYPE 6E RECORDS FROM CMRDETL WHERE T6ETRID = 'CSSN' OR 'CSSF' OR 'CSTT' AND T6ETMID = ' '
NOTE
Use AND last for NOT ( ) conditions.
Example
SET WIDTH SET TITLE = '80' = 'T E R M I N A L E R R O R S'
SELECT TYPE 6E RECORDS FROM CMRDETL WHERE T6EPGNM = 'DFHACP ' USING CMRDATE CMRTIME T6ETRID T6ETMID T6EOPID T6ECPUR T6ERESP REPORT END
Chapter 10
201
Example
202
Chapter
11
11
PRL programming techniques

This chapter describes additional programming techniques to produce specialized PRL reports. Using these techniques, you can create programs that do the following functions:
s s s s s s s s s s s s
use CICS/TS 3.2 time fields route report output format report output select archive data expedite record retrieval index detail (CMRDETL) records index by qualifying variable names produce graphs with PRL reports summarize data sort by more than one field report fourth-generation language (4GL) database activity create worst transaction response time reports
Routing report output

The output class and format of a user-defined report can be defined with a DD statement in the JCL and the PRL SET REPORTID statement. The output class and format also can be defined with the PRL statements SET CLASS and SET FORM, described in SET (assign a value) on page 191. If neither the JCL nor the PRL report allocation is specified, PERFORMANCE REPORTER allocates the report to a default output class of A and a default form of STD. The user-written PRL program (DD statement named ISYPROG) and the JCL output listing (MSGCLASS parameter of JOB statement) are allocated as specified in the MVS JCL, described in Generating PRL reports on page 159.
Chapter 11 PRL programming techniques
203
Defining output with JCL and PRL SET REPORTID
Defining output with JCL and PRL SET REPORTID

To define report output with JCL, the ddname of the DD statement must be the same as the value of the SET REPORTID statement or its default (the execution job name). For example, if the PRL program specified SET REPORTID = 'TRANAB', the name of the DD statement defining the report output must be
//TRANAB DD userParametersForReportOutput
NOTE
The user parameters must follow the rules and conventions for coding a DD statement, as described in the IBM publication, MVS Extended Architecture JCL Reference Manual.
If more than one report is created and DD statements are used to define each report output, there must be a corresponding report ID for each named DD allocation. An example follows:
//TRANAB //TRANSUM //TRANDET DD ...... DD ...... DD ......
The preceding specification allocates the report as specified in the JCL if

s
a user-written PRL program that contains a SET REPORTID = 'TRANAB' declaration for the TRANAB allocation exists a user-written PRL program that contains a SET REPORTID = 'TRANSUM' declaration for the TRANSUM allocation exists a user-written PRL program that contains a SET REPORTID = 'TRANDET' declaration for the TRANDET allocation exists
If a DD allocation for the report in the JCL exists and SET REPORTID is not specified or the value is not the same as the ddname of the report output DD statement, the JCL DD allocation is ignored. When no SET REPORTID declaration exists in a PRL program , the execution job name is the default value.
204
Defining output with PRL SET CLASS and SET FORM
Defining output with PRL SET CLASS and SET FORM

You can define report output with JCL as described in Defining output with JCL and PRL SET REPORTID on page 204 or you can use the PRL SET CLASS and SET FORM statements. However, if SET CLASS and SET FORM allocate the report output and the JCL contains a named DD statement that matches a SET REPORTID value, the report is allocated as specified by the DD statement. The output defaults to a CLASS of A and a form of STD when there is no user-specified allocation.
Formatting report output

A PRL report is formatted according to the parameters specified with the SET and USING statements. SET can be used to size and title the report and to define a variable to be reported. USING specifies the CMRDETL and CMRSTATS records to be reported as reformatted by the BBPLIB data dictionary. See the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference. The data dictionary descriptions for each record field are printed as column headings with each specified field name.
Predefined data dictionary descriptions

PERFORMANCE REPORTER provides a data dictionary in a partitioned data set called BBPLIB that automatically creates column headings for the CMRDETL and CMRSTATS record fields specified with the PRL. A description of each CMRDETL and CMRSTATS record field is stored by the PRL processor in the data dictionary. These descriptions are used in the report as column headings. The USING statement (discussed in USING (designate column headings and record fields) on page 197) and the field names of a CMRDETL or CMRSTATS record (shown in the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference) specify the data dictionary headings to be printed. For example, as shown in the Time versus activity graph on page 211, the following specification selects the CMRSTATS type CC record and specifies the data dictionary descriptions for the CMRDATE, CMRTIME, and T6FTRNCT fields:
SELECT TYPE CC RECORDS FROM CMRSTATS . . . USING CMRDATE CMRTIME T6FTRNCT REPORT END
205
User-defined data
The CMRDATE, CMRTIME, and T6FTRNCT fields correspond to the following column headings: RECORDED DATE RECORDED TIME TOTAL TASKS
User-defined data
To define data to be reported, assign a value to a variable, and then specify the variable with the USING statement. For example, the following specification prints the variable name, TRANTERM, as a column heading, and its assigned value as the column contents:
SELECT TYPE 6E RECORDS FROM CMRDETL . . SET TRANTERM = '&T6ETRID. &T6ETMID' USING TRANTERM . . REPORT END
The transaction identifier (&T6ETRID field of the CMRDETL record) and the terminal identifier (&T6ETMID field of the CMRDETL record) are reported in the TRANTERM column. The variable name must start with an alphabetic character and can comprise any alphanumeric string up to a maximum of eight characters. If the name is longer than the width of the column data, the name is truncated to the column width. PERFORMANCE REPORTER right-justifies user-defined headings and adds two blank characters at the beginning of the column heading. The column headings can be aligned with the column data either by examining the BBPLIB data dictionary or running the report. The printed report shows where to align the headings. The character string assigned to a column heading variable can then be adjusted to the width of the column data. Or the BBPLIB data dictionary can be examined for the length and type of each CMRDETL or CMRSTATS field, which together determine the column width. For a description of the type key definition, see the MAINVIEW for CICS PERFORMANCE REPORTER Data Reference.
206
Selecting archive data
Selecting archive data

As described in Archiving data (CMRPURG) on page 34, CMRDETL and CMRSTATS records can be archived to tape. The SELECT statement does not have to be changed to distinguish a tape data set from VSAM; the PRL does this automatically. Therefore, change the CMRDETL or CMRSTATS DD statement to point to a tape data set, as shown in Figure 54. Figure 54 DD statement change to point to a tape data set
//jobname JOB user parameters <-- CHANGE AS NEEDED /*JOBPARM user parameters <-- CHANGE AS NEEDED //CMRL PROC //PRL EXEC PGM=CMRPRL //STEPLIB DD DISP=SHR,DSN=CMR.CMRV3.BBLINK <-- CHANGE AS NEEDED //SYSPRINT DD SYSOUT=* //SYSUDUMP DD SYSOUT=* //SYSOUT DD SYSOUT=* //ISYPROG DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //ISYSOUT DD SYSOUT=*,DCB=(LRECL=80,BLKSIZE=160,RECFM=FBA) //PNLLIB DD DISP=SHR,DSN=CMR.CMRV3.BBPLIB <-- CHANGE AS NEEDED //CMRDETL DD DSN=CMR.MERGED.ARCHIVE,DISP=OLD,UNIT=TAPE,VOL=SER=TP0001 //SORTWK01 DD UNIT=SYSDA,SPACE=(CYL,5) //SORTWK02 DD UNIT=SYSDA,SPACE=(CYL,5) //SORTWK03 DD UNIT=SYSDA,SPACE=(CYL,5) //SYSIN DD * SET TITLE = 'TRANSACTION ANALYSIS FOR ALL TRANSACTIONS' SET CLASS = 'R' SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = '09/17/1999' SET COUNT = COUNT + 1 USING T6ETRID CMRDATE CMRTIME T6ERESP T6ECPUR T6ETMID T6EOPID ORDER BY T6ETRID DETAIL REPORT PRINT COUNT END
Expediting record retrieval

The data collected in CMRDETL and CMRSTATS is arranged in chronological order with date and time stamps. A record can be selected and the selection process can start at a specific time or date by use of STARTING AT with the SELECT statement; for example, the following specification begins selection of only CC records collected on 09/30/2004 at 8:00 A.M.:
207
Indexing detail (CMRDETL) records
SELECT TYPE CC RECORDS FROM CMRSTATS STARTING AT 08:00:00 ON 09/30/2004
NOTE
If records are being selected from an archive tape, STARTING AT has no effect and should not be used. Records can be selected from an archive tape by using WHERE. An example follows: SELECT TYPE CC RECORDS FROM CMRSTATS WHERE CMRDATE = '09/30/2004'
The archive data set name must be defined in the JCL as shown in Archiving data (CMRPURG) on page 34.

CMRDETL record file data can be indexed to obtain information from the T6E and T6D records. To report the data in these fields, use a FOR-NEXT statement to set up a counter. Within the FOR-NEXT loop, index the fields by the variable counter, assign the value of the indexed fields to variables, and report each variable using the variable names and column headings.
NOTE
A CMRDETL data set may contain two forms of T6E records, original and expanded. Be sure to check the expanded record flag (field T6EQUAL) to determine whether a given record is expanded or not. The sample program CMRPRL16, which is distributed in the BBSAMP data set, illustrates how to process both original and expanded format records in the same file.
Table 21 lists the T6E and T6D fields that can be used for indexing. Table 21
Field name T6EQUAL
T6E and T6D fields for indexing detail records (part 1 of 2)

Use format of the file data, as a decimal value of
s s
127 or less for an original format record 128 or greater for an expanded format record
T6EFCNT
number of files per task The maximum value is 40 for expanded format records and 255 for original format records.
208
Table 21
Field name T6EFI
T6E and T6D fields for indexing detail records (part 2 of 2)

Use File type, identified by
s s s s s s s s s
A (Adabas) C (CICS) D (DATACOM/DB) K (System 2000) P (MAINVIEW for CICS) S (SUPRA) U (user value) 1 (DL/I) 2 (DB/2)
T6EFN T6EFT T6EFC T6ERD T6ERU T6EWR T6ERW T6EDL T6EUL T6ESB T6ERN T6ERP T6EEB T6ERB T6EOT T6EV1 T6EV2 T6EV3
file or database name elapsed I/O request time number of I/O requests to the file number of file read or DB2 select requests number of file read for update or DB2 open requests number of file write or DB2 close requests number of file rewrite or DB2 insert requests number of file delete or DB2 delete requests number of file unlock or DB2 update requests number of file start browse or DB2 fetch requests number of file read next or DB2 commit requests number of file read previous or DB2 other requests number of file end browse requests number of file reset browse requests number of file other requests first volume serial number associated with the file second volume serial number associated with the file third volume serial number associated with the file
209
Indexing by qualified variable name
Indexing by qualified variable name

The parenthetical value used to qualify a variable is appended to the variable name. This qualified name must be used consistently by the TRANS variable within the PRL program, as shown in Figure 55. Figure 55 Qualified variable names PRL statement example
SET TITLE = '-*- TOTAL USER TRANSACTION SUMMARY -*-' SELECT TYPE 6E RECORDS FROM CMRDETL SET PGMNAM = T6EPGNM(1:4) SET TRAN = T6ETRID(1:1) SET HOUR = CMRTIME(1:2) WHERE PGMNAM = 'FACT' AND TRAN = 'C' AND CMRDATE = '09/04/1999' COMPUTE TRANS(&HOUR) = TRANS(&HOUR) + 1 USING T6ETRID T6ERESP T6EDIST T6EDISTC T6EFCWT T6EFCWTC T6ESUST ORDER BY T6ETRID SUMMARY REPORT PRINT 'TOTAL TRANSACTIONS PER HOUR' FOR &X = 00 TO 24 SET &Y = &X + 01 PRINT 'FROM' &X ': 00 TO ' &Y ': 00 TOTAL:' &TRANS(&X) NEXT &X END
Producing graphs with PRL reports

The following case studies show how to create PRL graph reports with CMRDETL or CMRSTATS data. The first case study can use any record stored on either data set. The PRL program described in the second study can use 1C records only.
210
Time versus activity graph

Member Objective Data Scenario CMRPRL12 to show graphically the activity over time within a given CICS Global Performance Records (Type CC) from the CMRSTATS file Every 5 minutes, MAINVIEW for CICS produces a Global Performance Record containing the task count for the 5minute interval (T6FTRNCT) along with other fields. PRL records these values in chronological order and represents them on a bar chart, as seen in Figure 57 on page 212.
The control statements shown in Figure 56 specify the input data and the format of the report: Figure 56 Time versus activity PRL statements
SET CLASS = 'R' SET REPORTID = 'PRL12' SET TITLE = 'T I M E V S A C T I V I T Y' SET WIDTH = '82' SET FORM = 'STD ' DEFINE GRAPH CHAR 52 DEFINE T6FTRNCT EDIT ZZZZZ9 DEFINE NUMSTAR EDIT 99999999999999 SET BB = '&vbar SELECT TYPE CC RECORDS FROM CMRSTATS WHERE CMRDATE = '09/17/1999' COMPUTE NUMSTAR = T6FTRNCT / 10 COMPUTE NUMSTAR = NUMSTAR + 1 IF NUMSTAR > 51 THEN BEGIN COMPUTE NUMSTAR = 52 END; SET &GRAPH = BB(1:&NUMSTAR) USING CMRDATE CMRTIME T6FTRNCT GRAPH REPORT END
NOTE
A variable BB is used to define a string of asterisks (*); the substring feature of the SET statement is used to pick as many asterisks (*) as needed to represent the value. The minimum activity is 15 transactions in the 5-minute period, since MAINVIEW for CICS uses this many transactions to monitor the system. T6FTRNCT was divided by 5 to extract activity for a particular day that was not busy. If the line were all asterisks, this would represent a count in excess of 250 transactions in the 5-minute interval. If your daily transaction rate is 50,000, 100,000, or more, divide by 10, 20, or even 30.
211
Figure 57
PRL12 T I M E
Time versus activity within one day

V S A C T I V I T Y VERS: REL5.4.0 TIME: 11:16:27 DATE: 09/08/1999 PAGE: 1
RECORDED DATE --------09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999 09/17/1999
RECORDED TOTAL TIME TASKS -------- -----12:14:00 12:19:00 12:24:00 12:29:00 12:34:00 12:39:00 12:44:00 12:49:00 12:54:00 12:59:00 13:04:00 13:09:00 13:14:00 13:19:00 13:24:00 13:29:00 13:34:00 13:39:00 13:44:00 13:49:00 13:54:00 13:59:00 14:04:00 14:09:00 14:14:00 14:19:00 14:24:00 14:29:00 14:34:00 14:39:00 79 94 127 147 81 104 86 102 171 147 113 232 253 199 237 229 273 383 279 312 275 399 360 401 337 424 339 347 328 288
GRAPH -------------------------------------------------|******* |******** |************ |**************** |******* |********* |******** |********* |**************** |************** |********** |********************** |************************ |******************* |*********************** |********************** |************************** |************************************* |*************************** |****************************** |*************************** |*************************************** |*********************************** |*************************************** |********************************* |***************************************** |********************************* |********************************** |******************************** |****************************
A look at the sample output in Figure 57 shows the highs and lows of the days activity. Use T6FTRMCT instead of T6FTRNCT if you are interested in only terminal activity as opposed to asynchronous tasks; add other CC fields to the USING statement as needed. Adding a particular time of day in a WHERE clause, as follows, permits you to chart, on a single page, information about peak periods, such as 12:00 P.M. to 12:59 P.M., over several months; see Figure 58 on page 213:
SELECT TYPE CC RECORDS FROM CMRSTATS SET TIME = CMRTIME(1:2) WHERE TIME = '12'
212
Service level analysis graph
Figure 58
PRL12 T I M E
Time versus activity over several months

V S A C T I V I T Y VERS: REL5.4.0 TIME: 11:05:36 DATE: 09/28/1999 PAGE: 1
RECORDED DATE --------09/03/1999 09/04/1999 09/05/1999 09/09/1999 09/07/1999 09/10/1999 10/09/1999 10/09/1999 10/13/1999 10/14/1999 10/17/1999 10/18/1999 10/19/1999 10/20/1999 10/21/1999 10/24/1999 10/25/1999 10/26/1999 10/27/1999 10/28/1999 11/02/1999 11/03/1999 11/04/1999 11/05/1999
TOTAL TASKS -----102 171 147 113 232 253 199 237 229 273 383 279 312 275 399 360 409 337 424 339 328 288 254 278
GRAPH --------------------------------------------------|********* |**************** |************** |********** |********************** |************************ |******************** |************************ |*********************** |*************************** |*************************** |************************************** |**************************** |******************************* |**************************** |**************************************** |************************************ |**************************************** |********************************** |****************************************** |*********************************** |********************************* |***************************** |********************************

Member Objective CMRPRL7 to graphically represent service levels as defined by CMRSOPT and collected by MAINVIEW for CICS
213
Data
Service Level Analysis records (Type 1C) from the CMRSTATS file Each record contains 18 service level counts.
Scenario
MAINVIEW for CICS produces service level records every hour, on the hour. CMRSOPT has 17 default service levels defined, ranging from half a second to 90 seconds. In addition, an extra service level is kept, recording response times greater than 90 seconds. The ranges of values in the various service levels can be changed through the customization macro CMRSOPT. A different recording frequency can be specified on the Service Level Analysis field of the PERFORMANCE REPORTER Statistics File Recording Frequencies screen. Note that the transaction counts only involve transactions that run at terminals and are, therefore, not biased by fast asynchronous transactions. The control statements shown in Figure 59 show how service levels can be represented. The example in Figure 59 is not intended to cover instances where there has been no activity whatsoever. Thus, the WHERE clause presumes that this time frame is idle when there has been no activity in the first three LTX values. If this assumption is incorrect for your site, repeat the OR clause with LTX004, LTX005, and so on as far as is necessary. The LTX fields in the FOR loop in the above example are represented by as many asterisks (*) as needed by substringing field BB. If there are more than 50 asterisks (*), the plus (+) is added. If you have a lot of asterisks (*) in your report, increase the frequency of 1C records on the file by reducing the interval between records through transaction FSET or by dividing the transaction count by a scaling factor. Each line containing the LTX values and their graphical representation is then printed. Ideally, the chart shows a high percentage of transactions appearing in the first service level, with the frequency rapidly diminishing as the higher service level buckets are reached. Any activity above the ninth or tenth service level slot, with a response time of seven seconds or more, is cause for concern. At this point, the time and date from the 6E records can be noted for a detailed analysis.
214
Figure 59
Service level analysis PRL statements
SET CLASS = 'R' SET REPORTID = 'PRL7' SET TITLE = 'S E R V I C E L E V E L A N A L Y S I S' SET WIDTH = '86' SET LINES = '43' DEFINE NUMTRAN EDIT Z,ZZZ,ZZ9 DEFINE NUMSTAR EDIT 99999999999999 SET BB = |***************************************************+ SET GRAPHS = '|....5....10...15...20...25...30...35...40...45...50' SET GRAPH = ' ' READ: SELECT TYPE CA RECORDS FROM CMRSTATS WHERE CMRDATE = '09/17/2007' AND LTX001 > 0 OR LTX002 > 0 OR LTX003 > 0 USING CMRDATE CMRTIME PRINT ' ' PRINT CMRDATE CMRTIME ' ' GRAPHS PRINT ' RESPONSE COUNT | (SCALE = 1 : 5 (ROUNDED))' FOR &IX = 001 TO 018 COMPUTE NUMSTAR = LTX(&IX) / 5 COMPUTE NUMSTAR = NUMSTAR + 1 IF LTIM(&IX) = 0 THEN SET LTIM(&IX) = ' 100.000' IF NUMSTAR > 51 THEN BEGIN COMPUTE NUMSTAR = 52 END; SET STARS = BB(1:&NUMSTAR) COMPUTE NUMTRAN = LTX(&IX) PRINT LTIM(&IX) NUMTRAN STARS NEXT &IX GOTO READ REPORT END
As shown in the report in Figure 60 on page 216, the CMRDATE and CMRTIME field contents are reported when the REPORT statement is processed. The headings, RESPONSE, COUNT, and the scale are printed when the PRINT statement is processed. Enclosing a character string in single quotation marks defines the characters to be printed. To print the value of the variable, specify a defined variable, not enclosed in single quotation marks, with PRINT.
NOTE
To use the PRINT statement, as in this example, the number of data lines per page must be defined, for example, SET LINES=21, and the reporting of data within those lines must be controlled, for example, FOR &IX = 001 TO 018. SET LINES determines each page break and generates a new page with titles and column headings.
215
Figure 60
PRL7 S E R V I C E
Service level analysis output

L E V E L A N A L Y S I S VERS: REL5.4.0 TIME: 12:10:56 DATE: 09/17/1999 PAGE: 1
RECORDED RECORDED DATE TIME -------- --------
09/17/1999 15:59:00 RESPONSE COUNT 0.500 4,687 1.000 281 1.500 153 2.000 123 2.500 36 3.000 5 4.000 2 5.000 2 6.000 2 7.000 4 8.000 2 9.000 2 10.000 3 20.000 12 30.000 0 60.000 0 90.000 0 34359.738 0
|....5....10...15...20...25...30...35...40...45...50 | (SCALE = 1 : 5 (ROUNDED)) |**************************************************+ |**************************************************+ |******************************* |************************* |******* |* | | | |* | |* |** | | | | |
216
Summarizing data
Summarizing data
Each selected record is reported on one line, which can cause the report to become quite large. Use the PRL statement SUMMARY to report all of the records on just two lines whenever there is a control-break in the report, as shown in Figure 61 and in Figure 62 on page 218. A control-break is the point in a report when the first variable specified with USING changes. For this reason, it is best to always make the first USING variable the same as the ORDER BY variable. Figure 61 SUMMARY PRL statement
SET CLASS = 'R' SET REPORTID = 'REPORT1' SET TITLE = 'TRANSACTION ANALYSIS FOR ALL TRANSACTIONS' SELECT TYPE 6E RECORDS FROM CMRDETL USING T6ETRID T6ERESP T6ECPUR T6EFCWT T6ESHWM T6EPAGCT ORDER BY T6ETRID SUMMARY REPORT END
The PRL for this report is in the CMRPRL3 member of the sample library (BBSAMP). As shown in Figure 62, the first of the two summary lines contains totals for all numeric fields. The second line contains the computed averages. The totals and averages are shown whenever there is a change in the first reported variable. These two lines of totals and averages also are shown when the PRL statement DETAIL is used. One selected record is reported per line, and when the first variable changes, the totals and averages are reported as SUMMARY.
NOTE
Character data only can be included in a summary report if it is specified as the first USING variable.
217
Sorting by more than one field
Figure 62
REPORT01
Transaction Analysis for All Transactions summary report

TRANSACTION ANALYSIS FOR ALL TRANSACTIONS VERS: REL5.4.0 TIME: 17:03:11 DATE: 09/17/1999 PAGE: 1
TRAN NAME TOTAL --------------CSGM 52 AVG ***** 6 AVG ***** 13 AVG ***** 33 AVG ***** 4 AVG ***** 5 AVG ***** 16 AVG ***** 3 AVG ***** 6 AVG ***** 28922 AVG ***** 4 AVG ***** 72 AVG ***** 56 AVG ***** 929 AVG ***** 62 AVG ***** 5360 AVG ***** 30 AVG ***** 16 AVG ***** 15 AVG *****
RESPONSE TIME ------------29.530 0.568 4.380 0.730 0.418 0.032 24.234 0.734 2.339 0.585 0.206 0.041 9.879 0.617 7.617 2.539 5.147 0.858 6788.137 0.235 2.539 0.635 15.729 0.218 156338.884 2791.766 10020.464 10.786 19.769 0.319 308.284 0.058 964.988 32.166 72.636 4.540 160.559 10.704
CPU REAL TIME ------------0.264 0.005 0.062 0.010 0.022 0.002 0.496 0.015 0.021 0.005 0.017 0.003 0.224 0.014 0.756 0.252 0.088 0.015 132.645 0.005 0.014 0.003 1.789 0.025 13.032 0.233 4.830 0.005 0.308 0.005 10.726 0.002 0.327 0.011 0.235 0.015 0.262 0.017
FILE_CONTROL I/O TIME ------------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 2275.220 0.079 0.000 0.000 8.092 0.112 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
STORAGE HI_WATER_MARK --------------354984 6827 72952 12159 102048 7850 321368 9738 31776 7944 39432 7886 99040 6190 39328 13109 88624 14771 305450768 10561 54400 13600 1259048 17487 409264 7308 599785928 645625 436824 7046 32212464 6010 911808 30394 483792 30237 455016 30334
TOTAL PAGE_COUNT --------------58 1 13 2 18 1 137 4 7 2 0 0 1 0 37 12 8 1 14799 1 28 7 23 0 397 7 5 0 33 1 2 0 17 1 14 1 3 0
CSMT
CSPG
CSSF
CSSN
CSST
DB2
DISP
FCD2
FIC2
FST2
HST2
JNL2
JPH1
SMN2
STAR
TSM2
TSM4
TSM5
Sorting by more than one field

To sort by more than one field and report these fields:
s s
Set a variable to more than one field. Specify the variable name with USING and ORDER BY.
218
Reporting 4GL database activity
An example is shown in Figure 63. Figure 63 Set variables to sort by more than one field PRL statement
*************************************************************** * * * This report selects transaction records and then places * * them into one of ten response ranges. These ranges are * * then summarized to show resource usage for a given * * transaction and response range. * * * *************************************************************** SET CLASS = 'R' SET REPORTID = 'REPORT1' SET TITLE = 'TRANSACTION RESPONSE ANALYSIS BY RESPONSE RANGE' SELECT TYPE 6E RECORDS FROM CMRDETL SET &VAR2 = '9> ' IF &T6ERESP < 9 THEN SET &VAR2 = '8-9' IF &T6ERESP < 8 THEN SET &VAR2 = '7-8' IF &T6ERESP < 7 THEN SET &VAR2 = '6-7' IF &T6ERESP < 6 THEN SET &VAR2 = '5-6' IF &T6ERESP < 5 THEN SET &VAR2 = '4-5' IF &T6ERESP < 4 THEN SET &VAR2 = '3-4' IF &T6ERESP < 3 THEN SET &VAR2 = '2-3' IF &T6ERESP < 2 THEN SET &VAR2 = '1-2' IF &T6ERESP < 1 THEN SET &VAR2 = '0-1' SET &VAR3 = T6ETRID SET ID-RANGE = '&VAR3. &VAR2. ' USING ID-RANGE T6ERESP T6ECPUR T6EDIST T6EFCWT T6ESUST T6ETSWT T6EPCWT ORDER BY ID-RANGE SUMMARY REPORT END
This technique is used in the case study described in Creating longest transaction response time reports on page 220.
Reporting 4GL database activity

To retrieve and process the fourth-generation language (4GL) database activity collected in the CMRDETL and CMRSTATS data sets, use PRL and JCL DD statements to specify the input for your reports. The databases and 4GLs are identified by the following field names used in PRL statements:
s
T6EPGNM specifies the program name of any 4GL or database (except MANTIS, which is identified through T6EMNTIS) represented by T6EPTYPE.
219
Creating longest transaction response time reports
T6EPTYPE specifies the program name type identified by a 1-character database or
4GL designation.
s
T6EFI specifies the file type identified by a 1-character file or database name
represented by T6EFN.
s
T6EFN specifies the file or database name.
For example, the statements shown in Figure 64, submitted as an MVS batch job, report on Software AG Natural performance on a given day. Figure 64 NATURAL performance PRL statements
SET TITLE = 'NATURAL RESPONSE TIMES' SET TITLE1 = 'On 09/17/1999' SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = '09/17/1999' AND T6EPTYPE = 'N' USING CMRTIME T6EPGNM T6EUSER T6ERESP T6EFCNT REPORT END -

Detailed transaction information stored on the CMRDETL data set can be used to create daily reports. Sample member CMRPRL14 of your BBSAMP data set creates a report from 6E detail transaction records. This report shows the 30 longest transactions recorded over a daily interval.
220
Figure 65 shows sample PRL statements from member CMRPRL14 to create a report of the transactions with the longest response time. In this example, the control statements restrict the report to terminal transactions that occurred on a single day. Figure 65 Longest transaction response time PRL statements
SET CLASS = 'R' SET REPORTID = 'PRL14' SET TITLE = 'WORST RESPONSE TIMES' SET TITLE1 = 'RESTRICTED TO 09/18/1999 ' SET TITLE2 = 'TERMINAL ATTACHED TRANSACTIONS SELECT TYPE 6E RECORDS FROM CMRDETL WHERE CMRDATE = '09/18/1999' AND T6ETMID = ' ' COMPUTE LINE_I/O = T6ECHARI + T6ECHARO USING CMRTIME T6ETRID T6EPGNM T6ERESP T6ECPUR T6ETMID T6EOPID T6EFCWT T6EFCCAL LINE_I/O ORDER BY T6ERESP IN DESCENDING SEQUENCE SET COUNT = COUNT + 1 IF COUNT > 30 THEN GOTO DDD REPORT DDD: PRINT ' * * * * * * * * * * * PRINT ' * PRINT ' * W O R S T 3 0 R E S P O PRINT ' * PRINT ' * * * * * * * * * * * END
ONLY'
N S E S * * * * *
* * * * *
' ' ' ' '
Figure 66 on page 222 shows the left-most portion of an example report created by these PRL control statements. This report lists 30 transactions with the longest response times over a daily period.
221
Figure 66
PRL14
Longest transaction response times report

VERS: REL5.4.0 TIME: 14:38:01 DATE: 09/19/1999 PAGE: 1 TERM ID. --OPR ID. --DJD RGN RGN RGN DJD FILE CONTROL I/O TIME --- ---0.377 0.000 0.000 0.000 0.000 0.000 2.095 0.000 0.000 0.000 4.279 0.000 0.000 0.068 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 3.823 0.248 0.000 0.000 0.000 0.000
WORST RESPONSE TIMES RESTRICTED TO 09/18/1999 TERMINAL ATTACHED TRANSACTIONS ONLY
RECORDED TIME ---10:22:23 16:55:30 13:10:21 16:55:42 16:55:58 17:14:10 10:22:45 17:08:02 12:03:52 17:06:45 15:58:43 17:02:34 15:58:35 11:28:59 16:13:18 13:27:07 12:02:53 16:52:40 16:10:26 10:22:08 17:14:06 16:04:32 15:58:41 16:42:13 15:28:49 17:12:50 16:04:31 16:13:05 13:28:47 13:31:58 * * * * W * * *
TRAN NAME ---DGNH PCXR DRVA GRSD PCXR GRSD DGNH GRSD FSET GRSD DGNH GRSD GRSD CSPK GRSD CEMT CSGM PCXR GRSD PCXR PCXR GRSD GRSD GRSD DGNH PCXR PCXR FRED CEMT GRSD * *
PROGRAM NAME ---LDG0000 PCXRDRVR ADRIVER GRS2500 PCXRDRVR GRS2500 SYSEDIR GRS2500 CMRMAP GRS2500 SYSEDIT DFB20001 GRS2500 DFHPRK GRS2500 DFHEMTP DFHGMM PCXRDRVR GRS2500 PCXRDRVR PCXRDRVR GRS2500 GRS2500 GRS2500 LDG0000 PCXRDRVR PCXRDRVR GRS2500 DFHEMTP GRS2500 * * * 3 0 * * * *
RESPONSE TIME ---33.515 20.658 16.632 16.577 13.263 11.990 8.708 8.537 8.370 7.713 7.201 6.490 6.402 5.696 5.368 4.620 4.612 4.494 4.418 4.270 4.253 4.151 4.034 3.984 3.949 3.907 3.615 3.570 3.561 3.559 * * *
CPU REAL TIME ----
O R S T * *
R E S P O N S E S * * * * * *
0.629 SC32 0.028 L072 0.156 L072 0.387 SB24 0.027 L072 0.384 SB24 0.398 SC32 0.388 SB24 0.078 SB24 0.405 SB24 0.087 SC32 0.020 SC26 0.418 SB24 0.113 L064 0.418 SB24 0.953 L072 0.017 SB24 0.026 L072 0.384 SB24 0.022 L064 0.025 L072 0.385 SB24 0.386 SB24 0.436 SB24 0.098 SC32 0.035 L072 0.021 L072 0.380 SB24 0.929 L072 0.390 SB24 * * * * * * * * *
DJD
RGN RGN
RGN
DJD RGN RGN RGN
Transactions are sorted in descending order by response time. The longest transaction is at the top of the report.
222
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Index
Symbols
& reserved word 172, 192 FCT 100 JCT 102 overview 97 PCT 100 PPT 99 SERVICE 98 TCT 104 TST 105 CMRASM 79 CMRCHRT batch program 91 CMRCHRT reports control statements CLASS 105 DATA 97, 105 END 105 keyword 9496 overview 93 creating 9192 description 91 examples 106110 JCL 9192 CMRCHRT reports control statements CLASS 97 CMRCJCL 31 CMRCMPR 45 CMRCMPR decompression program 48 CMRCMPRN 46 CMRCMPW 45, 138 CMRCMPW decompression program control statements 48 CMRCMPWN 39, 45, 138 CMRCMPWN and T6E conversion 39 CMRCMPWN decompression program description 46 CMRCOB11 124 CMRCOB11 program 128 CMRCOB11N 128 CMRCOB12 124 CMRCOB12 program 129 CMRCOB12N 129 CMRCOB13 124 CMRCOB13 program 130 CMRCOB13N 130 CMRCOB14 124 CMRCOB14 program 131 CMRCOB14N 131
A
abended transactions, sample PRL report 155 activity for a specific program, sample PRL summary report 155 ALARM reserved word 172 ALTERVS reserved word 172 AND reserved word 172 APPL CLASS control statement 97, 98 applications class control statement 97, 98 APPLID control statement 44 archive data, selecting 207 archiving CMRDETL data single data set 37
B
BBPLIB data set specifying in JCL 160 using with PRL 205 BBSAMP PRL sample programs 155 BBSAMP data set using 22 BLINK reserved word 172 BMC Software, contacting 2
C
CALC=PERCENT 101 CCMD reserved word 172 CDATE reserved word 158, 172 CICS 110 SMF format. 38 CICS CLASS control statement 98 CLASS control statements APPL 97, 98 CICS 98 CSA 101 CSO 102 DCT 102 DLZ 103
Index
223
CMRCOB15 124 CMRCOB15 program 132 CMRCOB15N 132 CMRCOB21 124 CMRCOB21 program 133 CMRCOB21N 133 CMRCOB23 124 CMRCOB23 program 135 CMRCOB23N 135 CMRCOB25 124 CMRCOB25 program 136 CMRCOB25N 136 CMRCOBCP 118 CMRCOBCP program 120121 CMRDETL data set accessing for COBOL reports 118121 archiving single 37 conversion to CICS 110 SMF 38 data types altered by CMRSUMD 74 description 28, 31 fields altered by CMRSUMD 7374 record types 31 reformatting compressed records 4548 summarizing records 4975 CMRDFSUM 51, 73 CMRFRED program 118119 FT messages 119 CMRFREDN 117 CMRL JCL member 154, 159161 CMRLTXR reports control statements 112 example 112 field descriptions 113 JCL 111112 CMRN2MCT 39 CMRN2SAS SAS field identifications 40 CMRNDATA 40 CMRNDB2 40 CMRNDMCT 39 CMRNDSAS 40 CMRNIFUE 138 CMRPRL sample programs 155, 156 CMRPRL28 66 CMRPRLN 153 CMRPRLn members in BBSAMP 155 CMRPURG 34 CMRPURG program purging CMRSUMD records 3738 CMRPURGE JCL member 34 CMRRAP 79 CMRRAPM 79 CMRRAPM macro 8182 CMRRAPR 79 CMRRAPR report 79 CMRRAPR reports control statements 8384 description 79 example 8486 field descriptions 8789 JCL 7980 macro assembly statements 8182 CMRSAS10 program 137 CMRSAS10N 137 CMRSAS11 139 CMRSAS11 program 141 CMRSAS11N 139, 141 CMRSAS12 139 CMRSAS12 program 142 CMRSAS12N 139, 142 CMRSAS13 139 CMRSAS13 program 143 CMRSAS13N 139, 143 CMRSAS14 139 CMRSAS14 program 144 CMRSAS14N 139, 144 CMRSAS15 139 CMRSAS15 N 145 CMRSAS15 program 145 CMRSAS15N 139 CMRSAS21 139 CMRSAS21 program 146 CMRSAS21N 146 CMRSAS23 program 147 CMRSAS23N 147 CMRSAS25 program 148 CMRSAS25N 148 CMRSAS6F program 149150 CMRSAS6FN 149 CMRSASxx 138 CMRSAxxN 138 CMRSIFUE 138 CMRSOPT startup table description 32 SERVICE control statement 111 CMRSTATS batch program 27 CMRSTATS data set converting records with CMRSTATS 4145 description 3234 record types 34 CMRSTATS program control statements 4344 error messages 4445 JCL 41 CMRSTATX JCL member 34 CMRSTJCL JCL member 41 CMRSUMD utility 31, 37 CMRDETL fields altered 7374 combining summarized records 74 control statements coding 51 COPY 62 COPYAPPL 62 DETAIL 58
224
CMRSUMD utility (continued) EXCLxxxx 56, 57 FILES 58 FROM DATE 59 FROM TIME 60 INCLxxxx 57 MAXCPU 62 MAXRESP 61 MINCPU 61 MINRESP 61 MINUTES 54 NEWAPPL 57 overview 51 PROCFLIO 63 REPLACE 58 SGROUP 56 SKIPFLIO 63 START TIME 60 STOP TIME 6061 TABLE 62 TO DATE 59 TO TIME 60 TYPE 55 USEREXIT 63 WORKLOAD 54 ZERODATA 59 creating PRL reports 6668 data types altered 74 description 4950, 53 detail file switch 70 JCL 5051, 64 purging summary records 3738 replacing summary data 71 sample jobstream 64 sample output 6566 selecting input records 6970 storing summary data 72 usage notes 6975 viewing summary records 7172 CMRSUMD1 JCL member 64 CMRTIME fields 200 CMRURAPM 79 CMRUSRD 63 COBOL reports accessing CMRDETL records 118121 CMRCOBCP program 120121 CMRFRED program 118119 common fields 127128 control statements 126 customizing 121123 description 117 examples Abended Transaction Report 136 File Analysis for All Files 133 overview 123124 Performance Analysis by Netname 131 Performance Analysis by Program 129 Performance Analysis by Terminal 130 Performance Analysis by User ID 132 Performance Analysis for All Transactions 128 Rogues Report 135 JCL 125127 collecting data 2734 COLOR reserved word 172 Common System Area class control statement 101 compressed CMRDETL records, reformatting 4548 COMPUTE PRL statement 173175 COMPUTE reserved word 172 control statements CLASS 97, 105 CMRSUMD utility 51, 63 COBOL 126 DATA 97, 105 histogram 93105 PRL 169170 Resource Tracking and Analysis 8384 SAS 140 Service Level Analysis 112 conventions, documentation 18 converting CMRSTATS records 4145 counts by record types from CMRDETL and CMRSTAT files sample PRL report 155 CSA CLASS control statement 101 CSO CLASS control statement 102 CTIME reserved word 172 CURSOR reserved word 172 customer support 3 CYCLE reserved word 172
D
DATA control statements applications class 9798 CICS system class 99 Common System Area class 101102 Destination Control Table class 102 DL/I class 103104 File Control Table class 101 Journal Control Table class 103 overview 97 Processing Program Table class 99 Program Control Table class 100 Temporary Storage Table class 105 Terminal Control Table class 104105 Transaction Class Table class 102 data dictionary specifying in JCL 160 using with PRL 205 Data Language/I (DL/I) CLASS control statement 103 DB2 COBOL File Analysis for All Files report 133 SAS File Analysis for All Files report 146
Index
225
DB2 data, from variable file entries 155 DBC$DATN-DBC$DATS 40 DBC$GH 40 DBC$GHN 40 DBC$GHNP 40 DBC$GHU 40 DBC$GNP 40 DBC$GU1 40 DBC$SCHT/16 40 DBC$TMIO/16 40 DBC$TOTC 40 DBCTL, sample PRL report 155 DCB, default 35 decompressing CMRDETL records 4548 DEFINE PRL statement 175176 DEFINE reserved word 172 DELAY reserved word 172 Destination Control Table (DCT), CLASS control statement 102 Destination Control Table Report 24 detail file switch 70 DETAIL PRL statement 177178 DETAIL reserved word 172 DFH$DLET 40 DFH$ISRT 40 DFH$REPL 40 DFHMNTDS dsect 40 DFSDSTA control block 40 DIAGNOSE PRL statement 178179 DIAGNOSE reserved word 172 DL/I class control statements 103 DLZ CLASS control statement 103 Dnnnnnnn reserved word 172 documentation, related 16 DUMP reserved word 172 FIND reserved word 172 FOR and NEXT PRL statements description 180181 examples 181, 208 FOR reserved word 172 FORMAT statement 46 formatting PRL output 205 fourth-generation language (4GL) COBOL report 129 PRL database activity report 219220 SAS report 142 FROM DATE control statement 44 FROM TIME control statement 44 FT256 message 119 FT257 message 119 FT258 message 119 FT260 message 119 FT261 message 119 FT582I message 44 FT583I message 45 FT584I message 45 FT585I message 45
G
GAPPLID control statement 44 GCMD reserved word 172 GET reserved word 172 GOTO PRL statement 182183 GOTO reserved word 172 GRAF reserved word 172 graph reports see also histogram reports using PRL 210216 Graphic Histogram Report 25
E
electronic documentation 15 EMNU 86 END PRL statement 179180 END reserved word 172 error messages, CMRSTATS 4445 EXCLxxxx control statement 69 EXEC reserved word 172 EXIT reserved word 172
H
HEADING reserved word 172 HELP reserved word 172 Help, online 15 highest users of CPU time, sample PRL report 155 HIGHLIGHT reserved word 172 histogram reports creating 9192 description 91 examples 106110 JCL 9192 histogram reports control statements CLASS 97, 105 DATA 97, 105 END 105 keyword 9496 overview 93 HISTORY reserved word 172
F
FACT reserved word 172 FCMD reserved word 172 file control sample PRL detail report 156 sample PRL summary report 156 File Control Table (FCT), CLASS control statement 100
226
HISTORY service 49, 7172
P
PAD reserved word 173 PANELID reserved word 173 PCT class control statement 100 PERFORMANCE REPORTER archiving data 3438 collecting data 2734 description 2123 report types 2426 Performance Reporting Language (PRL) control statement verbs COMPUTE 173175 DEFINE 175176 DETAIL 177178 DIAGNOSE 178179 END 179180 example 161, 165167 FOR and NEXT 180181 GOTO 182183 IF and THEN 183184 LET 185 ORDER BY 185186 overview 169170 PRINT 187 REPORT 188 SCAN 188189 SELECT 189191 SET 191195 SUMMARY 195197 syntax 170172 USING 197198 WHERE 199201 JCL 159161 operators 156 overview 153154 processing flow 162167 program syntax 154159 programming techniques creating response time reports 220222 formatting report output 205 indexing by qualified variable 210 indexing CMRDETL records 208209 producing graphs 210216 reporting 4GL database activity 219220 retrieving selected records 207208 routing report output 203205 selecting archive data 207 sorting data 218219 summarizing data 217218 reserved words 172 sample programs 155, 156 substrings 156157 using CMRSUMD utility 6668 variables 158159 PRINT PRL statement 187 PRINT reserved word 173
I
I/O activity sample PRL report 155 IF and THEN PRL statements 183184 IF reserved word 172 INCLxxxx control statement 69 indexing with PRL CMRDETL records 208209 by qualified variable 210
J
JCL members CMRLJCL 154 CMRPURGE 34 CMRSTATX 34 CMRSTJCL 41 CMRSUMD1 64 Journal Control Table (JCT), CLASS control statement 102
L
LEAVE= 35 LET PRL statement 185 LET reserved word 172 LIN001 reserved word 172 LOGOFF reserved word 173
M
MAINVIEW for CICS Interactive Guide 17 MODNAME value 79 MQ data from variable file entries sample PRL report 155
N
NEW reserved word 173 NEWAPPL control statement 71 NEXT reserved word 173
O
online Help 15 OR reserved word 173 ORDER BY PRL statement description 185186 examples 186, 218219 ORDER reserved word 173
Index
227
PRL, See Performance Reporting Language 153 Processing Program Table (PPT), CLASS control statement 99 product support 3 PROFCNT reserved word 173 PROFMEM reserved word 173 PROFRC reserved word 173 program control detail sample PRL report 155 summary sample PRL report 156 Program Control Table (PCT), CLASS control statement 100 program timing data sample PRL report 155 publications, related 16 PUT reserved word 173 reserved words 172 Resource Analysis Report 25 Resource Tracking and Analysis reports control statements 8384 description 79 example 8486 field descriptions 8789 JCL 7980 macro assembly statements 8182 resource usage, summary sample PRL report 155 response time analysis by response range sample PRL report 155 for MRO data sample PRL report 155 for terminal attached transactions sample PRL report 155 RETURN reserved word 173 REVERSE reserved word 173 Rogues Reports COBOL 135 SAS 147 routing PRL output 203205 RVIDEO reserved word 173
R
RAND reserved word 173 records creating CICS statistical 4145 purging CMRSUMD 3738 SMF 110 34 Type 6D and 6E transaction detail 31 reformatting compressed CMRDETL records 4548 REPLACE= 71 REPORT PRL statement 188 REPORT reserved word 173 reports creating 2123 custom 26 histogram 91110 long-term 25 predefined 2425 Resource Tracking and Analysis 7989 Service Level Analysis 111113 short-term 24 reports, COBOL Abended Transaction Report 136 File Analysis for All Files 133 Performance Analysis by Netname 131 Performance Analysis by Program 129 Performance Analysis by Terminal 130 Performance Analysis by User ID 132 Performance Analysis for All Transactions 128 Rogues Report 135 reports, SAS Abended Transaction Roster 148 File Analysis for All Files 146 Operational Statistics Report 149150 Performance Analysis by Node Name 144 Performance Analysis by Program 142 Performance Analysis by Terminal 143 Performance Analysis by User ID 145 Performance Analysis for All Transactions 141 Rogues Report 147
S
SAPPLID control statement 44 SAS reports Abended Transaction Roster example 148 common fields 140141 control statements 140 description 137 File Analysis for All Files example 146 Operational Statistics Report example 149150 overview 137140 Performance Analysis by Node Name example 144 Performance Analysis by Program example 142 Performance Analysis by Terminal example 143 Performance Analysis by User ID example 145 Performance Analysis for All Transactions example 141 Rogues Report example 147 SASI6ExN 138 SCAN PRL statement 188189 SCAN reserved word 173 SCMD reserved word 173 SELECT PRL statement description 189 examples 191, 207208 SELECT reserved word 173 SERVICE CLASS control statement 98 service level analysis graph, sample PRL report 155 Service Level Analysis reports control statements 112 description 111 example 112 field descriptions 113
228
Service Level Analysis reports (continued) JCL 111112 sample PRL report 155 session data sample PRL report 155 SET PRL statement 191195 SET reserved word 173 SMF 110 records 34 SQL activity, by DB2 plan PRL report 155 summarization utility 69 summarized data from CMRSUMD sample PRL report 155 summarizing CMRDETL records 4975 SUMMARY PRL statement description 195 examples 196, 217218 SUMMARY reserved word 173 SUMMDATA 71 support, customer 3 SYMBOLIZE reserved word 173 system management facility (SMF) 32 terminal control sample PRL detail report 156 sample PRL summary report 156 Terminal Control Table (TCT), CLASS control statement 104 terminal errors, sample PRL report 155 TO DATE control statement 44 TO TIME control statement 44 TRACE reserved word 173 transaction analysis for all transactions, sample PRL report 155 transaction class CLASS control statement 102 transaction class sample PRL report 156 Transaction Class Table class control statements 102 transaction file control wait analysis, sample PRL report 155 transaction response time, sample PRL report 155 TRIM reserved word 173 TROW reserved word 173 TUTOR reserved word 173 Type 6D and 6E transaction detail records overview 31
T
T6EDL 209 T6EEB 209 T6EFC 209 T6EFCNT 208 T6EFI 209 T6EFN 209 T6EFT 209 T6EOT 209 T6EQUAL 208 T6ERB 209 T6ERD 209 T6ERN 209 T6ERP 209 T6ERU 209 T6ERW 209 T6ESB 209 T6EUL 209 T6EV1 209 T6EV2 209 T6EV3 209 T6EWR 209 T6FTRNCT 205 TABLE control statement 79 TCOL reserved word 173 technical support 3 temporary storage sample PRL detail report 156 sample PRL summary report 156 Temporary Storage Table (TST), CLASS control statement 105 terminal attached transactions sample PRL report 155
U
UNDERSCORE reserved word 173 Unnnnnnn reserved word 173 USEREXIT control statement 70 USEREXIT= 66 USING PRL statement description 197 examples 198, 205 USING reserved word 173
V
VSAM storing CMRSUMD data 72
W
WHERE PRL statement 199201 WHERE reserved word 173
Z
ZCARD reserved word 173 ZCICS reserved word 173 ZCSA reserved word 173 ZCT reserved word 173 ZCURSOR reserved word 173 ZCW reserved word 173 ZDATE reserved word 158, 173
Index
229
ZERRMSG reserved word 173 ZIFX reserved word 173 ZIFY reserved word 173 ZLAST reserved word 173 ZLINE reserved word 173 ZLINES reserved word 173 ZLOC reserved word 173 ZMEMBER reserved word 173 ZMSG reserved word 173 ZOPTERR reserved word 173 ZPF reserved word 173 ZREQUEST reserved word 173 ZSCREEN reserved word 173 ZSHOW reserved word 173 ZTIME reserved word 158, 173 ZTITLE reserved word 173 ZTRAN reserved word 173 ZUSER reserved word 173 ZWRAP reserved word 173 ZXDT reserved word 173
230
Notes
*100606* *100606* *100606* *100606*

*100606*

100606

Uploaded by

Copyright:

Available Formats

100606

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

100606

Uploaded by

Copyright:

Available Formats

MAINVIEW for CICS

PERFORMANCE REPORTER User Guide

Contacting BMC Software

United States and Canada

Outside United States and Canada

Restricted rights legend

Support by telephone or e-mail

Before contacting BMC

License key and password information

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Introduction and report data

MAINVIEW for CICS reports

COBOL and SAS reports

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Performance Reporting Language reports

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

About this book

About this book

MAINVIEW Common Customization Guide

MAINVIEW for CICS Customization Guide

MAINVIEW Security Guide

MAINVIEW Security Reference Manual

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Category core documents

Document MAINVIEW for CICS Getting Started

MAINVIEW for CICS Online Services Reference Manual

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

describes how to produce a variety of batch reports, including

MAINVIEW User Guide

MAINVIEW Alarm Management Guide

notices and other documents

About this book

Online services documentation

Online services documentation

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Introduction and report data

Introduction and report data

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Introduction to PERFORMANCE REPORTER

Chapter 1 Introduction to PERFORMANCE REPORTER

Report creation process

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Running report programs

Running report programs

Chapter 1 Introduction to PERFORMANCE REPORTER

An overview of PERFORMANCE REPORTER reports

An overview of PERFORMANCE REPORTER reports

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Chapter 1 Introduction to PERFORMANCE REPORTER

Custom PRL reports

Custom PRL reports

MAINVIEW for CICS PERFORMANCE REPORTER User Guide

Collecting and archiving report data

Collecting report data

Collecting and archiving report data

Transaction detail data (CMRDETL)

Transaction detail data (CMRDETL)

MAINVIEW for CICS PERFORMANCE REPORTER User Guide