IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

46
© 2013 Fundi Software The (R)Evolution Continues IMS12 – Updates to IMS Performance Tools Rafael Avigad, Fundi Software

Transcript of IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Page 1: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

© 2013 Fundi Software

The (R)Evolution Continues

IMS12 – Updates to IMS Performance Tools Rafael Avigad, Fundi Software

Page 2: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Disclaimer: Information regarding potential future products is intended to

outline our general product direction and it should not be relied on in making a purchasing decision. The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any material, code or functionality. Information about potential future products may not be incorporated into any contract. The development, release, and timing of any future features or functionality described for our products remains at our sole discretion.

Page 3: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Dimensions of performance management

• Empowerment • Audit and control

• Transparency • Communication

• Optimizing • Redundancy • Automation

• Monitoring • Identifying

problems

Reactive Proactive

Development Operations

3

Page 4: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Challenges for tools

4

Collection • Finding

performance artefacts

• Actively capturing information

Attribution • Mining

performance information

• Providing innate knowledge

• Attribution

Access • Guidance • Availability from

non traditional contexts

Resolution • Autonomics • Instrumentation

Page 5: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Tools for the challenges

IMS Connect Extensions: manage TCP/IP access to IMS OTMA and IMS database (ODBM)

Transaction Analysis Workbench: Analyze transactions at the business level: across IMS, DB2, and CICS

IMS Performance Analyzer: report on IMS transaction and system performance and health

IMS Problem Investigator: interactively interrogate and interpret IMS and related logs

5

Page 6: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

6

Time to market • Increase connectivity options • Directly access transactions and data • Utilize familiar SQL syntax

Performance and cost saving • Reduce network costs • Improve performance • Simplify capacity management

Development challenges • Chasm between IMS skills and Java skills • Control and management of sandbox

systems • Tracing and tracking requests

Broader challenges • Disruptive to existing processes • Performance monitoring and optimizations

requirements • Organizational impact

IMS Connect

Page 7: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

7

Event collection • Information about transaction and open database

activity • Basic reporting • Integration with IMS system programmer tools (PA

and PI)

Routing • Improves availability • Insulates developers from the network • Allows some problems to be fixed in real time

Centralized management • Instrumentation • Enhanced security • Control of traffic options

Eclipse GUI • Monitoring system status • Managing in-flight sessions • Controlling IMS Connect instances

IMS Connect Extensions

Page 8: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

A single workspace to manage and control various projects

Project A

IMS

IMS Connect

IMS

IMS

IMS Connect

Project B

IMS

IMS Connect

IMS

IMS

IMS Connect

File Option Help

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

_ Stop IMS 1 link

_ View sessions

_ Stop datastore

F1 – Help F2 - Scroll

Eclipse or ISPF

8

Page 9: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

9

Any IMS Connect instance from anywhere that is accessible via TCP/IP

Logical groups reflecting projects, teams, geographies

Control systems datastore, aliases, ports, exits, and more

Activate traces

View basic statistics

Page 10: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Single consolidated view of sessions running on all systems (or view activity for individual systems or user-defined groups)

Extensive details on individual sessions

Highlight sessions matching any criteria Export to CSV

Open database session

View your IMS Connect session activity in real time

Page 11: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

11

Messages Routing rule

OTMA TCB TCB TCB TCB

IMS Connect Datastore E Datastore F Datastore G Datastore H

Logical target Physical target

Transaction routing: logical partitioning

Converts a logical destination to a

physical destination

OTMA TCB TCB TCB TCB

IMS Connect Datastore A Datastore B Datastore C Datastore D

Simple setup and configuration

Page 12: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Centrally-managed

client options

Transaction expiration

IRM Timer values

Message translation

maps Client ID

cancellation

Extended RSM

feedback

Simplify client management

Page 13: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

© 2012 Fundi Sofftware

IMS Tools 2012

Scenario: using the Eclipse-based plugin

13

• Shell-sharing allows integration with development tools such as Rational Developer for System z and the (free) IMS Explorer for Development

• Provides an improved experience for developers

• Includes commonly expected controls (with more coming)

Page 14: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

14

Console with status information

Status information

Control the server Manage basic aspects of the configuration

Basic features of client server development

Page 15: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

15

Console with status information Stop and start

components

Manage basic aspects of the configuration

Rational IDE extended with IMS Connect Extensions

Page 16: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

New features

Routing based on OTMA flood condition

Datastore drain: prevent new activity on a datastore while allowing existing activity to complete

Session rebalancing: enhance support for sysplex distributor when sessions use persistent sockets

16

CEX230 PM69452 - Datastore Route Drain Support - Technote CEX230 PM64619 - Session Message Limit - Technote IMS APAR PM69452 autoresume enablement.

Page 17: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

17

Scenario:

• Identify an OTMA flood condition

• Investigate the causes of the flood…

Page 18: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

18

Correct the problem. For example: • Restart the PSB • Increase the flood

threshold • Change capacity

weights to help reduce the overall load.

Page 19: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Scenario: Datastore needs to be taken offline

Problem: cannot take a datastore offline without potentially disrupting clients with active sessions

Mark the datastore as requiring a drain

Status changed to suspended: no new requests will be routed to this system

19

Page 20: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

20

Monitor the session counts until they reach an acceptable level

Use any criteria to find the sessions that matter

Page 21: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Restart the datastore

Click to stop the datastore

Perform maintenance

Click to start the datastore

21

Page 22: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

IMS

VSAM

DB2

Problem: where did the delay occur?

A single transaction can involve activity across many subsystems

Subsystem-specific tools offer a limited perspective

To quickly identify performance issues, you need to track activity across subsystems

Each subsystem has its own activity log

© 2012

22

CICS

Page 23: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Solution: Transaction Analysis Workbench for z/OS Combines logs from many

subsystems to provide a single, consolidated timeline of transaction activity.

Shows records that are related to the same transaction instance.

Automatically locates the log data sets for the problem time range (for some subsystems).

© 2012

CICS

DB2

IMS

VSAM

0.000000s

0.002000s

0.004000s

0.005000s

23

Page 24: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Currently supported subsystems and their logs

© 2012 IBM Corporation and Fundi Software 24

Work bench

CICS IMS

DB2 Web Sphere

z/OS IMS log IMS transaction index IMS monitor and DB monitor IMS Connect event data CQS log stream CMF performance (SMF)

WebSphere log extract Statistics (SMF) Accounting (SMF)

Selected SMF record types (in either log streams or data sets) OPERLOG (log stream)

DB2 log Accounting (SMF) Performance (SMF)

Page 25: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Combining CICS and IMS performance data

Report and extractJCL:

Report and extractJCL:

Report and extractJCL:

SMF file(CMF records)

IMS log

CICStransaction index

IMStransaction index

REPORT ...CICS-DBCTL

REPORT CICS-DBCTL ... ..COMBINED.

REPORT ...IMS-DBCTL

1. CMF records

3. Combined

2. IMS log records

CICS-DBCTLlist and summaryreport

CICS-DBCTLcombined summaryreport

IMS-DBCTLlist and summaryreport

This report combinesCMF and IMS data

25

Page 26: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Step 3: CICS-DBCTL combined summary report

© 2012 IBM C

26

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

Page 27: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: combines CICS and IMS data

© 2012 IBM C

27

ss

CICS

ss

IMS

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

Page 28: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: CICS data (section 1)

© 2012 IBM C

28

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

CICS

Data from CICS monitoring facility (CMF) performance class (SMF type 110) records. The report is summarized by CICS transaction ID and APPLID. You can specify whether the report shows the actual CICS transaction ID or an application (“umbrella”) transaction ID.

Page 29: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: CICS data

© 2012 IBM C

29

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

CICS

The number of CMF records summarized in this section.

Page 30: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: CICS data

© 2012 IBM C

30

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

CICS

Averages for the summarized transactions: response time, CPU time, number of DL/I requests issued, and IMS wait time. Comparing the response time to the IMS wait time helps you to identify whether delays occurred in CICS or in IMS. The IMS sections of the report (described later) show the types of IMS requests, and separate components of the IMS wait time, to help you identify the cause of long IMS wait times.

Page 31: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: CICS data

© 2012 IBM C

31

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

CICS

The number of transactions that abended. This example shows that, of the 60 transactions summarized here (BANK transactions that ran on APPLID CICSP1), 10 abended.

Page 32: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: CICS data

© 2012 IBM C

32

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

CICS

Number of transactions per second (for this transaction ID running on this APPLID). A value of 0 indicates a rate of less than 1 transaction per second.

Page 33: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: IMS data (sections 2-5)

33

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

IMS

Section 2: IMS thread elapsed and CPU times, and PSB schedule delays

Page 34: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: IMS data

34

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

IMS

Section 3: Full Function database DL/I call activity and delays

Page 35: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: IMS data

© 2012 IBM C

35

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

IMS

Section 4: Fast Path database DL/I call activity and delays

Page 36: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

CICS-DBCTL combined summary report: IMS data

© 2012 IBM C

36

2012-03-28 Wednesday CICS-DBCTL Summary Page 1 Tran APPLID CMF Count Response CPU Time IMS Reqs IMS Wait ABEND Rate/Sec ---- -------- --------- -------- -------- -------- -------- -------- -------- BANK CICSP1 60 11.12982 0.008967 35 4.256977 10 0 08 Count Elapsed CPU Time StaDelay Schedule IC Wait PS Wait --------- -------- -------- -------- -------- -------- -------- 42 10.94999 0.004092 0.011668 0.000183 0 0 07 Count DB call DB Gets DB Upds IO Count IO Time LockWait --------- -------- -------- -------- -------- -------- -------- 41 33 13 19 4 0.003438 3.980170 FP Count FP call FP Gets FP Upds FP Wait FP Fail --------- -------- -------- -------- -------- -------- 41 19 7 11 0 7 Synctime Phase 1 Phase 2 FP PH2 OTHREAD -------- -------- -------- -------- -------- 0.011938 0.006555 0.005383 0.002232 0.017659

ss

IMS

Section 5: Syncpoint elapsed time breakdown, including the async Fast Path OTHREAD overhead

Page 37: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Tracking a CICS DBCTL transaction in the log browser Select the following three files for browsing:

– CICS transaction index – IMS transaction index – IMS log

© 2012

37

File Menu Edit Help ────────────────────────────────────────────────────────────────────────────── Process Log Files Row 1 of 298 More: < > Command ===> Scroll ===> CSR Select a Log File to browse. IMS Release 111 + Zone +0800 / Log File Rel + Filter + Zone SS 'CICS.CMF.EXCEPT.EXTRACT' 6E13 'IMS.DBCTL.EXCEPT.EXTRACT' CA01 SS 'IMSA.SLDS' 08

Page 38: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Tracking a CICS DBCTL transaction in the log browser (continued)

© 2012 IBM C

38

File Mode Filter Time Labels Options Help ────────────────────────────────────────────────────────────────── BROWSE CICS.CMF.EXCEPT.EXTRACT + Record 00017856 More: < > Command ===> Scroll ===> CSR Slice . . Duration Date 2011-04-06 Time 08.41.45.732610 Code Description < 00.00.00.000000 > 2011-04-06 Wednesday Time (Relative) / ---- ------------------------------------------------------ ---------------- TX 6E13 CICS Transaction TranCode=BANK Task=203 08.51.16.302265 Program=BANKP1 Userid=JIM LTerm=VAPFUW2B Terminal=UW2B RecToken=CICSP1/C7945AC9E865D801 Resp=8.233590 CPU=0.005672 IMS=37 ACCT=FTS3.VAPFUW2B.945AC9E865D8 Task=203 PSB=BANKPSB1 ---------------------------------------------------------------------------- CA01 IMS Transaction TranCode=BANK Task=203 PSB=BANKPSB1 +0.001923 RecToken=CICSP1/C7945AC9E865D801 Region=0001 CPU=0.004973 Process=8.228251 RegTyp=DBC DBCalls=36 ---------------------------------------------------------------------------- 08 Application Start TranCode=BANK Program=DFHTWM04 +0.001923 5607 Start of UOR Program=BANKP1 +0.001924 50 Database Update Database=BANKDB21 +3.245731 50 Database Update Database=BANKDB22 +3.246030 50 Database Update Database=BANKDB23 +3.246065 50 Database Update Database=BANKDB24 +3.246233 50 Database Update Database=BANKDB25 +3.246281

Page 39: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Tracking a CICS DBCTL transaction in the log browser (continued)

© 2012 IBM C

39

50 Database Update Database=BANKDB26 +4.293974 50 Database Update Database=BANKDB27 +4.294279 50 Database Update Database=BANKDB28 +4.294315 50 Database Update Database=BANKDB29 +4.294481 50 Database Update Database=BANKDB31 +4.294527 50 Database Update Database=BANKDB32 +7.177249 50 Database Update Database=BANKDB33 +7.177533 50 Database Update Database=BANKDB34 +7.177581 50 Database Update Database=BANKDB35 +7.177792 50 Database Update Database=BANKDB36 +7.177841 5610 Syncpoint Start of Phase 1 +8.224356 5950 FP Database Update Database=BANKFP3 +8.224384 5950 FP Database Update Database=BANKFP4 +8.224387 5950 FP Database Update Database=BANKFP5 +8.224388 5950 FP Database Update Database=BANKFP6 +8.224391 3730 Syncpoint End of Phase 1 +8.228253 5937 FP Syncpoint Program=BANKPGM1 +8.229158 56FA Transaction Statistics +8.229168 07 Application Terminate +8.230174 5612 Syncpoint End of Phase 2 Program=BANKPGM1 +8.233899 ******************************** Bottom of Data *******************************

Page 40: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

BROWSE JCH1.FUW745.UPDATE.CICS.INDEX + Record 00000456 More: < > Command ===> Scroll ===> CSR Slice . . Duration Date 2013-05-31 Time . . 15.02.14.735246 Code Description < 00.04.45.000000 > 2013-05-31 Friday Time (Elapsed) / ———— —————————————————————————————————————————————————————— ———————————————— 6E13 CICS Transaction TranCode=FB66 Task=930 16.26.19.715404 053 SQL DESCRIBE/COMMIT/ROLL SQLCODE=0000 STMT=000158 DBA6 0.004614 380 SP Entry FBOSP006 DBA6 0.001755 015 Index scan begin DBA6 0.431828 018 Scan end DBA6 0.000255 053 SQL DESCRIBE/COMMIT/ROLL SQLCODE=0000 STMT=000281 DBA6 0.000127 060 SQL SELECT STMT=000344 DBA6 0.000181 017 Sequential scan begin DBA6 0.000126 018 Scan end DBA6 1.725756 058 SQL call completion SQLCODE=0000 STMT=000344 DBA6 0.000067 061 SQL UPDATE STMT=000423 DBA6 0.000358 017 Sequential scan begin DBA6 0.000132 018 Scan end DBA6 0.066364 058 SQL call completion SQLCODE=0000 STMT=000423 DBA6 0.000276 499 SP statement execution detail DBA6 0.000284 380 SP Exit FBOSP006 SQLCODE=0000 DBA6 0.000035 053 SQL DESCRIBE/COMMIT/ROLL SQLCODE=0000 STMT=000196 DBA6 0.000044 003 Thread accounting DBA6 0.002966 TranCode=FB66 Userid=TWM ClientID=FUWTCIC RESP=2.230828 CPU1=0.001276 CPU2=0.000866 I/O3=0.000359 ACCT=FTS3.SC0TCP06.70FA74C4CAA8 Source=CICS GtPgRq=14616 SyPgUp=1 Suspnd=0 DeadLk=0 TimOut=0 MxPgLk=2 Sel=1 Ins=0 Upd=1 Del=0 LUWID=FTS3/DBA6LU/CB70FA74C5AE/0002

Workbench coming soon: Improved DB2 diagnosis for CICS and IMS transactions

40

Page 41: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

********************************* Top of data ********************************* +00CC Code... 018 Scan end DBA6 +00D8 STCK... CB70FA7D5032F384 LSN.... 000000000001035D Date... 2013-05-31 Friday Time... 16.26.28.675375.219 +0048 QW0018ID... Scan details +0048 Scan type.... 'SEQD' Rows processed... +780753 +0054 Rows examined.... +780753 +005C Rows qualified after stage 1... +1 +0064 Rows qualified after stage 2... +1 +006C Rows inserted.... +0 +0074 Rows updated... +0 +007C Rows deleted... +0 +0084 Pages scanned.... +14610 +0088 Pages scanned (RI)... +0 +008C Rows deleted (RI).... +0 +0094 Rows skipped (incompatible lock)... +0 +00C8 QWHS....... Product section standard header +00C8 Standard header length... +86 Header type.... 01 +00CB RMID....... 0E IFCID...... +18 NSDA....... +3 +00CF DB2 release.... A1 ACE address.... 1A94B680 +00D4 DB2 subsystem.... 'DBA6' +00D8 Time stamp... CB70FA7D5032F384 +00E0 IFCID sequence... +3855809 +00E4 Destination sequence... 00B3FBFE +00E8 Active trace number mask... 00000004 +00EC DB2 location... 'DB2ALOC'

DB2 analysis continued: Long sequential data scan caused the problem

41

Page 42: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

BROWSE IBB1.SLDSP.IBB1.D10110.T1046282.V13 Record 00000385 More: < > Command ===> Scroll ===> CSR Slice . . Duration Date 2010-04-20 Time . . 02.46.28.354791 Code Description < 00.04.45.000000 > 2010-04-20 Tuesday Time (Elapsed) / ———— —————————————————————————————————————————————————————— ———————————————— 08 Application Start Program=DFHTWM04 Region=0002 03.01.08.462818 5050 Database DLET Database=DI21PART Region=0002 0.003205 5050 Database DLET Database=DI21PART Region=0002 0.000035 5052 Database insert into KSDS Database=DI21PART 0.000445 5050 Database ISRT Database=DI21PART Region=0002 0.000057 AA DLI Database call: GHU Region=0002 0.000000 AA DLI Database call: REPL Region=0002 0.000003 AA DLI Database call: GHU Region=0002 0.000000 . . . AA DLI Database call: ISRT Region=0002 0.000003 AA DLI Database call: GHU Region=0002 0.000003 5050 Database DLET Database=DI21PART Region=0002 0.034917 5050 Database DLET Database=DI21PART Region=0002 0.000043 5052 Database insert into KSDS Database=DI21PART 0.000569 5050 Database ISRT Database=DI21PART Region=0002 0.000055 5610 Syncpoint Start of Phase 1 Region=0002 3.201498 5950 FP Database Update Database=IVPDB3 Region=0002 0.000032 5950 FP Database Update Database=IVPDB3 Region=0002 0.000001 5950 FP Database Update Database=IVPDB3 Region=0002 0.000001 3730 Syncpoint End of Phase 1 Region=0002 0.003366 5937 FP Syncpoint Program=DFHTWM04 Region=0002 0.001752 56FA Transaction Statistics Region=0002 0.000009 07 Application Terminate Region=0002 0.001785 5612 Syncpoint End of Phase 2 Program=DFHTWM04 Region=0002 0.005594 ******************************** Bottom of Data *******************************

42

IMSPI: Incorporate the IMS trace into your Log analysis

Page 43: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Extract Request Command ===> Specify the required options then press enter to perform the extract. Extract data set . . 'JCH.FUW.P0000002.D130617.T152530.EXTRACT' Size attributes for a new data set: Space units . . . . . TRKS (BLKS, TRKS, CYLS, MB) Primary quantity . . . 1 (In above units) Secondary quantity . . 1 (In above units) Options to prevent a run-away extract: Maximum record count 10000 (Stop after this many records) Find limit . . . . . . 5000 (Override for this request only) Timeout . . . . . . . 0 (Override for this request only)

IMSPI: Easily create an extract during analysis

43

Extract the: 1. Entire transaction you are currently tracking 2. Selected or a range of records only Control with: 1. Size and limit options 2. Generated extract dataset names according to your standards

Page 44: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

IMSPA: Reporting CPU service units

Q: How do you compare transaction CPU utilization when upgrading to a new machine type?

A: CPU service units – the standard measure of CPU cycles

The following transaction list report shows CPU time and service units together.

Tran IMSID Response CPUtime CPUSU

BANK OLD1 0.629 1.880 28047

BANK NEW1 0.425 1.234 27505

While the CPU time for the same transaction running on two different systems shows a 52% improvement, the service units has standardized that difference to around 2%.

44

Page 45: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Future directions

Workbench proposal: a GUI for first responders. Unprecedented empowerment of frontline staff

Workbench proposal: DB2 scenarios and improved cross-product integration; better integration with other DB2 tools

IMS Connect Extensions proposal: Exploitation of dynamically added datastores: incorporation into routing (dynamically create new targets for routing without needing to stop ICON; add “as-needed” capacity)

45

Page 46: IMS11 IMS Performance Tools - IMS UG June 2013 Sydney

Summary: the vision for performance tools

Increasingly focused on not just mining the data but guiding users on how to interpret and collect this information

Paradigm changing from individual systems (silos) to federated applications

Increase reliance on GUI for analytics and response

Solving, not just identifying, problems

46