Edit
Attach
P
rintable
r26 - 2014-04-02 - 15:10:38 - Main.gcovell
You are here:
TWiki
>
Deployment Web
>
DeploymentPlanningAndDesign
>
PerformanceDatasheetsAndSizingGuidelines
>
RationalTeamConcertOnZOSPerformanceImpactAfterJavaJFSAPIAdoptionIn405
<div id="header-title" style="padding: 10px 15px; border-width:1px; border-style:solid; border-color:#FFD28C; background-image: url(<nop>https://jazz.net/wiki/pub/Deployment/WebPreferences/TLASE.jpg); background-size: cover; font-size:120%"> ---+!! <img src="https://jazz.net/wiki/pub/Deployment/WebPreferences/new.png" alt="new.png" width="50" height="50" align="right"> Rational Team Concert for z/OS performance impact of Java JFS API adoption in 4.0.5 %DKGRAY% Authors: Main.SuHui <br> Last updated: December 6, 2013 <br> Build basis: Rational Team Concert on ZOS 4.0.5 & 4.0.4 %ENDCOLOR%</div></sticky> <!-- Page contents top of page on right hand side in box --> <sticky><div style="float:right; border-width:1px; border-style:solid; border-color:#DFDFDF; background-color:#F6F6F6; margin:0 0 15px 15px; padding: 0 15px 0 15px;"> %TOC{title="Page contents"}% </div></sticky> <sticky><div style="margin:15px;"></sticky> ---++ Introduction RTC for z/OS has recently introduced a Java API to access JFS (Jazz Foundation Services) instead of using the HTTP APIs which provides the potential for significant performance improvements. This report compares the performance before and after RTC for z/OS adopted the Java JFS API in part of the resource operations and queries. Comparison is made between the 4.0.5 RC1 development version and the previous 4.0.4 release. The test objective is to verify the performance improvement for the new Java JFS API adoption. ---++ Disclaimer %INCLUDE{"PerformanceDatasheetDisclaimer"}% ---++ Findings Based on the test data performance of the RTC for z/OS has improved significantly from 4.0.5 to 4.0.4 on some user scenarios: * The *promotion time improved about 20%* (50% of "Generating list of binaries to promote" time and 15% of "Finalize build maps") * A build property (team.enterprise.build.dependency.forceFullAnalysis) was added during team build to reflect the impact of the changed queries, the *preprocessing build time shows an improvement of about 60%* For other scenarios performance stayed consistent between the two versions. ---++ Topology The RTC server was set up based on !WebSphere and DB2 on Linux for System z. The build machine with Rational Build Agent was on zOS. <table class="gray-table"> <tr> <th><strong>Test Environment</strong></th> <th></th> <th></th> </tr> <tr> <td><strong>RTC Server</strong></td> <td>Operating System & Version: Linux for System z (SUSE Linux Enterprise Server 10 (s390x))</br> System Resource : 10 GB Storage, 4 CPs (20000 mips, CPU type : 2097.710, CPU model: E12) </br> CLM: 4.0.5 RC1 (CALM-I20131009-0549), 4.0.4 GA (CALM-I20130806-2343), 4 GB heap size </br> DB2: 9.7.0.5 </br> WAS: 8.0.0.3 </br> </td> <tr> <td><strong>Build Forge Agent</strong></td> <td>Operating System & Version: z/OS 01.12.00</br> System Resource: 6 GB Storage, 4 CPs (20000 mips, CPU type : 2097.710, CPU model: E12)</br> Build System Toolkit: 4.0.5 RC1 (RTC-I20131009-0158), 4.0.4 GA (RTC-I20130803-2156) </td> </tr> </table> The CLM deployment topology is derived from and is part of the standard E5 topology documented at [[AlternativeCLMDeploymentTopologies#CLM_E5_Enterprise_zOS_Linux_on_S][(CLM-E5) Enterprise - zOS / Linux on System z / DB2]]. Unlike the E5 topology, DB2 is setup on the same LPAR as IBM !WebSphere Application Server. <img src="%ATTACHURLPATH%/Enterprise_zOS_Linux_on_z_DB2_topology.png" alt="Enterprise_zOS_Linux_on_z_DB2_topology.png"/> ---++ Methodology Build time and individual activity time were compared by getting test start date and time. For CPU and memory utilization monitoring tools NMON was used for RTC server and RMF on ZOS was used for Rational Build Agent. The sample project for the test was Mortgage*100 which is 100 duplicates of the [[https://jazz.net/wiki/bin/view/Main/ZOSBuildSamplesV4][Mortgage sample application]] with 800 compiles. In the repository the source code is stored in one stream with one single component which includes 4 zComponent Projects. Each test scenario is executed twice against the two versions. ---+++Test Scenarios <table class="gray-table"> <tbody> <tr> <th align="center">Test Scenario</th> <th align="center">Description</th> </tr> <tr> <td>RTC EE features</td> <td>1. Request full dependency build</br>2. Request component-based promotion, after full dependency build in step 1 with 800 files</td> </tr> <tr> <td>Incremental Build</td> <td>1. Change a copybook (!MortgageApplication-Common\zOSsrc\COPYBOOK\A00MTCOM.cpy) and request build</br>2. Change a COBOL file (!MortgageApplication-EPSCMORT\zOSsrc\COBOL\A00CMORT.cbl) and request build</br>3. Change all copybooks and request build</td> </tr> <tr> <td>Enterprise build with build property turned on to force full analysis</td> <td>Run enterprise build with build property team.enterprise.build.dependency.forceFullAnalysis=true</td> </tr> <tr> <td>Concurrent build</td> <td>Change a COBOL file and run 5 personal builds concurrently</td> </tr> </tbody> </table> ---++ Results ---+++ Run duration These data tables show the run duration comparing between 4.0.4 and 4.0.5 with the test start and completion time. The build time with full analysis enabled, promotion time, and the test build time after the promotion show the significant performance improvements. The promotion time was reduced about 20%. Test build time after promotion was reduced by 35%. Build with team.enterprise.build.dependency.forceFullAnalysis property time reduced about 60%. <img src="%ATTACHURLPATH%/build_time.png" alt="build_time.png" width="80%" height="80%" /> <img src="%ATTACHURLPATH%/promotion_time.png" alt="promotion_time.png" width="80%" height="80%" /> <img src="%ATTACHURLPATH%/incremental_time.png" alt="incremental_time.png" width="80%" height="80%" /> <img src="%ATTACHURLPATH%/concurrent_time.png" alt="concurrent_time.png" width="80%" height="80%" /> ---+++ Build Activities These data tables display the detailed build activity run time, improvements are contributed by "Collecting buildable files" in the build. "Generating list of binaries to promote" and "Finalize build maps" Activities are the major improvements for the promotion time. The improvements reflected the changing build map service in RTC for z/OS which adopted the Java JFS API. <img src="%ATTACHURLPATH%/activity_build.png" alt="activity_build.png" width="70%" height="70%" /> <img src="%ATTACHURLPATH%/activity_build_fullanalysis.png" alt="activity_build_fullanalysis.png" width="70%" height="70%" /> <img src="%ATTACHURLPATH%/activity_promotion.png" alt="activity_promotion.png" width="70%" height="70%" /> <img src="%ATTACHURLPATH%/activity_incrementalbuild.png" alt="activity_incrementalbuild.png" width="80%" height="80%" /> <img src="%ATTACHURLPATH%/activity_concurrent.png" alt="activity_concurrent.png" width="80%" height="80%" /> ---+++ CPU and Memory for RTC server This graph shows CPU and memory utilization for RTC server, data is collected by NMON tool. 4.0.4 and 4.0.5 show similar CPU and memory utilization. <img src="%ATTACHURLPATH%/404_server.png" alt="404_server.png" width="70%" height="70%" /> <img src="%ATTACHURLPATH%/405_server.png" alt="405_server.png" width="70%" height="70%" /> ---+++ CPU and Memory for Build Agent This graph shows CPU and memory utilization for Build Forge Agent on ZOS machine, data is collected by RMF tool. 4.0.4 and 4.0.5 show similar CPU and memory utilization. <img src="%ATTACHURLPATH%/404_BFA.png" alt="404_BFA.png" width="70%" height="70%" /> <img src="%ATTACHURLPATH%/405_BFA.png" alt="405_BFA.png" width="70%" height="70%" /> ---++ Appendix A - Key Tuning Parameters #AppendixA <table class="gray-table"> <tbody> <tr> <th align="left" width="200"><strong>Product</strong><br></th> <th align="left" width="100"><strong>Version</strong></th> <th align="left" width="600"><strong>Highlights for configurations under test</strong></th> </tr> <tr> <td style="vertical-align: top;">IBM !WebSphere Application Server</td> <td style="vertical-align: top;">8.0.0.3 </td> <td style="vertical-align: top;"><strong>JVM settings:</strong> * GC policy and arguments, max and init heap sizes: <verbatim> -Xmn512m -Xgcpolicy:gencon -Xcompressedrefs -Xgc:preferredHeapBase=0x100000000 -Xmx4g -Xms4g</verbatim> <strong>OS configuration:</strong> <verbatim> * hard nofile 120000 * soft nofile 120000</verbatim> Refer to http://pic.dhe.ibm.com/infocenter/clmhelp/v4r0m4/topic/com.ibm.jazz.install.doc/topics/c_special_considerations_linux.html for details </td> </tr> <tr> <td>DB2</td> <td>DB2 Enterprise Server 9.7.0.5</td> <td>Tablespace is stored on the same machine as IBM !WebSphere Application Server</td> </tr> <tr> <td>License Server</td> <td>Same as CLM version</td> <td>Hosted locally by JTS server</td> </tr> <tr> <td>Network</td> <td> </td> <td>Shared subnet within test lab</td> </tr> </tbody> </table> ---++ Related Links: * [[Rtceeperftestresultrootpage][RTC EE performance datasheets and sizing reports]] ---++++!! About the authors Main.SuHui -------------------- ---+++++!! Questions and comments: * What other performance information would you like to see here? * Do you have performance scenarios to share? * Do you have scenarios that are not addressed in documentation? * Where are you having problems in performance? %COMMENT{type="below" target="PerformanceDatasheetReaderComments" button="Submit"}% %INCLUDE{"PerformanceDatasheetReaderComments"}% <sticky></div></sticky>
Edit
|
Attach
|
P
rintable
|
V
iew topic
|
Backlinks:
We
b
,
A
l
l Webs
|
H
istory
:
r27
<
r26
<
r25
<
r24
<
r23
|
More topic actions...
Deployment
Deployment web
Planning and design
Installing and upgrading
Migrating and evolving
Integrating
Administering
Monitoring
Troubleshooting
Community information and contribution guidelines
Create new topic
Topic list
Search
Advanced search
Notify
RSS
Atom
Changes
Statistics
Web preferences
NOTE: Please use the Sandbox web for testing
Status icon key:
To do
Under construction
New
Updated
Constant change
None - stable page
Smaller versions of status icons for inline text:
Copyright © by IBM and non-IBM contributing authors. All material on this collaboration platform is the property of the contributing authors.
Contributions are governed by our
Terms of Use.
Please read the following
disclaimer
.
Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more
here
.