[infinispan-issues] [JBoss JIRA] (ISPN-3534) Investigate performance regressions in Infinispan 6.0.0
Michal Linhard (JIRA)
jira-events at lists.jboss.org
Fri Nov 15 09:11:07 EST 2013
[ https://issues.jboss.org/browse/ISPN-3534?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12924010#comment-12924010 ]
Michal Linhard commented on ISPN-3534:
--------------------------------------
The max_credits didn't seem to help they just smoothed the client stress test curve a bit, but the maximum is still quite far below 6.1.0.GA
http://jawa36.mw.lab.eng.brq.redhat.com:8080/repo/reports/compare/exec?q=640|931|939|946
I realized I haven't done proper comparison in radargun tests yet. I wanna see how we're standing there.
Changing of the bundler in default jgroups config improved the results: https://url.corp.redhat.com/fccbc87
I wanna see how we're standing in comparison to 6.1.0.GA in those same tests
> Investigate performance regressions in Infinispan 6.0.0
> -------------------------------------------------------
>
> Key: ISPN-3534
> URL: https://issues.jboss.org/browse/ISPN-3534
> Project: Infinispan
> Issue Type: Enhancement
> Affects Versions: 5.3.0.Final, 6.0.0.CR1
> Reporter: Mircea Markus
> Assignee: Dan Berindei
> Priority: Critical
> Labels: 620
> Fix For: 6.0.0.Final
>
>
> As per QE:
> 30% performance regression in distributed, udp, tx, async library stress tests
> 20% performance regression in distributed, udp, tx, sync library stress tests
> 20% performance regression in memcached client stress tests
> 50%+ performance regression in REST client stress
> tests
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
More information about the infinispan-issues
mailing list