You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@beam.apache.org by "Etienne Chauchot (JIRA)" <ji...@apache.org> on 2019/04/30 10:09:00 UTC

[jira] [Commented] (BEAM-7184) Performance regression on spark-runner

    [ https://issues.apache.org/jira/browse/BEAM-7184?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16830151#comment-16830151 ] 

Etienne Chauchot commented on BEAM-7184:
----------------------------------------

Performance has improved lately thanks to the use of new AccumulatorsV2 API in this PR https://github.com/apache/beam/pull/8387 (only commit in master on the non-portable runner since performance regression) but the underlying source of the initial performance regression is not found yet.

> Performance regression on spark-runner
> --------------------------------------
>
>                 Key: BEAM-7184
>                 URL: https://issues.apache.org/jira/browse/BEAM-7184
>             Project: Beam
>          Issue Type: Bug
>          Components: runner-spark
>            Reporter: Etienne Chauchot
>            Priority: Blocker
>
> There is a performance degradation of+200% in spark runner starting on 04/10 for all the Nexmark queries. See https://apache-beam-testing.appspot.com/explore?dashboard=5138380291571712



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)