[
https://issues.apache.org/jira/browse/BEAM-1815?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15945198#comment-15945198
]
ASF GitHub Bot commented on BEAM-1815:
--------------------------------------
Github user asfgit closed the pull request at:
https://github.com/apache/beam/pull/2334
> Avoid shuffling twice in GABW
> -----------------------------
>
> Key: BEAM-1815
> URL: https://issues.apache.org/jira/browse/BEAM-1815
> Project: Beam
> Issue Type: Bug
> Components: runner-spark
> Reporter: Amit Sela
> Assignee: Amit Sela
>
> Spark runner implementation of GABW includes a "built-in" groupByKey, but
> BOBK before it already groups, so in order to avoid an unnecessary shuffle we
> need to force a {{Partitioner}} on the RDDs involved.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
