[
https://issues.apache.org/jira/browse/SPARK-37267?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17441578#comment-17441578
]
Apache Spark commented on SPARK-37267:
--------------------------------------
User 'ulysses-you' has created a pull request for this issue:
https://github.com/apache/spark/pull/34542
> OptimizeSkewInRebalancePartitions support optimize non-root node
> ----------------------------------------------------------------
>
> Key: SPARK-37267
> URL: https://issues.apache.org/jira/browse/SPARK-37267
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 3.3.0
> Reporter: XiDuo You
> Priority: Major
>
> `OptimizeSkewInRebalancePartitions` now is applied if the
> `RebalancePartitions` is the root node, but sometimes, we expect a local sort
> after do RebalancePartitions that can improve the compression ratio.
> After SPARK-36184, we make validate easy that whether the rule introduces
> extra shuffle or not and the output partitioning is ensured by
> `AQEShuffleReadExec.outputPartitioning`.
> So it is safe to make `OptimizeSkewInRebalancePartitions` support optimize
> non-root node.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
