[
https://issues.apache.org/jira/browse/HIVE-13293?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15238556#comment-15238556
]
Rui Li commented on HIVE-13293:
-------------------------------
Thanks [~xuefuz] for the review.
I mean it can work with queries that have only one ShuffleMapStage. It will
definitely work with queries that have multiple ShuffleMapStage too. But as I
said in previous comment, what we care about here is just the last
ShuffleMapStage because that's what gets re-computed in parallel order by.
On the other hand, splitting task that has only one ShuffleMapStage seems weird
and may be bad for performance. That's why I chose to cache the RDD.
> Query occurs performance degradation after enabling parallel order by for
> Hive on Spark
> ---------------------------------------------------------------------------------------
>
> Key: HIVE-13293
> URL: https://issues.apache.org/jira/browse/HIVE-13293
> Project: Hive
> Issue Type: Bug
> Components: Spark
> Affects Versions: 2.0.0
> Reporter: Lifeng Wang
> Assignee: Rui Li
> Attachments: HIVE-13293.1.patch
>
>
> I use TPCx-BB to do some performance test on Hive on Spark engine. And found
> query 10 has performance degradation when enabling parallel order by.
> It seems that sampling cost much time before running the real query.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
