[
https://issues.apache.org/jira/browse/HIVE-9561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14308560#comment-14308560
]
Xuefu Zhang commented on HIVE-9561:
-----------------------------------
Yeah, hasSortBy and hasOrderBy are for the whole query. Thus, they are not
reliable for either detection. I think we can add a flag in ReduceSinkDesc,
similar to enforceSort, telling about sortBy/orderBy/None.
Yes, I think that optimization makes sense. Please do some prototyping. If all
is well, create an JIRA for it.
> SHUFFLE_SORT should only be used for order by query [Spark Branch]
> ------------------------------------------------------------------
>
> Key: HIVE-9561
> URL: https://issues.apache.org/jira/browse/HIVE-9561
> Project: Hive
> Issue Type: Sub-task
> Components: Spark
> Reporter: Rui Li
> Assignee: Rui Li
> Attachments: HIVE-9561.1-spark.patch
>
>
> The {{sortByKey}} shuffle launches probe jobs. Such jobs can hurt performance
> and are difficult to control. So we should limit the use of {{sortByKey}} to
> order by query only.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
