[
https://issues.apache.org/jira/browse/HIVE-10901?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15616435#comment-15616435
]
Ashutosh Chauhan commented on HIVE-10901:
-----------------------------------------
We can use old method implemented in AggregateExpandDistinctAggregatesRule
which does this via computing distinct count on each branch and then doing a
join. Likely grouping set approach may be more efficient but join approach may
be an improvement on state of art in certain cases.
> Optimize mutli column distinct queries
> ----------------------------------------
>
> Key: HIVE-10901
> URL: https://issues.apache.org/jira/browse/HIVE-10901
> Project: Hive
> Issue Type: New Feature
> Components: CBO, Logical Optimizer
> Affects Versions: 1.2.0
> Reporter: Mostafa Mokhtar
> Assignee: Ashutosh Chauhan
> Attachments: HIVE-10901.patch
>
>
> HIVE-10568 is useful only when there is a distinct on one column. It can be
> expanded for multiple column cases too.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
