[
https://issues.apache.org/jira/browse/HIVE-20199?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16548091#comment-16548091
]
Vihang Karajgaonkar commented on HIVE-20199:
--------------------------------------------
Thanks for reporting the issue and providing a patch. Can you provide a patch
which works for all the different databases supported? In order to provide a
patch you will need to provide a .sql following the naming convention similar
to *.sql files in metastore/scripts/upgrade/ directory. Also, you will have to
modify the hive-schema-4.0.0.sql file to add the index above.
> Improved filtering performance for a large number of partitions in a single
> table.
> ----------------------------------------------------------------------------------
>
> Key: HIVE-20199
> URL: https://issues.apache.org/jira/browse/HIVE-20199
> Project: Hive
> Issue Type: Improvement
> Components: Metastore
> Affects Versions: 1.2.1
> Reporter: Biao Wu
> Assignee: Biao Wu
> Priority: Major
> Attachments: 021-HIVE-20199.mysql.sql
>
>
> eg:
> {code:sql}
> select * from test where dt = '20180606'
> {code}
> The filter 'dt=20180606' will be pushed down to mysql for execution, but the
> test table contains a large number of partitions, this is a poor performance.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
