hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Phabricator (JIRA)" <>
Subject [jira] [Commented] (HIVE-3536) Output of sort merge join is no longer bucketed
Date Fri, 05 Oct 2012 04:07:47 GMT


Phabricator commented on HIVE-3536:

njain has commented on the revision "HIVE-3536 [jira] Output of sort merge join is no longer

  Otherwise, it looks good

  ql/src/test/queries/clientpositive/smb_mapjoin_11.q:32 Unfortunately, this does not verify
that the data is bucketed.
  Can you perform a join between test_table3 and test_table1 for bucket 2 for both of them.
  That would return 0 rows if the data was not bucketed correctly.


To: JIRA, njain, kevinwilfong

> Output of sort merge join is no longer bucketed
> -----------------------------------------------
>                 Key: HIVE-3536
>                 URL:
>             Project: Hive
>          Issue Type: Bug
>          Components: Query Processor
>    Affects Versions: 0.10.0
>            Reporter: Kevin Wilfong
>            Assignee: Kevin Wilfong
>         Attachments: HIVE-3536.1.patch.txt
> I don't know if this was a feature or a happy coincidence, but before HIVE-3230, the
output of a sort merge join on two partitions would be bucketed, even if hive.enforce.bucketing
was set to false.  This could potentially save a reduce phase when inserting into a bucketed
> This would be good to have back.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message