You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@hive.apache.org by "Ashutosh Chauhan (JIRA)" <ji...@apache.org> on 2013/06/30 10:02:21 UTC

[jira] [Commented] (HIVE-3629) Try bucket split more localized by referencing block locations of small aliases

    [ https://issues.apache.org/jira/browse/HIVE-3629?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13696299#comment-13696299 ] 

Ashutosh Chauhan commented on HIVE-3629:
----------------------------------------

[~navis] Did you get a chance to benchmark this? How much speedup you are getting because of this?
                
> Try bucket split more localized by referencing block locations of small aliases
> -------------------------------------------------------------------------------
>
>                 Key: HIVE-3629
>                 URL: https://issues.apache.org/jira/browse/HIVE-3629
>             Project: Hive
>          Issue Type: Improvement
>          Components: Query Processor
>            Reporter: Navis
>            Assignee: Navis
>            Priority: Minor
>         Attachments: HIVE-3629.D6273.1.patch, HIVE-3629.D6273.2.patch
>
>
> Currently splits for SMBMapJoin does not concern about location of small aliases. If all inputs could be on single server, joining would be more faster.

--
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: http://www.atlassian.com/software/jira