Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Date: Tue, 28 Aug 2012 03:45:11 +1100 (NCT)
From: "Namit Jain (JIRA)" <jira@apache.org>
To: hive-dev@hadoop.apache.org
Message-ID: <1939626739.1925.1346085911138.JavaMail.jiratomcat@arcas>
In-Reply-To: <1590697659.40159.1340302243241.JavaMail.jiratomcat@issues-vm>
Subject: [jira] [Commented] (HIVE-3171) Bucketed sort merge join doesn't
 work when multiple files exist for small alias
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HIVE-3171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13442517#comment-13442517 ] 

Namit Jain commented on HIVE-3171:
----------------------------------

I know the policy, but in most of the cases, I think we don't commit our own patches.
Anyway, it is not a big deal - I don't feel very comfortable about it, but have no
reservations if you want to take that path.
                
> Bucketed sort merge join doesn't work when multiple files exist for small alias
> -------------------------------------------------------------------------------
>
>                 Key: HIVE-3171
>                 URL: https://issues.apache.org/jira/browse/HIVE-3171
>             Project: Hive
>          Issue Type: Bug
>          Components: Query Processor
>    Affects Versions: 0.10.0
>            Reporter: Joey Echeverria
>            Assignee: Navis
>              Labels: bucketing, joins, partitioning
>         Attachments: HIVE-3171.1.patch.txt
>
>
> Executing a query with the MAPJOIN hint and the bucketed sort merge join optimizations enabled:
> {noformat}
> set hive.input.format=org.apache.hadoop.hive.ql.io.BucketizedHiveInputFormat;
> set hive.optimize.bucketmapjoin = true;
> set hive.optimize.bucketmapjoin.sortedmerge = true;
> {noformat}
> works fine with partitioned tables if there is only one partition in the table. However, if you add a second partition, Hive attempts to do a regular map-side join which can fail because the tables are too large. Hive ought to be able to still do the bucketed sort merge join with partitions.

--
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