[ 
https://issues.apache.org/jira/browse/HIVE-865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12761667#action_12761667
 ] 

Ning Zhang commented on HIVE-865:
---------------------------------

The mem leak turns out to be caused by the MRU cache manager in JDBM. Currently 
Hive caches 25000 rows in JDBM through MapJoin by default. I changed it to 1 
and it seems works fine. I'll looking into MRU.

> mapjoin: memory leak for same key with very large number of values
> ------------------------------------------------------------------
>
>                 Key: HIVE-865
>                 URL: https://issues.apache.org/jira/browse/HIVE-865
>             Project: Hadoop Hive
>          Issue Type: Bug
>          Components: Query Processor
>            Reporter: Namit Jain
>            Assignee: Ning Zhang
>


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to