hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sergey Shelukhin (JIRA)" <>
Subject [jira] [Updated] (HIVE-6430) MapJoin hash table has large memory overhead
Date Sat, 15 Mar 2014 01:17:43 GMT


Sergey Shelukhin updated HIVE-6430:

    Attachment: HIVE-6430.02.patch

Addressed major review and discussion feedback. I kept the list bit in the ref though, because
putting it in the array results in huge pita w/retrieval of the union. Removed the "split"
long, now everything is in one place.
Probably need to write some unit tests, q files do not cover all cases. Will do so later today
or maybe sunday

> MapJoin hash table has large memory overhead
> --------------------------------------------
>                 Key: HIVE-6430
>                 URL:
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Sergey Shelukhin
>            Assignee: Sergey Shelukhin
>         Attachments: HIVE-6430.01.patch, HIVE-6430.02.patch, HIVE-6430.patch
> Right now, in some queries, I see that storing e.g. 4 ints (2 for key and 2 for row)
can take several hundred bytes, which is ridiculous. I am reducing the size of MJKey and MJRowContainer
in other jiras, but in general we don't need to have java hash table there.  We can either
use primitive-friendly hashtable like the one from HPPC (Apache-licenced), or some variation,
to map primitive keys to single row storage structure without an object per row (similar to

This message was sent by Atlassian JIRA

View raw message