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

Posted to user@hive.apache.org by Ryan LeCompte <le...@gmail.com> on 2009/12/10 22:44:46 UTC

Hive not using the full mapper capacity for certain jobs

Hello all,

The cluster has a capacity of 28 concurrent mappers. It looks like on larger
tables Hive will use all 28 mappers, but on a smaller table it only uses 4
mappers maximum. Any idea how to override this?

Thanks,
Ryan

RE: Hive not using the full mapper capacity for certain jobs

Posted by Zheng Shao <zs...@facebook.com>.

Try this:
set mapred.map.tasks=28;

Zheng
From: Ryan LeCompte [mailto:lecompte@gmail.com]
Sent: Thursday, December 10, 2009 1:45 PM
To: hive-user@hadoop.apache.org
Subject: Hive not using the full mapper capacity for certain jobs

Hello all,

The cluster has a capacity of 28 concurrent mappers. It looks like on larger tables Hive will use all 28 mappers, but on a smaller table it only uses 4 mappers maximum. Any idea how to override this?

Thanks,
Ryan