You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Siddharth Seth (JIRA)" <ji...@apache.org> on 2016/06/27 21:02:16 UTC
[jira] [Updated] (HIVE-14102) LLAP Improvements to per executor
memory computation in daemons
[ https://issues.apache.org/jira/browse/HIVE-14102?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Siddharth Seth updated HIVE-14102:
----------------------------------
Labels: llap (was: )
> LLAP Improvements to per executor memory computation in daemons
> ---------------------------------------------------------------
>
> Key: HIVE-14102
> URL: https://issues.apache.org/jira/browse/HIVE-14102
> Project: Hive
> Issue Type: Improvement
> Reporter: Siddharth Seth
> Labels: llap
>
> Currently set to 80% of maxAvailableMemory - irrespective of container size. This can end up with a lot of wastage.
> Also - the 80% / overhead may already be accounted for in YARN container sizing calculations. This should not be double counted.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)