[jira] [Commented] (MAPREDUCE-5785) Derive task attempt JVM max heap size and io.sort.mb automatically from mapreduce.*.memory.mb

["Karthik Kambatla (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAPREDUCE-5785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14215266#comment-14215266 ] 

Karthik Kambatla commented on MAPREDUCE-5785:
---------------------------------------------

Sorry for the dropping the ball on this. Few comments:
# For those moving from MR1 to MR2, specifying Xmx might be easier. In those cases, it would be nice to estimate the memory.mb values. I am open to doing this in another JIRA, and would be happy to work on it and post a patch.
# Multiplying container-size by 0.8 to arrive at the heapsize does read better. We will have to divide by 0.8 to do it in the other direction. 
# In the patch itself, deprecating mapred.child.java.opts is fine but I would refer to .map.java.opts and .reduce.java.opts instead of the container size values. Those properties instead could point to the memory.mb properties. 

I understand it has been a while since we last looked at this. [~jira.shegalov] - will you be able to address these comments? I can help with addressing the comments if you are otherwise caught up. 

> Derive task attempt JVM max heap size and io.sort.mb automatically from mapreduce.*.memory.mb
> ---------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-5785
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5785
>             Project: Hadoop Map/Reduce
>          Issue Type: New Feature
>          Components: mr-am, task
>            Reporter: Gera Shegalov
>            Assignee: Gera Shegalov
>         Attachments: MAPREDUCE-5785.v01.patch, MAPREDUCE-5785.v02.patch, MAPREDUCE-5785.v03.patch
>
>
> Currently users have to set 2 memory-related configs per Job / per task type.  One first chooses some container size map reduce.\*.memory.mb and then a corresponding maximum Java heap size Xmx < map reduce.\*.memory.mb. This makes sure that the JVM's C-heap (native memory + Java heap) does not exceed this mapreduce.*.memory.mb. If one forgets to tune Xmx, MR-AM might be 
> - allocating big containers whereas the JVM will only use the default -Xmx200m.
> - allocating small containers that will OOM because Xmx is too high.
> With this JIRA, we propose to set Xmx automatically based on an empirical ratio that can be adjusted. Xmx is not changed automatically if provided by the user.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)