[
https://issues.apache.org/jira/browse/YARN-5900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15673838#comment-15673838
]
Jason Lowe commented on YARN-5900:
----------------------------------
I don't believe this would simplify anything, rather just make it more
complicated within the scheduler and more complicated to explain to users
what's going on. In addition it will add to internal fragmentation of
container allocations, since we will force container allocations that are
smaller than the normal minimum to a larger value even though the task within
the container will not use it, reducing the overall utilization of the cluster.
> Configuring minimum-allocation-mb at queue level
> ------------------------------------------------
>
> Key: YARN-5900
> URL: https://issues.apache.org/jira/browse/YARN-5900
> Project: Hadoop YARN
> Issue Type: Improvement
> Reporter: Manikandan R
>
> Motivation for proposing minimum-allocation-mb at queue level in the form of
> yarn.scheduler.capacity.<queue-path>.minimum-allocation-mb
> is, when the queue structure has been designed on resource usages (only
> memory for now). For example, there could be three segments like small,
> medium & large jobs and queues can be created for each segment accordingly.
> With this, it would be good to configure the min container size of each queue
> separately. For example, For small, it is 1 GB, For medium, it is 3 GB, For
> large, it can be 6 GB.
> Would this simplify container release process and its overall management,
> eventually reduces no. of containers running at any moment?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
