[
https://issues.apache.org/jira/browse/YARN-4512?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15115701#comment-15115701
]
Karthik Kambatla commented on YARN-4512:
----------------------------------------
Sorry for the delay here, I was out of the country for the past week.
Thanks Inigo for pointing out ResourceOption, I was not aware of it. Just
looked through how it works. Using the rmadmin CLI to update a node's configs
(through the RM) is not very intuitive to me. Also, an NM restart overwrites
the values with those in the configuration. Given that work was done before
work-preserving NM restart was implemented, I feel we might be better off
making the configuration change and restarting the node.
In any case, unless folks are very particular on this, I would like for us to
commit the patch posted here and revisit the configs in another JIRA before the
merge. Glad to file that JIRA if you guys think we can get this in.
> Provide a knob to turn on over-allocation
> -----------------------------------------
>
> Key: YARN-4512
> URL: https://issues.apache.org/jira/browse/YARN-4512
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager
> Reporter: Karthik Kambatla
> Assignee: Karthik Kambatla
> Attachments: YARN-4512-YARN-1011.001.patch,
> yarn-4512-yarn-1011.002.patch, yarn-4512-yarn-1011.003.patch
>
>
> We need two configs for overallocation - one to specify the threshold upto
> which it is okay to over-allocate, another to specify the threshold after
> which OPPORTUNISTIC containers should be preempted.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
