[
https://issues.apache.org/jira/browse/YARN-4512?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15097544#comment-15097544
]
Wangda Tan commented on YARN-4512:
----------------------------------
Thanks for explanations, [~kasha]/[~elgoiri].
Make sense to me, looking forward your doc, and I suggest to add more details
to your new added protos.
One question to [~elgoiri] about your comment:
bq. After thinking about it, I think it makes more sense to send this
overallocaiton parameter in the registration instead of the heartbeat. If we
want to dynamically change the overallocaiton we may want to do it as the
ResourceOption does.
I'm not very familiar with ResourceOption, is it only used by RM? I think it is
possible that NM will use its own data to decide how many resources could be
over-allocated.
> 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)
