[
https://issues.apache.org/jira/browse/YARN-5356?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15637606#comment-15637606
]
Jason Lowe commented on YARN-5356:
----------------------------------
Thanks for updating the patch!
I don't think getNMResourceCalculatorPlugin does quite the right thing for this
case. Note that it tries to get the configured container monitor resource
calculator with a fallback to the node monitor resource calculator. The
resource monitor only uses the node monitor resource calculator, not the
container monitor resource calculator. So theoretically we could be doing the
wrong thing if the container monitor calculator did something different than
the node monitor calculator. We need to use the same calculator that the
NodeResourceMonitorImpl does, otherwise we can still have the mismatched
calculators issue.
We could add two utility methods for creating the calculator, one for the node
monitor and one for the container monitor, and make sure we call the same one
that NodeResourceMonitorImpl does.
> NodeManager should communicate physical resource capability to ResourceManager
> ------------------------------------------------------------------------------
>
> Key: YARN-5356
> URL: https://issues.apache.org/jira/browse/YARN-5356
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: nodemanager, resourcemanager
> Affects Versions: 3.0.0-alpha1
> Reporter: Nathan Roberts
> Assignee: Inigo Goiri
> Labels: oct16-medium
> Attachments: YARN-5356.000.patch, YARN-5356.001.patch,
> YARN-5356.002.patch, YARN-5356.002.patch, YARN-5356.003.patch,
> YARN-5356.004.patch, YARN-5356.005.patch, YARN-5356.006.patch,
> YARN-5356.007.patch, YARN-5356.008.patch, YARN-5356.009.patch,
> YARN-5356.010.patch
>
>
> Currently ResourceUtilization contains absolute quantities of resource used
> (e.g. 4096MB memory used). It would be good if the NM also communicated the
> actual physical resource capabilities of the node so that the RM can use this
> data to schedule more effectively (overcommit, etc)
> Currently the only available information is the Resource the node registered
> with (or later updated using updateNodeResource). However, these aren't
> really sufficient to get a good view of how utilized a resource is. For
> example, if a node reports 400% CPU utilization, does that mean it's
> completely full, or barely utilized? Today there is no reliable way to figure
> this out.
> [~elgoiri] - Lots of good work is happening in YARN-2965 so curious if you
> have thoughts/opinions on this?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
