[
https://issues.apache.org/jira/browse/YARN-9756?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16911513#comment-16911513
]
Manikandan R commented on YARN-9756:
------------------------------------
{quote}These new metrics will be similar to AggregateMemoryMBSecondsPreempted,
AggregateVcoreSecondsPreempted, etc. I propose to process the total preempted
resources in the same way that is done for preempted seconds (memory, vcores,
etc).
LeafQueue#updateQueuePreemptionMetrics will aggregate the total preempted
resources just like it does for preempted resource seconds.{quote}
Ok, I was bit confused with "per round". Attached a quick WIP patch for your
review.
{quote}The challenge I have encountered is making this work for extended
resources (like gpu, etc.){quote}
Can {{QueueMetricsForCustomResources}} be used to generate this metric like
other metrics for GPU? Have covered this also in the patch. Please share your
views.
> Create metric that sums total memory/vcores preempted per round
> ---------------------------------------------------------------
>
> Key: YARN-9756
> URL: https://issues.apache.org/jira/browse/YARN-9756
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: capacity scheduler
> Affects Versions: 3.2.0, 2.9.2, 3.0.3, 2.8.5, 3.1.2
> Reporter: Eric Payne
> Assignee: Eric Payne
> Priority: Major
> Attachments: YARN-9756.WIP.patch
>
>
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
