[
https://issues.apache.org/jira/browse/YARN-10154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17057771#comment-17057771
]
Sunil G commented on YARN-10154:
--------------------------------
Thanks [~clayb].
Regarding to the configuration
_+yarn.scheduler.capacity.<queue-path>.leaf-queue-template.maximum-allocation-mb+_
and
+_yarn.scheduler.capacity.<queue-path>.leaf-queue-template.maximum-allocation-vcores_+,
there may be a problem with respect to multiple resources design model. As we
may have more resources such like GPU or FPGA, then this configuration model
may not scale to this.
Could we use below model like
+_yarn.scheduler.capacity.<queue-path>.leaf-queue-template.capacity_+ =
"[memory=10240, vcores=10]"
This is exactly similar to normal absolute resource design we have done in
YARN-5881. In that case as well, we were configuring
+_yarn.scheduler.capacity.<queue-path>.capacity_+ = "[memory=10240, vcores=10,
gpu=6]".
[~clayb] does this make sense ? Kindly share your thoughts.
> CS Dynamic Queues cannot be configured with absolute resources
> --------------------------------------------------------------
>
> Key: YARN-10154
> URL: https://issues.apache.org/jira/browse/YARN-10154
> Project: Hadoop YARN
> Issue Type: Bug
> Affects Versions: 3.1.3
> Reporter: Sunil G
> Assignee: Manikandan R
> Priority: Major
> Attachments: YARN-10154.001.patch
>
>
> In CS, ManagedParent Queue and its template cannot take absolute resource
> value like
> [memory=8192,vcores=8]
> Thsi Jira is to track and improve the configuration reading module of
> DynamicQueue to support absolute resource values.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
