[
https://issues.apache.org/jira/browse/YARN-9120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16721027#comment-16721027
]
Zhankun Tang commented on YARN-9120:
------------------------------------
[~snemeth] , Thanks for the explanation! Agree that it's valuable to enable
updating of the available GPU devices at runtime. Three questions:
# Could you share the new property value of
"yarn.nodemanager.resource-plugins.gpu.allowed-gpu-devices"?
# Per my understanding, YARN NM cannot reload configuration at runtime for
now. I'm afraid the newly added value cannot work that perfectly. And
personally, restarting NM is not a big thing.
# I'm also thinking in what scenario would a user wants a diverse
configuration in the cluster. As far as I know, the Ambari can only update
configuration at the cluster level. Maybe this is useful for a manually managed
Hadoop cluster.
> Need to have a way to turn off GPU auto-discovery in GpuDiscoverer
> ------------------------------------------------------------------
>
> Key: YARN-9120
> URL: https://issues.apache.org/jira/browse/YARN-9120
> Project: Hadoop YARN
> Issue Type: Improvement
> Reporter: Szilard Nemeth
> Assignee: Szilard Nemeth
> Priority: Major
>
> GpuDiscoverer.getGpusUsableByYarn either parses the user-defined GPU devices
> or should have the value 'auto' (from property:
> yarn.nodemanager.resource-plugins.gpu.allowed-gpu-devices)
> In some circumstances, users would want to exclude a node from scheduling, so
> they should have an option to turn off auto-discovery.
> It's straightforward that this is possible by removing the GPU
> resource-plugin from YARN's config along with GPU-related config in
> container-executor.cfg, but doing that with a dedicated value for
> yarn.nodemanager.resource-plugins.gpu.allowed-gpu-devices is a more
> lightweight approach.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
