[
https://issues.apache.org/jira/browse/YARN-10503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17303206#comment-17303206
]
Qi Zhu commented on YARN-10503:
-------------------------------
Thanks [~gandras] [~ebadger] for review:
Actually, i just want to support absolute conf to support GPU in our production
cluster, and i think if we don't support this, the gpu will be consistent with
(memory / total) * GPU total, it's not reasonable, and YARN-9936 still not
going now.
So i proposal for this, to make gpu reasonable in absolute mode.
What your opinion about "the gpu will be consistent with (memory / total) * GPU
total, it's not reasonable" ?
Thanks.
> Support queue capacity in terms of absolute resources with gpu resourceType.
> ----------------------------------------------------------------------------
>
> Key: YARN-10503
> URL: https://issues.apache.org/jira/browse/YARN-10503
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Qi Zhu
> Assignee: Qi Zhu
> Priority: Critical
> Attachments: YARN-10503.001.patch, YARN-10503.002.patch
>
>
> Now the absolute resources are memory and cores.
> {code:java}
> /**
> * Different resource types supported.
> */
> public enum AbsoluteResourceType {
> MEMORY, VCORES;
> }{code}
> But in our GPU production clusters, we need to support more resourceTypes.
> It's very import for cluster scaling when with different resourceType
> absolute demands.
>
> This Jira will handle GPU first.
--
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
