You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Qi Zhu (Jira)" <ji...@apache.org> on 2021/03/12 08:10:00 UTC

[jira] [Comment Edited] (YARN-10503) Support queue capacity in terms of absolute resources with gpu resourceType.

    [ https://issues.apache.org/jira/browse/YARN-10503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17300121#comment-17300121 ] 

Qi Zhu edited comment on YARN-10503 at 3/12/21, 8:09 AM:
---------------------------------------------------------

Updated a patch for review.

cc [~pbacsko] [~gandras] [~ebadger] [~Jim_Brennan] [~epayne] [~shuzirra] 

Could you help review this?

I think it's very important for GPU users, in our GPU production cluster, we want to set GPU resource just like memory and vcores in absolute resource mode.

Thanks.

 


was (Author: zhuqi):
Updated a patch for review.

cc [~pbacsko] [~gandras] [~ebadger] [~Jim_Brennan] [~epayne]

Could you help review this?

I think it's very important for GPU users, in our GPU production cluster, we want to set GPU resource just like memory and vcores in absolute resource mode.

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: Improvement
>            Reporter: Qi Zhu
>            Assignee: Qi Zhu
>            Priority: Critical
>         Attachments: YARN-10503.001.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-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org