[
https://issues.apache.org/jira/browse/YARN-10503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17307421#comment-17307421
]
Eric Badger commented on YARN-10503:
------------------------------------
bq. Do we want to treat GPUs and FPGAs like that? In other parts of the code,
we have mem/vcore as primary resources, then an array of other resources.
I believe the correct approach is to leave memroy and vcores as "first class"
resources and then add on logic to add arbitrary extended resources, such as
GPU or FPGA. The arbitrary extended resources should not be hardcoded values.
The point is that we're doing the work right now to support GPUs. But in 2
years if some new resource needs to be tracked and used, we don't want to have
to redo all of this work again. We should make sure that our work right here is
extended to any future arbitrary resources
> Support queue capacity in terms of absolute resources with custom
> 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,
> YARN-10503.003.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: [email protected]
For additional commands, e-mail: [email protected]
