[ https://issues.apache.org/jira/browse/YARN-10503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17303185#comment-17303185 ]
Andras Gyori commented on YARN-10503: ------------------------------------- I agree with [~ebadger]. Not only do we need to extend this functionality to arbitrary resource types, but we also need to support these types on other metrics as well (max/min/absolute capacity). This initiative has already been started as per YARN-9936. I would be inclined to do this approach instead, with flexibility in mind (to support not only GPU, but arbitrary resource types). > 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