[ https://issues.apache.org/jira/browse/YARN-6384?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15943508#comment-15943508 ]
dengkai commented on YARN-6384: ------------------------------- This may be confused, the yarn.nodemanager.resource.percentage-physical-cpu-limit can limit sum of cpu-usage of all containers, but there has no upper limit for each one container. So a container with few vcores will use much cpu more than promised, I think it's unreasonable even the cluster has massive spare resources. https://issues.apache.org/jira/browse/YARN-6384?focusedCommentId=15940051&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15940051 > Add configuratin to set max cpu usage when strict-resource-usage is false > with cgroups > -------------------------------------------------------------------------------------- > > Key: YARN-6384 > URL: https://issues.apache.org/jira/browse/YARN-6384 > Project: Hadoop YARN > Issue Type: Improvement > Reporter: dengkai > > When using cgroups on yarn, if > yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage is > false, user may get very more cpu time than expected based on the vcores. > There should be a upper limit even resource-usage is not strict, just like a > percentage which user can get more than promised by vcores. I think it's > important in a shared cluster. -- This message was sent by Atlassian JIRA (v6.3.15#6346) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org