[ https://issues.apache.org/jira/browse/YARN-4948?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15366158#comment-15366158 ]
jialei weng commented on YARN-4948: ----------------------------------- Hi, [~Naganarasimha], Thanks for reply. The main scenario for this is to give a way to decouple yarn with HDFS. Since nodelabel is a very important data for yarn, if hdfs down, yarn still fail to start up. So I think this change is meaningful for make yarn much independence when user serve both yarn and HDFS. I make a way to beyond the zknode 1MB storage limitation. In UT, it can support 100000 nodemanager, recover all 100000 node nodelabel cost 4 seconds. It is really efficiency and reliable. > Support node labels store in zookeeper > -------------------------------------- > > Key: YARN-4948 > URL: https://issues.apache.org/jira/browse/YARN-4948 > Project: Hadoop YARN > Issue Type: Sub-task > Components: resourcemanager > Reporter: jialei weng > Assignee: jialei weng > Attachments: YARN-4948.001.patch, YARN-4948.002.patch, > YARN-4948.003.patch, YARN-4948.004.patch > > > Support node labels store in zookeeper -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org