[ https://issues.apache.org/jira/browse/HBASE-2782?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12909530#action_12909530 ]
stack commented on HBASE-2782: ------------------------------ This patch changes the api for rpc. It likely messes up Gary's work on making rpc pluggable; i.e. secure and non-secure rpc. Let me ask him what he thinks of this. Whats this change mean? - private static final int MAX_QUEUE_SIZE_PER_HANDLER = 100; + private static final int MAX_QUEUE_SIZE_PER_HANDLER = 1000; QoSFunction has to be in HRS? We've been doing work to break up the massive classes. How about a unit test? > QOS for META table access > ------------------------- > > Key: HBASE-2782 > URL: https://issues.apache.org/jira/browse/HBASE-2782 > Project: HBase > Issue Type: Improvement > Components: regionserver > Affects Versions: 0.89.20100621 > Reporter: Todd Lipcon > Assignee: ryan rawson > Fix For: 0.90.0 > > Attachments: HBASE-2782.txt > > > I'd like to brainstorm some ideas on how we can prioritize reads and writes > to META above reads and writes to other tables. I've noticed that if the > regionserver hosting META is under heavy load, then lots of other operations > take much longer than they should. For example, I'm currently running 120 > threads of YCSB across 3 client nodes hitting a 5-node cluster. Doing a full > scan of META (only 600 rows) takes upwards of 30 seconds in the shell, since > all of the handler threads are tied up and there's a long RPC queue. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.