[ https://issues.apache.org/jira/browse/HBASE-11209?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Purtell updated HBASE-11209: ----------------------------------- Attachment: HBASE-11209.patch A suitable patch. Will commit with credit to N after the SVN->GIT migration is complete. > Increase the default value for hbase.hregion.memstore.block.multipler from 2 > to 4 > --------------------------------------------------------------------------------- > > Key: HBASE-11209 > URL: https://issues.apache.org/jira/browse/HBASE-11209 > Project: HBase > Issue Type: Brainstorming > Components: regionserver > Affects Versions: 0.99.0, 0.98.2 > Reporter: Nicolas Liochon > Assignee: Nicolas Liochon > Fix For: 0.99.0, 0.98.3 > > Attachments: HBASE-11209.patch > > > On a YCSB test, I saw a 33% performance increase, both on the max latency and > on the throughput. I'm convinced enough that this value is better that I > think it makes sense to change it on 0.98 as well. > More fundamentally, but outside of the scope of this patch, I think this > parameter should be changed to something at the region server level: today, > we have: > - global memstore check: if we're other 40%, we flush the biggest memstore > - local: no more than 2 (proposed: 4) memstore size per region. > But if we have enough memory and a spike on a region, there is no reason for > not taking the write. -- This message was sent by Atlassian JIRA (v6.2#6252)