[ https://issues.apache.org/jira/browse/HBASE-16890?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15606424#comment-15606424 ]
stack commented on HBASE-16890: ------------------------------- I did a primitive compare with 100 threads, 10 columns on WALPE. The classic FSHLog in JFR shows a few instances of FSOutputSummer#write and DFSOS#flushOrSync in samples (7 and 3 respectively). The asyncwal by contrast has two orders of magnitude more points of contention showing up in samples at append and sync (900 and 200). The sync on waitingConsumePayloads is a throughput killer. > Analyze the performance of AsyncWAL and fix the same > ---------------------------------------------------- > > Key: HBASE-16890 > URL: https://issues.apache.org/jira/browse/HBASE-16890 > Project: HBase > Issue Type: Sub-task > Components: wal > Affects Versions: 2.0.0 > Reporter: ramkrishna.s.vasudevan > Assignee: ramkrishna.s.vasudevan > Fix For: 2.0.0 > > Attachments: contention.png, contention_defaultWAL.png > > > Tests reveal that AsyncWAL under load in single node cluster performs slower > than the Default WAL. This task is to analyze and see if we could fix it. > See some discussions in the tail of JIRA HBASE-15536. -- This message was sent by Atlassian JIRA (v6.3.4#6332)