[ https://issues.apache.org/jira/browse/HBASE-15971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15326666#comment-15326666 ]
stack edited comment on HBASE-15971 at 6/12/16 10:43 PM: --------------------------------------------------------- HBASE-11355 "a couple of callQueue related improvements" when applied on branch-1 actually 3x'd the perf but throughput numbers were super low before it was applied; we went from 80k to 230k. The following commit dropped the branch-1 throughput from 180k to 80k: commit de98955c16d71f2486f54f88c9ca6c88b91a3812 Author: mbertozzi <mbertozzi@unknown> Date: Wed May 7 21:05:04 2014 +0000 HBASE-10993 Deprioritize long-running scanners Checking to see if I revert the above and use scheduler from 0.98, see if I have same throughput as 0.98 branch was (Author: stack): HBASE-11355 "a couple of callQueue related improvements" when applied on branch-1 actually 3x'd the perf but throughput numbers were super low before it was applied; we went from 80k to 230k. The following commit dropped the branch-1 throughput from 180k to 80k: commit de98955c16d71f2486f54f88c9ca6c88b91a3812 Author: mbertozzi <mbertozzi@unknown> Date: Wed May 7 21:05:04 2014 +0000 HBASE-10993 Deprioritize long-running scanners Checking to see if I revert both of the above, does branch-1 perf come back up to 0.98s. > Regression: Random Read/WorkloadC slower in 1.x than 0.98 > --------------------------------------------------------- > > Key: HBASE-15971 > URL: https://issues.apache.org/jira/browse/HBASE-15971 > Project: HBase > Issue Type: Sub-task > Components: rpc > Reporter: stack > Assignee: stack > Priority: Critical > Attachments: 098.hits.png, 098.png, HBASE-15971.branch-1.001.patch, > Screen Shot 2016-06-10 at 5.08.24 PM.png, Screen Shot 2016-06-10 at 5.08.26 > PM.png, branch-1.hits.png, branch-1.png, > flight_recording_10172402220203_28.branch-1.jfr, > flight_recording_10172402220203_29.09820.0.98.20.jfr, handlers.fp.png, > hits.fp.png, hits.patched1.0.vs.unpatched1.0.vs.098.png, run_ycsb.sh > > > branch-1 is slower than 0.98 doing YCSB random read/workloadC. It seems to be > doing about 1/2 the throughput of 0.98. > In branch-1, we have low handler occupancy compared to 0.98. Hacking in > reader thread occupancy metric, is about the same in both. In parent issue, > hacking out the scheduler, I am able to get branch-1 to go 3x faster so will > dig in here. -- This message was sent by Atlassian JIRA (v6.3.4#6332)