[jira] [Updated] (HBASE-8316) JoinedHeap for non essential column families should reseek instead of seek
[ https://issues.apache.org/jira/browse/HBASE-8316?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Hofhansl updated HBASE-8316: - Resolution: Fixed Status: Resolved (was: Patch Available) Committed to 0.94, 0.95, and 0.98. Thanks for looking and improving, Ted! > JoinedHeap for non essential column families should reseek instead of seek > -- > > Key: HBASE-8316 > URL: https://issues.apache.org/jira/browse/HBASE-8316 > Project: HBase > Issue Type: Sub-task > Components: Filters, Performance, regionserver >Reporter: Lars Hofhansl >Assignee: Lars Hofhansl > Fix For: 0.98.0, 0.94.7, 0.95.1 > > Attachments: 8316-0.94.txt, 8316-trunk.txt, 8316-trunk.txt, > FDencode.png, noencode.png > > > This was raised by the Phoenix team. During a profiling session we noticed > that catching the joinedHeap up to the current rows via seek causes a > performance regression, which makes the joinedHeap only efficient when either > a high or low percentage is matched by the filter. > (High is fine, because the joinedHeap will not get behind as often and does > not need to be caught up, low is fine, because the seek isn't happening > frequently). > In our tests we found that the solution is quite simple: Replace seek with > reseek. Patch coming soon. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-8316) JoinedHeap for non essential column families should reseek instead of seek
[ https://issues.apache.org/jira/browse/HBASE-8316?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Hofhansl updated HBASE-8316: - Attachment: noencode.png FDencode.png Benchmark results (replacing seek with reseek, not requestSeek, but that will only make it better). The dotted line were the result before the change. > JoinedHeap for non essential column families should reseek instead of seek > -- > > Key: HBASE-8316 > URL: https://issues.apache.org/jira/browse/HBASE-8316 > Project: HBase > Issue Type: Sub-task > Components: Filters, Performance, regionserver >Reporter: Lars Hofhansl >Assignee: Lars Hofhansl > Fix For: 0.98.0, 0.94.7, 0.95.1 > > Attachments: 8316-0.94.txt, 8316-trunk.txt, 8316-trunk.txt, > FDencode.png, noencode.png > > > This was raised by the Phoenix team. During a profiling session we noticed > that catching the joinedHeap up to the current rows via seek causes a > performance regression, which makes the joinedHeap only efficient when either > a high or low percentage is matched by the filter. > (High is fine, because the joinedHeap will not get behind as often and does > not need to be caught up, low is fine, because the seek isn't happening > frequently). > In our tests we found that the solution is quite simple: Replace seek with > reseek. Patch coming soon. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Updated] (HBASE-8316) JoinedHeap for non essential column families should reseek instead of seek
[ https://issues.apache.org/jira/browse/HBASE-8316?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Anoop Sam John updated HBASE-8316: -- Summary: JoinedHeap for non essential column families should reseek instead of seek (was: JoinedHeap for essential column families should reseek instead of seek) > JoinedHeap for non essential column families should reseek instead of seek > -- > > Key: HBASE-8316 > URL: https://issues.apache.org/jira/browse/HBASE-8316 > Project: HBase > Issue Type: Sub-task > Components: Filters, Performance, regionserver >Reporter: Lars Hofhansl > Fix For: 0.98.0, 0.94.7, 0.95.1 > > Attachments: 8316-0.94.txt, 8316-trunk.txt, 8316-trunk.txt > > > This was raised by the Phoenix team. During a profiling session we noticed > that catching the joinedHeap up to the current rows via seek causes a > performance regression, which makes the joinedHeap only efficient when either > a high or low percentage is matched by the filter. > (High is fine, because the joinedHeap will not get behind as often and does > not need to be caught up, low is fine, because the seek isn't happening > frequently). > In our tests we found that the solution is quite simple: Replace seek with > reseek. Patch coming soon. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira