[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-06-23 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16521376#comment-16521376 ] stack commented on HBASE-20236: --- Attached latest (after HBASE-20780). [^2.0623.111354.lock.svg] >

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-06-23 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16521375#comment-16521375 ] stack commented on HBASE-20236: --- So, MutableSegment locking has been cleaned up in HBASE-20411. The fast

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-04-13 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16437860#comment-16437860 ] stack commented on HBASE-20236: --- I moved the MutableSegment locking to its own issue, HBASE-20411 >

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-04-03 Thread Hadoop QA (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16423678#comment-16423678 ] Hadoop QA commented on HBASE-20236: --- | (x) *{color:red}-1 overall{color}* | \\ \\ || Vote || Subsystem

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-04-02 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16423317#comment-16423317 ] stack commented on HBASE-20236: --- [~eshcar] This showed up in a write-only run where I'm using the default

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-03-28 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16418474#comment-16418474 ] stack commented on HBASE-20236: --- Attached no_semaphore_vs_semaphore.png. First run is no semaphore. Seems

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-03-28 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16418472#comment-16418472 ] stack commented on HBASE-20236: --- Hmm... No. Semaphore is slightly faster (3-4%) even though it has this

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-03-28 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16418453#comment-16418453 ] stack commented on HBASE-20236: --- I just tried a run where we skip the Semaphore and just go to the queue

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-03-28 Thread stack (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16417809#comment-16417809 ] stack commented on HBASE-20236: --- [~eshcar] Thank you. Funny. I was looking at this yesterday. Sorry the doc

[jira] [Commented] (HBASE-20236) [locking] Write-time worst offenders

2018-03-28 Thread Eshcar Hillel (JIRA)
[ https://issues.apache.org/jira/browse/HBASE-20236?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16417228#comment-16417228 ] Eshcar Hillel commented on HBASE-20236: --- I can take a look into this. I see the semaphor used in