[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17080066#comment-17080066 ] Hudson commented on HBASE-24099: Results for branch branch-2.2 [build #835 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17079985#comment-17079985 ] Hudson commented on HBASE-24099: Results for branch branch-1.4 [build #1220 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17079344#comment-17079344 ] Hudson commented on HBASE-24099: Results for branch master [build #1693 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17079245#comment-17079245 ] Hudson commented on HBASE-24099: Results for branch branch-1 [build #1280 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17079220#comment-17079220 ] Hudson commented on HBASE-24099: Results for branch branch-2.3 [build #23 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17078933#comment-17078933 ] Hudson commented on HBASE-24099: Results for branch branch-2 [build #2591 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=1707#comment-1707 ] Hudson commented on HBASE-24099: Results for branch branch-2.1 [build #1851 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17078787#comment-17078787 ] Hudson commented on HBASE-24099: Results for branch branch-1.3 [build #1172 on

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17077656#comment-17077656 ] Andrew Kyle Purtell commented on HBASE-24099: - I don't think there is a real scan penalty.

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076908#comment-17076908 ] Reid Chan commented on HBASE-24099: --- Thanks for the efforts [~apurtell]!  I'm still a fan of para

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076842#comment-17076842 ] David Manning commented on HBASE-24099: --- Yep, understood. I was just hoping someone could explain

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076831#comment-17076831 ] Andrew Kyle Purtell commented on HBASE-24099: - bq. I'm guessing very few of those

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076828#comment-17076828 ] Andrew Kyle Purtell commented on HBASE-24099: - bq. Put another way, there should be

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076799#comment-17076799 ] David Manning commented on HBASE-24099: --- I am not an expert on the locks, but looking through the

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076776#comment-17076776 ] Andrew Kyle Purtell commented on HBASE-24099: - So here's the trade off: We might starve the

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17076769#comment-17076769 ] Andrew Kyle Purtell commented on HBASE-24099: - Digging into scan degradation with PE, let me

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074247#comment-17074247 ] Reid Chan commented on HBASE-24099: --- bq. I’ve got more perf tests to do. If they turn up a significant

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074245#comment-17074245 ] Reid Chan commented on HBASE-24099: --- What i'm thinking about is the scan penalty may not be tasty to

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074244#comment-17074244 ] Andrew Kyle Purtell commented on HBASE-24099: - I’ve got more perf tests to do. If they turn

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074240#comment-17074240 ] Reid Chan commented on HBASE-24099: --- >> This is because sometimes we don’t have the confidence to make

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074227#comment-17074227 ] Andrew Kyle Purtell commented on HBASE-24099: - We could, however we have way too many

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074206#comment-17074206 ] Reid Chan commented on HBASE-24099: --- Can we make this a parameter. e.g, key is .lock.fair.mode,

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074198#comment-17074198 ] Andrew Kyle Purtell commented on HBASE-24099: - YCSB results show mostly within variance I

[jira] [Commented] (HBASE-24099) Use a fair ReentrantReadWriteLock for the region close lock

[ https://issues.apache.org/jira/browse/HBASE-24099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17074116#comment-17074116 ] Xu Cang commented on HBASE-24099: - >LTT results look like any difference is within normal variance.