[
https://issues.apache.org/jira/browse/HBASE-18144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16034056#comment-16034056
]
stack commented on HBASE-18144:
-------------------------------
[~allan163] sorry. Should have been more explicit. Thread dumping we we're
stuck getting locks. I was not able to do a good reproduction of the same
scenario. In my measurements current locking seemed to take only 20% longer but
in the production install the read write lock was plainly taking much longer
than this. Doing thousands of increments a second from hundreds of handlers on
same small set of rows seems to have this effect. Cited quotes and article
indicate acquisition times can degenerate. Hth
> Forward-port the old exclusive row lock; there are scenarios where it
> performs better
> -------------------------------------------------------------------------------------
>
> Key: HBASE-18144
> URL: https://issues.apache.org/jira/browse/HBASE-18144
> Project: HBase
> Issue Type: Bug
> Components: Increment
> Affects Versions: 1.2.5
> Reporter: stack
> Assignee: stack
> Fix For: 2.0.0, 1.3.2, 1.2.7
>
>
> Description to follow.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
