hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Varun Sharma (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-7295) Contention in HBaseClient.getConnection
Date Fri, 07 Dec 2012 00:53:21 GMT

    [ https://issues.apache.org/jira/browse/HBASE-7295?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13526052#comment-13526052
] 

Varun Sharma commented on HBASE-7295:
-------------------------------------

I dont know if we can use putIfAbsent() without doing a createPool since thats the value we
would want to put. Also I thought that get() and put() combination would do worse than putIfAbsent()
in ConcurrentHashMap and would also be less atomic.

We can fix the bug by making the "put" synchronized so that it updates size and adds at the
same time. The get() operation would still not be synchronized and we probably wouldn't want
to synchronize that since we will hit that much more often. The put() will be hit only when
we create connections initially or recreate connections.
                
> Contention in HBaseClient.getConnection
> ---------------------------------------
>
>                 Key: HBASE-7295
>                 URL: https://issues.apache.org/jira/browse/HBASE-7295
>             Project: HBase
>          Issue Type: Improvement
>    Affects Versions: 0.94.3
>            Reporter: Varun Sharma
>            Assignee: Varun Sharma
>             Fix For: 0.94.3
>
>         Attachments: 7295-0.94.txt
>
>
> HBaseClient.getConnection() synchronizes on the connections object. We found severe contention
on a thrift gateway which was fanning out roughly 3000+ calls per second to hbase region servers.
The thrift gateway had 2000+ threads for handling incoming connections. Threads were blocked
on the syncrhonized block - we set ipc.pool.size to 200. Since we are using RoundRobin/ThreadLocal
pool only - its not necessary to synchronize on connections - it might lead to cases where
we might go slightly over the ipc.max.pool.size() but the additional connections would timeout
after maxIdleTime - underlying PoolMap connections object is thread safe.

--
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

Mime
View raw message