hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Daryn Sharp (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-9955) RPC idle connection closing is extremely inefficient
Date Fri, 13 Sep 2013 17:26:52 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-9955?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13766683#comment-13766683
] 

Daryn Sharp commented on HADOOP-9955:
-------------------------------------

Sure, I can make changes 1 & 3.  I think maintaining #2 is still worthwhile.  {{size}}
on the concurrent collection isn't necessarily very speedy.  It has to compute the size which
may involve locking all the internal segments which make it concurrent.  While currently not
called much outside of log statements, it will be called more frequently in a future patch
to limit connections to avoid running out of fds since this and HADOOP-9956 increase the odds
that a stampede will exhaust fds.
                
> RPC idle connection closing is extremely inefficient
> ----------------------------------------------------
>
>                 Key: HADOOP-9955
>                 URL: https://issues.apache.org/jira/browse/HADOOP-9955
>             Project: Hadoop Common
>          Issue Type: Sub-task
>          Components: ipc
>    Affects Versions: 2.0.0-alpha, 3.0.0
>            Reporter: Daryn Sharp
>            Assignee: Daryn Sharp
>         Attachments: HADOOP-9955.patch
>
>
> The RPC server listener loops accepting connections, distributing the new connections
to socket readers, and then conditionally & periodically performs a scan for idle connections.
 The idle scan choses a _random index range_ to scan in a _synchronized linked list_.
> With 20k+ connections, walking the range of indices in the linked list is extremely expensive.
 During the sweep, other threads (socket responder and readers) that want to close connections
are blocked, and no new connections are being accepted.

--
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