hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vinod Kumar Vavilapalli (Updated) (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (MAPREDUCE-3572) MR AM's dispatcher is blocked by heartbeats to ResourceManager
Date Fri, 16 Dec 2011 01:13:30 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-3572?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Vinod Kumar Vavilapalli updated MAPREDUCE-3572:

    Attachment: MAPREDUCE-3572-20111215.txt

Raw patch moving the event processing in RMContainerAllocator into a separate thread so that
the rest of the AM can do its work. The patch may need a bit of cleanup and tests.
> MR AM's dispatcher is blocked by heartbeats to ResourceManager
> --------------------------------------------------------------
>                 Key: MAPREDUCE-3572
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-3572
>             Project: Hadoop Map/Reduce
>          Issue Type: Sub-task
>          Components: mr-am, mrv2, performance
>    Affects Versions: 0.23.0
>            Reporter: Vinod Kumar Vavilapalli
>            Assignee: Vinod Kumar Vavilapalli
>             Fix For: 0.23.1
>         Attachments: MAPREDUCE-3572-20111215.txt
> All the heartbeat processing is done in {{RMContainerAllocator}} locking the object.
The event processing is also locked on this, causing the dispatcher to be blocked and the
rest of the AM getting stalled.
> The event processing should be in a separate thread.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message