hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sunil G (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-3136) getTransferredContainers can be a bottleneck during AM registration
Date Thu, 05 Feb 2015 06:46:35 GMT

    [ https://issues.apache.org/jira/browse/YARN-3136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14306748#comment-14306748
] 

Sunil G commented on YARN-3136:
-------------------------------

Hi [~rohithsharma]

*getLiveContainers()* is giving a copy as you mentioned. Earlier I was looking for getter
method to fetch FicaSchedulerApp from *applications* with schedule lock. But that also not
needed.
{code}
Map<ApplicationId, SchedulerApplication<T>> applications

this.applications =
         new ConcurrentHashMap<ApplicationId,
             SchedulerApplication<FiCaSchedulerApp>>();
{code}
*applications* is a concurrent hash map. Hence synchronized can be removed from *getTransferredContainers*
without any change in the method itself. Also as mentioned by [~jianhe], a check will be added
on *registerApplicationMaster()*.

> getTransferredContainers can be a bottleneck during AM registration
> -------------------------------------------------------------------
>
>                 Key: YARN-3136
>                 URL: https://issues.apache.org/jira/browse/YARN-3136
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: scheduler
>    Affects Versions: 2.6.0
>            Reporter: Jason Lowe
>            Assignee: Sunil G
>
> While examining RM stack traces on a busy cluster I noticed a pattern of AMs stuck waiting
for the scheduler lock trying to call getTransferredContainers.  The scheduler lock is highly
contended, especially on a large cluster with many nodes heartbeating, and it would be nice
if we could find a way to eliminate the need to grab this lock during this call.  We've already
done similar work during AM allocate calls to make sure they don't needlessly grab the scheduler
lock, and it would be good to do so here as well, if possible.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message