[
https://issues.apache.org/jira/browse/YARN-3136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14306627#comment-14306627
]
Sunil G commented on YARN-3136:
-------------------------------
Hi [~jianhe]
bq. getTransferredContainers does not need to be invoked if
ApplicationSubmissionContext#getKeepContainersAcrossApplicationAttempts is
false.
Yes. It could be skipped.
bq.synchronization on getTransferredContainers is not needed
Yes. This is to be removed. But it internally invokes *applications* map. So a
synchronized getter can be invoked from *getTransferredContainers* to get the
application object for list of containers. Pls share your opinion.
> 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)
