[
https://issues.apache.org/jira/browse/YARN-9927?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17334565#comment-17334565
]
Qi Zhu commented on YARN-9927:
------------------------------
[~ebadger] [~gandras] [~epayne]
Updated a patch to improve:
1. The event produce to event queue :
Each eventType will have single async dispatcher thread, the thread number is
small which i have counted. And each dispatcher has single event queue, events
will not affect each other anymore.
2. The event consume from event queue and process:
I add an example to multi thread processing in handler just for RMNodeEvent.
Different RMNodeImpl object has different write lock, i think it can be faster
to make multi thread handle with different RMNode object.
What's your opinions about this?
Thanks.
> RM multi-thread event processing mechanism
> ------------------------------------------
>
> Key: YARN-9927
> URL: https://issues.apache.org/jira/browse/YARN-9927
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: yarn
> Affects Versions: 3.0.0, 2.9.2
> Reporter: hcarrot
> Assignee: Qi Zhu
> Priority: Major
> Attachments: RM multi-thread event processing mechanism.pdf,
> YARN-9927.001.patch, YARN-9927.002.patch, YARN-9927.003.patch,
> YARN-9927.004.patch, YARN-9927.005.patch
>
>
> Recently, we have observed serious event blocking in RM event dispatcher
> queue. After analysis of RM event monitoring data and RM event processing
> logic, we found that
> 1) environment: a cluster with thousands of nodes
> 2) RMNodeStatusEvent dominates 90% time consumption of RM event scheduler
> 3) Meanwhile, RM event processing is in a single-thread mode, and It results
> in the low headroom of RM event scheduler, thus performance of RM.
> So we proposed a RM multi-thread event processing mechanism to improve RM
> performance.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
