[
https://issues.apache.org/jira/browse/YARN-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Wangda Tan updated YARN-6244:
-----------------------------
Attachment: YARN-6244.preliminary.0.patch
Attached preliminary patch.
> Introduce AtomicResource in ResourceUsage to avoid read-write lock.
> -------------------------------------------------------------------
>
> Key: YARN-6244
> URL: https://issues.apache.org/jira/browse/YARN-6244
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: capacityscheduler, fairscheduler, resourcemanager,
> scheduler
> Reporter: Wangda Tan
> Attachments: YARN-6244.preliminary.0.patch
>
>
> While doing SLS tests of YARN-5139, I found when multiple threads are doing
> scheduling at the same time, read/write lock of ResourceUsage becomes
> bottleneck. This could be improved.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
