[
https://issues.apache.org/jira/browse/YARN-6670?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16054487#comment-16054487
]
Karthik Kambatla commented on YARN-6670:
----------------------------------------
I like the idea of making a decision based on the usage over a period of time.
However, that is definitely more complicated and can be added at a later stage.
Unless the applications are have strong synchronization barriers to ensure they
all start/stop at the same time, the approach of using the snapshot usage -
over tasks from multiple applications - should not be too far off from the
optimal over a period. [~haibo.chen] - can we file another JIRA to continue the
conversation around time periods?
+1 on the latest patch.
> Add separate NM overallocation thresholds for cpu and memory
> ------------------------------------------------------------
>
> Key: YARN-6670
> URL: https://issues.apache.org/jira/browse/YARN-6670
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager
> Affects Versions: 3.0.0-alpha3
> Reporter: Haibo Chen
> Assignee: Haibo Chen
> Attachments: YARN-6670-YARN-1011.00.patch,
> YARN-6670-YARN-1011.01.patch, YARN-6670-YARN-1011.02.patch,
> YARN-6670-YARN-1011.03.patch, YARN-6670-YARN-1011.04.patch,
> YARN-6670-YARN-1011.05.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
