[
https://issues.apache.org/jira/browse/YARN-6668?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Miklos Szegedi updated YARN-6668:
---------------------------------
Attachment: YARN-6668.009.patch
Thank you for the reviews [~haibochen]. I attached a test that is more like a
functional test comparing the results of the procfs and the cgroups based
resource calculators. In general CPU is good. We have some differences in
memory calculations, though. The checks in the code show the details.
> Use cgroup to get container resource utilization
> ------------------------------------------------
>
> Key: YARN-6668
> URL: https://issues.apache.org/jira/browse/YARN-6668
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager
> Affects Versions: 3.0.0-alpha3
> Reporter: Haibo Chen
> Assignee: Miklos Szegedi
> Attachments: YARN-6668.000.patch, YARN-6668.001.patch,
> YARN-6668.002.patch, YARN-6668.003.patch, YARN-6668.004.patch,
> YARN-6668.005.patch, YARN-6668.006.patch, YARN-6668.007.patch,
> YARN-6668.008.patch, YARN-6668.009.patch
>
>
> Container Monitor relies on proc file system to get container resource
> utilization, which is not as efficient as reading cgroup accounting. We
> should in NM, when cgroup is enabled, read cgroup stats instead.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
