[
https://issues.apache.org/jira/browse/YARN-6384?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15943704#comment-15943704
]
Miklos Szegedi commented on YARN-6384:
--------------------------------------
bq. This may be confused, the
yarn.nodemanager.resource.percentage-physical-cpu-limit can limit sum of
cpu-usage of all containers, but there has no upper limit for each container.
If you look at CGroupsCpuResourceHandlerImpl.preStart it does exactly what you
are missing, a limit per each container:
{code}
if (strictResourceUsageMode) {
if (nodeVCores != containerVCores) {
float containerCPU =
(containerVCores * yarnProcessors) / (float) nodeVCores;
int[] limits = getOverallLimits(containerCPU);
cGroupsHandler.updateCGroupParam(CPU, cgroupId,
CGroupsHandler.CGROUP_CPU_PERIOD_US, String.valueOf(limits[0]));
cGroupsHandler.updateCGroupParam(CPU, cgroupId,
CGroupsHandler.CGROUP_CPU_QUOTA_US, String.valueOf(limits[1]));
}
}
{code}
> Add configuratin to set max cpu usage when strict-resource-usage is false
> with cgroups
> --------------------------------------------------------------------------------------
>
> Key: YARN-6384
> URL: https://issues.apache.org/jira/browse/YARN-6384
> Project: Hadoop YARN
> Issue Type: Improvement
> Reporter: dengkai
>
> When using cgroups on yarn, if
> yarn.nodemanager.linux-container-executor.cgroups.strict-resource-usage is
> false, user may get very more cpu time than expected based on the vcores.
> There should be a upper limit even resource-usage is not strict, just like a
> percentage which user can get more than promised by vcores. I think it's
> important in a shared cluster.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
