[
https://issues.apache.org/jira/browse/MESOS-2652?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ian Downes updated MESOS-2652:
------------------------------
Description:
The CPU isolator needs to properly set limits for revocable and non-revocable
containers.
The proposed strategy is to use a two-way split of the cpu cgroup hierarchy --
normal (non-revocable) and low priority (revocable) subtrees -- and to use a
biased split of CFS cpu.shares across the subtrees, e.g., a 20:1 split.
Containers would be present in only one of the subtrees. CFS quotas will *not*
be set on subtree roots, only cpu.shares. Each container would set CFS quota
and shares as done currently.
If we end up using different hierarchies for revocable vs non-revocable, this
needs extensive testing to figure out the right knobs to use to minimize
interference.
was:
The CPU isolator needs to properly set limits for revocable and non-revocable
containers.
If we end up using different hierarchies for revocable vs non-revocable, this
needs extensive testing to figure out the right knobs to use to minimize
interference.
> Update Mesos containerizer to understand revocable cpu resources
> ----------------------------------------------------------------
>
> Key: MESOS-2652
> URL: https://issues.apache.org/jira/browse/MESOS-2652
> Project: Mesos
> Issue Type: Task
> Reporter: Vinod Kone
>
> The CPU isolator needs to properly set limits for revocable and non-revocable
> containers.
> The proposed strategy is to use a two-way split of the cpu cgroup hierarchy
> -- normal (non-revocable) and low priority (revocable) subtrees -- and to use
> a biased split of CFS cpu.shares across the subtrees, e.g., a 20:1 split.
> Containers would be present in only one of the subtrees. CFS quotas will
> *not* be set on subtree roots, only cpu.shares. Each container would set CFS
> quota and shares as done currently.
> If we end up using different hierarchies for revocable vs non-revocable, this
> needs extensive testing to figure out the right knobs to use to minimize
> interference.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
