[
https://issues.apache.org/jira/browse/YARN-8292?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16477832#comment-16477832
]
Wangda Tan commented on YARN-8292:
----------------------------------
[~eepayne] / [~sunilg],
Since both of you're asking similar questions, I think I should add more
explanations here:
bq. In this proposal, should preemption continue after the above? After a
container is subtracted in the above example, resToObtain == <20, 10, -10> and
the next container request is <10, 15, 5>, would the process continue so we
would have the following?
Yes, it will continue preemption. Assume res-to-obtain is x, and next-container
has y resource. We will stop preemption if all resource types of {{x - y}} are
negative.
Even though this behavior is more aggressive than existing logic, but in my
opinion, this is consistent to how we consider guaranteed capacity (when
dominant resource calculator is enabled).
But for intra-queue preemption, I still want to have less aggressive behavior,
that's why I kept the original logic.
Please let me know your thoughts on this.
> Fix the dominant resource preemption cannot happen when some of the resource
> vector becomes negative
> ----------------------------------------------------------------------------------------------------
>
> Key: YARN-8292
> URL: https://issues.apache.org/jira/browse/YARN-8292
> Project: Hadoop YARN
> Issue Type: Bug
> Components: yarn
> Reporter: Sumana Sathish
> Assignee: Wangda Tan
> Priority: Critical
> Fix For: 3.2.0, 3.1.1
>
> Attachments: YARN-8292.001.patch
>
>
>
> This is an example of the problem: (Same if we have more than 2 resources)
>
> Let's say we have 3 queues A/B/C. All containers with equal size <2,3>
>
> ||Queue||Guaranteed||Used ||Pending||
> |A|<20, 10>|<20,30>| |
> |B|<20, 10>|0|0|
> |C|<20, 10>|0|<20, 30>|
> | | | | |
>
> Under current logic, A's calculated to-preempt (how much resource other queue
> can preempt) will be <0, 20>. The preemption will not happen. However, under
> the context of DRC, queue A is using more resource than guaranteed, so queue
> C will be starved
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
