[
https://issues.apache.org/jira/browse/YARN-8095?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16429595#comment-16429595
]
Weiwei Yang commented on YARN-8095:
-----------------------------------
Hi [~kyungwan nam]
Here is your config:
||Queue||Priority||Capacity||Accessible-Parition||
|root.longlived|0|50~100|default|
|root.batch|1|50~100|default|
|root.label1|0|0|label1|
Looks like the used-resource by {{root.batch}} and {{root.longlived}} is bigger
than total resource in {{default}} partition, then it starts to use free
resource from {{root.label1}}. When an app is submitted to {{root.label1}},
then it needs to kick-off the preemption for resource. But P(root.batch) >
P(root.longlived) && U(root.longlived) > 50%, that's why container from
{{root.longlived}} was selected for preemption right? In this case, why don't
you set {{root.longlived}} priority higher if you "don’t want long-lived apps
to be killed abruptly"?
Thanks
> Allow disable non-exclusive allocation
> --------------------------------------
>
> Key: YARN-8095
> URL: https://issues.apache.org/jira/browse/YARN-8095
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: capacity scheduler
> Affects Versions: 2.8.3
> Reporter: kyungwan nam
> Priority: Major
>
> We have 'longlived' Queue, which is used for long-lived apps.
> In situation where default Partition resources are not enough, containers for
> long-lived app can be allocated to sharable Partition.
> Since then, containers for long-lived app can be easily preempted.
> We don’t want long-lived apps to be killed abruptly.
> Currently, non-exclusive allocation can happen regardless of whether the
> queue is accessible to the sharable Partition.
> It would be good if non-exclusive allocation can be disabled at queue level.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
