[
https://issues.apache.org/jira/browse/YARN-6194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Karthik Kambatla updated YARN-6194:
-----------------------------------
Attachment: YARN-6193.000.patch
Patch that:
# Uses policy-based check instead of fitsIn to see if a container can be
preempted
# Move some of the SchedulingPolicy methods to the abstract base class instead
of sub-classes
# Augmented tests to consider DRF for both minshare and fairshare preemption
> Cluster capacity in SchedulingPolicy is updated only on allocation file reload
> ------------------------------------------------------------------------------
>
> Key: YARN-6194
> URL: https://issues.apache.org/jira/browse/YARN-6194
> Project: Hadoop YARN
> Issue Type: Improvement
> Components: fairscheduler
> Affects Versions: 2.8.0
> Reporter: Karthik Kambatla
> Assignee: Yufei Gu
> Attachments: YARN-6193.000.patch
>
>
> Some of the {{SchedulingPolicy}} methods need cluster capacity which is set
> using {{#initialize}} today. However, {{initialize()}} is called only on
> allocation reload. If nodes are added between reloads, the cluster capacity
> is not considered until the next reload.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
