Manikandan R created YUNIKORN-3133:
--------------------------------------
Summary: [Umbrella] Quota change enforcement via preemption
Key: YUNIKORN-3133
URL: https://issues.apache.org/jira/browse/YUNIKORN-3133
Project: Apache YuniKorn
Issue Type: New Feature
Components: core - scheduler
Reporter: Manikandan R
Assignee: Manikandan R
Fix For: 1.8.0
When a quota is lowered we currently just stop scheduling workloads in that
queue or set of queues. We then rely on the natural pod exits to lower the
usage below the new quota set and then continue scheduling.
This work with short lived applications. It does not work if the workload is
long lived.
Need to look at implementing preemption to lower the usage to the queue quota
within some guidelines
* preempt most recent container?
* grace period allowed?
* daemon set exclusion
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
