[
https://issues.apache.org/jira/browse/FLINK-22254?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Till Rohrmann updated FLINK-22254:
----------------------------------
Parent: FLINK-22255
Issue Type: Sub-task (was: Improvement)
> Only trigger scale up if the resources have stabilized
> ------------------------------------------------------
>
> Key: FLINK-22254
> URL: https://issues.apache.org/jira/browse/FLINK-22254
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination
> Affects Versions: 1.13.0
> Reporter: Till Rohrmann
> Priority: Major
> Fix For: 1.14.0
>
>
> In order to improve the scale up behaviour of the {{AdaptiveScheduler}} we
> should only trigger a scale up operation if the resource set has stabilized.
> That way it should be possible to immediately restart the job after its
> cancellation if the resource set has not changed in the meantime.
> One way to implement this behaviour is to only notify the {{State}} about new
> resources after they have stabilized.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
