You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Maximilian Michels (Jira)" <ji...@apache.org> on 2023/03/06 18:08:00 UTC
[jira] [Reopened] (FLINK-30575) Autoscaler doesn't scale down on 0 load
[ https://issues.apache.org/jira/browse/FLINK-30575?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Maximilian Michels reopened FLINK-30575:
----------------------------------------
> Autoscaler doesn't scale down on 0 load
> ---------------------------------------
>
> Key: FLINK-30575
> URL: https://issues.apache.org/jira/browse/FLINK-30575
> Project: Flink
> Issue Type: New Feature
> Components: Autoscaler, Kubernetes Operator
> Reporter: Gyula Fora
> Assignee: Maximilian Michels
> Priority: Major
> Labels: pull-request-available
> Fix For: kubernetes-operator-1.5.0
>
>
> The logic for computing true processing rates rely on recordsProcessed / recordProcessingTime style computations which do not really work well when everything is 0.
> This leads to no scaling actions when the load suddenly drops to 0. We should handle these special cases and scale the pipeline to some minimal parallelism.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)