[
https://issues.apache.org/jira/browse/FLINK-27555?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17534281#comment-17534281
]
Roman Khachatryan commented on FLINK-27555:
-------------------------------------------
Thanks for the clarification [~zhuzh].
Besides of the description, there is also a regression script added in
[https://github.com/apache/flink-benchmarks/pull/50]
which actually reported this change as a regression.
To prevent such false positives, I see three options:
# modify the benchmark (e.g. report 1/time)
# modify the script so that it would treat distinguish "lower is better"
benchmarks differently
# ignore the benchmark in script
WDYT?
> Performance regression in schedulingDownstreamTasks on 02.05.2022
> -----------------------------------------------------------------
>
> Key: FLINK-27555
> URL: https://issues.apache.org/jira/browse/FLINK-27555
> Project: Flink
> Issue Type: Bug
> Components: Benchmarks
> Affects Versions: 1.16.0
> Reporter: Roman Khachatryan
> Priority: Major
> Attachments: Screenshot_2022-05-09_10-33-11.png
>
>
> http://codespeed.dak8s.net:8000/timeline/#/?exe=5&ben=schedulingDownstreamTasks.BATCH&extr=on&quarts=on&equid=off&env=2&revs=200
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
