Rui Fan

> LGTM 👍 Thanks for your efforts, @1996fanrui, and for keeping up with my nitty comments. I enjoyed the discussions in this PR. Thanks @XComp for the patient review, I...

> Fair enough. What about squashing the [FLINK-33565](https://issues.apache.org/jira/browse/FLINK-33565)-related commits into a single one? Hi @XComp , it's fine for me. Let me explain why I split them into multiple commits:...

Thanks @XComp for the patient review and a series of great suggestions again! Merging~ A good experience~

After this fixing, if the minimum bound isn't satisfied, Adaptive Scheduler won't start the new rescale. It means flink job still run with old parallelism. I'm not sure should we...

Hi @gyfora @mxm , this PR is related to Adaptive Scheduler. Would you mind helping take a look? Big thanks~

https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=55925&view=logs&j=0da23115-68bb-5dcd-192c-bd4c8adebde1&t=24c3384f-1bcb-57b3-224f-51bf973bbee8&l=7264 The CI fails due to the change of this PR, right?

> Hi @1996fanrui, Since this PR is quite large, can you help double-check it? Thanks @GOODBOY008 for the hard work and @Jiabao-Sun for the hard review. I will review this...

@flinkbot run azure

Hi @GOODBOY008 , would you mind rebasing the master branch? If I remember correctly, this ci isuue has been fixed in master branch. Btw, it's better to squash all commits...

Hi @KarmaGYZ , thanks for your hard review! > I think this PR contains two components. First would be a supplement of [FLINK-33448](https://issues.apache.org/jira/browse/FLINK-33448). Second is part of the TASKS strategy....