[FLINK-33388][runtime] Support tasks balancing at TM level for Default Scheduler
What is the purpose of the change
- [FLINK-33387][runtime] Introduce the abstraction and the interface about loading
- [FLINK-33388][runtime] Support tasks balancing at TM level for Default Scheduler
- [FLINK-35966][runtime] Introduce the TASKS for TaskManagerLoadBalanceMode enum and add the users-oriented interface for default scheduler balanced tasks scheduling.
Brief change log
- [FLINK-33387][runtime] Introduce the abstraction and the interface about loading
- [FLINK-33388][runtime] Support tasks balancing at TM level for Default Scheduler
- [FLINK-35966][runtime] Introduce the TASKS for TaskManagerLoadBalanceMode enum and add the users-oriented interface for default scheduler balanced tasks scheduling.
Verifying this change
Add the corresponding test cases.
Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): (yes / no)
- The public API, i.e., is any changed class annotated with
@Public(Evolving): (yes / no) - The serializers: (yes / no / don't know)
- The runtime per-record code paths (performance sensitive): (yes / no / don't know)
- Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn, ZooKeeper: (yes / no / don't know)
- The S3 file system connector: (yes / no / don't know)
Documentation
- Does this pull request introduce a new feature? (yes / no)
- If yes, how is the feature documented? (not applicable / docs / JavaDocs / not documented)
CI report:
- 101761e317766f417394440c3d4dea33d94bf16b Azure: SUCCESS
Bot commands
The @flinkbot bot supports the following commands:@flinkbot run azurere-run the last Azure build
Hi, @zhuzhurk @1996fanrui Could you help take a look if you had the free time ? Any comments would be great appreciated ! :)
https://docs.google.com/document/d/14WhrSNGBdcsRl3IK7CZO-RaZ5KXU2X1dWqxPEFr3iS8/edit?disco=AAAA2t58EuA
In order to achieve slot balance between TMs, should ResourceMatchingStrategy consider usingPrioritizedResourceMatchingStrategy when 'taskManagerLoadBalanceMode' is TASKS
LGTM, +1 to merge. Since the last commit was made about two months ago, could you please rebase onto the latest master branch to trigger the CI on the current codebase, and update the hotfix commit message to use FLINK-33388?
Thanks for your review and suggestions. That makes sense to me. And the state of the current pr was rebased with latest master-branch and the commit message was updated