SinBex opened a new pull request, #24170: URL: https://github.com/apache/flink/pull/24170
## What is the purpose of the change Currently, if users do not set the `execution.batch.adaptive.auto-parallelism.default-source-parallelism` configuration option, the AdaptiveBatchScheduler defaults to a parallelism of 1 for source vertices. In [FLIP-379](https://cwiki.apache.org/confluence/display/FLINK/FLIP-379%3A+Dynamic+source+parallelism+inference+for+batch+jobs#FLIP379:Dynamicsourceparallelisminferenceforbatchjobs-IntroduceDynamicParallelismInferenceinterfaceforSource), the value of `execution.batch.adaptive.auto-parallelism.default-source-parallelism` will act as the upper bound for inferring dynamic source parallelism, and continuing with the current policy is no longer appropriate. We plan to change the effectiveness strategy of `execution.batch.adaptive.auto-parallelism.default-source-parallelism`; when the user does not set this config option, we will use the value of `execution.batch.adaptive.auto-parallelism.max-parallelism` as the upper bound for source parallelism inference. If `execution.batch.adaptive.auto-parallelism.max-parallelism` is also not configured, the value of `parallelism.default` will be used as a fallback. ## Brief change log - *Change the logic for determining the default source parallelism value; when `execution.batch.adaptive.auto-parallelism.default-source-parallelism` is not configured, fall back to the global max parallelism.* ## Verifying this change This change added tests and can be verified as follows: - *Added unit test in DefaultVertexParallelismAndInputInfosDeciderTest#testComputeSourceParallelismUpperBound to test the case where default source parallelism is not configured.* ## Does this pull request potentially affect one of the following parts: - Dependencies (does it add or upgrade a dependency): (no) - The public API, i.e., is any changed class annotated with `@Public(Evolving)`: ( no) - The serializers: (no ) - The runtime per-record code paths (performance sensitive): (no ) - Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn, ZooKeeper: (no) - The S3 file system connector: (no) ## Documentation - Does this pull request introduce a new feature? (no) - If yes, how is the feature documented? (docs) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
