lsyldliu opened a new pull request, #23251:
URL: https://github.com/apache/flink/pull/23251
## What is the purpose of the change
In the predecessor PR, for the purpose of delayed computation, for Broadcast
HashJoin, we strongly assume that the build side won' spill to disk because it
less than 10MB. However, during the testing of TPC-DS, part of the query, due
to the problem of the inaccurate derivation of statistics by the optimizer,
although the optimizer chose Broadcast HashJoin, the spill behavior happened
during actual runtime pahse, so we need to remove this assumption and make it
support spill as well. For this fix, the gains obtained through TPC-DS are
basically the same as expected
## Brief change log
- *Fix the bug of broadcast hash join doesn't support spill to disk when
enabling operator fusion codegen*
## Verifying this change
This change is already covered by existing tests in
OperatorFusionCodegenITCase.
## 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? (not documented)
--
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
