alamb opened a new issue, #18781: URL: https://github.com/apache/datafusion/issues/18781
### Is your feature request related to a problem or challenge? - Part of https://github.com/apache/datafusion/issues/18779 The [`LimitedBatchCoalescer`](https://github.com/apache/datafusion/blob/e4bc5148171ef3c43698662b1674e6b1ae111366/datafusion/physical-plan/src/coalesce/mod.rs#L27) can now be used to integrate batch coalescing directly in `ExecutionPlan` implementations, which is better than using an external operator for the reasons described on https://github.com/apache/datafusion/issues/18779 Currently `HashJoinExec` needs a `CoalesceBatches` operator after for performance, see https://github.com/apache/datafusion/blob/bd30fe2375318d13fc5b66e06e53da63ca4d0928/datafusion/physical-optimizer/src/coalesce_batches.rs#L61 We should integrate coalescing directly in the operator ### Describe the solution you'd like 1. Add batch coalescing into HashJoinExec 2. Remove HashJoinExec from the `CoalesceBatches` optimizer rule: https://github.com/apache/datafusion/blob/bd30fe2375318d13fc5b66e06e53da63ca4d0928/datafusion/physical-optimizer/src/coalesce_batches.rs#L61 ### Describe alternatives you've considered @Dandandan did this for `FilterExec` in these two PRs, so we can probably follow a similar model - https://github.com/apache/datafusion/pull/18604 - https://github.com/apache/datafusion/pull/18630 ### Additional context _No response_ -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
