2010YOUY01 opened a new pull request, #25312: URL: https://github.com/apache/datafusion/pull/25312
…ateStream ## Which issue does this PR close? <!-- We generally require a GitHub issue to be filed for all bug fixes and enhancements and this helps us generate change logs for our releases. You can link an issue to this PR using the GitHub syntax. For example `Closes #123` indicates that this PR will close issue #123. --> part of https://github.com/apache/datafusion/issues/25157 ## Rationale for this change <!-- Why are you proposing this change? If this is already explained clearly in the issue then this section is not needed. Explaining clearly why changes are proposed helps reviewers understand your changes and offer better suggestions for fixes. Please explain the problem you are trying to solve in terms of the user-visible behavior, rather than the implementation. For example, "The code in `foo.rs` doesn't handle nulls" is a symptom of the implementation. "COUNT(DISTINCT) returns wrong results when the column contains nulls" is the user-visible problem. --> ### Cause See issue for the target query. The query plan looks like ``` > explain SELECT count(*) FROM ( SELECT DISTINCT d_year, brand, class, cat, manu, cnt, amt FROM src ); +---------------+-------------------------------+ | plan_type | plan | +---------------+-------------------------------+ | physical_plan | ┌───────────────────────────┐ | | | │ ProjectionExec │ | | | │ -------------------- │ | | | │ count(*): │ | | | │ count(Int64(1)) │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ AggregateExec │ | | | │ -------------------- │ | | | │ aggr: count(1) │ | | | │ mode: Final │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ CoalescePartitionsExec │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ AggregateExec │ | | | │ -------------------- │ | | | │ aggr: count(1) │ | | | │ mode: Partial │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ ProjectionExec │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ AggregateExec │ | | | │ -------------------- │ | | | │ group_by: │ | | | │ d_year, brand, class, cat,│ | | | │ manu, cnt, amt │ | | | │ │ | | | │ mode: │ | | | │ FinalPartitioned │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ RepartitionExec │ | | | │ -------------------- │ | | | │ partition_count(in->out): │ | | | │ 14 -> 14 │ | | | │ │ | | | │ partitioning_scheme: │ | | | │ Hash([d_year@0, brand@1, │ | | | │ class@2, cat@3, manu@4 │ | | | │ , cnt@5, amt@6], 14) │ | | | │ │ | | | │ preserve_order: true │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ AggregateExec │ | | | │ -------------------- │ | | | │ group_by: │ | | | │ d_year, brand, class, cat,│ | | | │ manu, cnt, amt │ | | | │ │ | | | │ mode: Partial │ | | | └─────────────┬─────────────┘ | | | ┌─────────────┴─────────────┐ | | | │ DataSourceExec │ | | | │ -------------------- │ | | | │ files: 14 │ | | | │ format: parquet │ | | | └───────────────────────────┘ | | | | +---------------+-------------------------------+ ``` It's slow due to inefficient output materializing in partial and final aggregation TODO: link internal mechanism ### Fix In order to fully bring back the performance, we have to fix: 1. Ordered partial aggregation (this PR) 2. Ordered final aggregation (maybe follow-up PR) 2 is almost the same mechanism as 1, so after this PR is reviewed, we can apply the pattern mechanically afterwards. After PR, this query runs in: (On M4 Pro Macbook Pro) ``` -- Still some gap due to final aggregation is not fixed yet Current main: 3.5s PR: 0.45s DataFusion 54.0: 0.37s ``` ## What changes are included in this PR? <!-- There is no need to duplicate the description in the issue here, but it is sometimes worth providing a summary of the individual changes in this PR. --> 1. Refactor the ordered-partial aggregation, so it's easier to implement incremental outputting with slicing 2. Implement the output materializing strategy mentioned above ## What is the testing strategy for this PR? <!-- We typically require tests for all PRs in order to: 1. Prevent the code from being accidentally broken by subsequent changes 3. Serve as another way to document the expected behavior of the code Briefly describe how this PR is tested, and point to the specific tests you added. For example: 'This new feature is covered by the `sqllogictest` cases added in `foo.slt`'. If this PR does not add tests, explain why. For example, if the change is already covered by existing tests, please mention it. You should also check the `codecov` bot reply on this PR to confirm the changed code is exercised. --> For correctness, existing tests have covered it. To prevent similar perf regression, we can do - https://github.com/apache/datafusion/issues/25310 ## Are there any user-facing changes? <!-- If there are user-facing changes then we may require documentation to be updated before approving the PR. If there are any breaking changes to public APIs, please add the `api change` label. --> -- 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]
