Yicong-Huang opened a new pull request, #55603: URL: https://github.com/apache/spark/pull/55603
### What changes were proposed in this pull request? Add ASV micro-benchmarks for `SQL_WINDOW_AGG_PANDAS_UDF` to `python/benchmarks/bench_eval_type.py`, mirroring the existing `SQL_WINDOW_AGG_ARROW_UDF` benchmarks added in [SPARK-56120](https://issues.apache.org/jira/browse/SPARK-56120). The new mixin `_WindowAggPandasBenchMixin` reuses the same scenario shapes and UDF set as the Arrow variant (sum, mean_multi) and writes the worker payload with `runner_conf={"window_bound_types": "unbounded"}`. ### Why are the changes needed? Part of [SPARK-55724](https://issues.apache.org/jira/browse/SPARK-55724). Establishes performance baselines before refactoring this eval type. ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? `./python/asv check --python=same` passes. Local quick run on `WindowAggPandasUDFTimeBench` and `WindowAggPandasUDFPeakmemBench`: ```text bench_eval_type.WindowAggPandasUDFTimeBench.time_worker ================ ========== ================ -- udf ---------------- --------------------------- scenario sum_udf mean_multi_udf ================ ========== ================ few_groups_sm 74.8 ms 75.3 ms few_groups_lg 115 ms 125 ms many_groups_sm 1.92 s 2.13 s many_groups_lg 578 ms 618 ms wide_cols 509 ms 539 ms ================ ========== ================ bench_eval_type.WindowAggPandasUDFPeakmemBench.peakmem_worker ================ ========= ================ -- udf ---------------- -------------------------- scenario sum_udf mean_multi_udf ================ ========= ================ few_groups_sm 471M 470M few_groups_lg 535M 533M many_groups_sm 501M 499M many_groups_lg 606M 604M wide_cols 587M 585M ================ ========= ================ ``` ### Was this patch authored or co-authored using generative AI tooling? No. -- 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]
