xiangfu0 commented on PR #14698: URL: https://github.com/apache/pinot/pull/14698#issuecomment-4158239058
@ankitsultana Yes — benchmarks are included in this PR under `pinot-perf/BenchmarkGroupByCombineOperator`. Results are posted in earlier comments on this PR. Summary: **10K records/segment, 32 segments, 8 threads:** | Groups | numPartitions=1 (ms) | numPartitions=8 (ms) | Speedup | |---:|---:|---:|---:| | 500 | 127.6 | **2.8** | **45.6x** | | 5K | 132.4 | **5.9** | **22.4x** | | 50K | 144.0 | **23.0** | **6.3x** | | 500K | 184.8 | **76.9** | **2.4x** | **100K records/segment (high cardinality):** | Groups | numPartitions=1 (ms) | numPartitions=8 (ms) | Speedup | |---:|---:|---:|---:| | 5M | 1,934 | **637** | **3.0x** | Key optimizations: cached `Key.hashCode()`, `Object2ObjectOpenHashMap` replacing `HashMap`, entry-set merge reusing existing Keys, and parallel partition-table tree reduction. -- 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]
