After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit e20ccb6.

Benchmarks with performance regressions

There were 0 possible performance regressions, according to the lookback z-score method.

All benchmark runs analyzed:

• Commit Run on arm64-t4g-2xlarge-linux at 2025-05-21 06:15:52Z
• Commit Run on amd64-c6a-4xlarge-linux at 2025-05-21 03:23:59Z
• Commit Run on amd64-m5-4xlarge-linux at 2025-05-21 04:12:07Z
• Commit Run on test-mac-arm at 2025-05-21 10:59:05Z

Unstable benchmarks with performance regressions

The following benchmark results indicate a possible performance regression, but are known to sometimes produce false positives when applying the lookback z-score analysis.

• Commit Run on arm64-t4g-2xlarge-linux at 2025-05-21 06:15:52Z

  • BitmapReaderSumWithOffset (C++) with params=64, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterSum (C++) with params=512, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterFourWordsSumWithOffset (C++) with params=8, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitmapReaderSum (C++) with params=65536, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterFourWordsSumWithOffset (C++) with params=512, source=cpp-micro, suite=arrow-bit-block-counter-benchmark

• Commit Run on amd64-c6a-4xlarge-linux at 2025-05-21 03:23:59Z

  • BM_DeltaBitPackingEncode_Int32_Narrow (C++) with params=32768, source=cpp-micro, suite=parquet-encoding-benchmark
  • AllocateDeallocate (C++) with params=/size:4096/real_time/threads:1, source=cpp-micro, suite=arrow-memory-pool-benchmark
  • BM_DeltaBitPackingEncode_Int32_Fixed (C++) with params=65536, source=cpp-micro, suite=parquet-encoding-benchmark