After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit 336fdb3.

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 2026-05-25 00:43:56Z
• Commit Run on test-mac-arm at 2026-05-25 00:43:24Z
• Commit Run on amd64-c6a-4xlarge-linux at 2026-05-24 23:52:16Z
• Commit Run on amd64-m5-4xlarge-linux at 2026-05-25 00:45:02Z

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 test-mac-arm at 2026-05-25 00:43:24Z

  • BitRunReader (C++) with params=0, source=cpp-micro, suite=arrow-bit-util-benchmark
  • BM_UnpackBool (C++) with params=DynamicAligned/1/64, source=cpp-micro, suite=arrow-bpacking-benchmark
  • BM_PlainDecodingDouble (C++) with params=32768, source=cpp-micro, suite=parquet-encoding-benchmark
  • BM_UnpackBool (C++) with params=NeonUnaligned/1/64, source=cpp-micro, suite=arrow-bpacking-benchmark
  • BM_UnpackBool (C++) with params=DynamicUnaligned/1/64, source=cpp-micro, suite=arrow-bpacking-benchmark

• Commit Run on amd64-c6a-4xlarge-linux at 2026-05-24 23:52:16Z

  • BM_UnpackUint64 (C++) with params=ScalarUnaligned/8/128, source=cpp-micro, suite=arrow-bpacking-benchmark
  • TakeChunkedChunkedFSBMonotonicIndices (C++) with params=524288/10/9, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • TakeChunkedChunkedFSBMonotonicIndices (C++) with params=524288/0/9, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • TakeChunkedChunkedFSBMonotonicIndices (C++) with params=524288/1000/9, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark