After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit e2d4dbf.

Benchmarks with performance regressions

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

All benchmark runs analyzed:

• Commit Run on ec2-c6a-4xlarge-us-east-2 at 2024-07-25 14:20:08Z
• Commit Run on test-mac-arm at 2024-07-25 14:26:17Z
• Commit Run on arm64-t4g-linux-compute at 2024-07-25 15:49:25Z
• Commit Run on ec2-m5-4xlarge-us-east-2 at 2024-07-25 15:45:08Z

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-linux-compute at 2024-07-25 15:49:25Z
  • ReferenceCompression (C++) with params=Compression::LZ4_FRAME, source=cpp-micro, suite=arrow-compression-benchmark
  • BitBlockCounterFourWordsSum (C++) with params=32768, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterSumWithOffset (C++) with params=4096, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterSum (C++) with params=32768, source=cpp-micro, suite=arrow-bit-block-counter-benchmark
  • BitBlockCounterFourWordsSumWithOffset (C++) with params=65536, source=cpp-micro, suite=arrow-bit-block-counter-benchmark