After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit 777797e.

Benchmarks with performance regressions

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

All benchmark runs analyzed:

• Commit Run on amd64-c6a-4xlarge-linux at 2025-01-09 19:45:00Z
• Commit Run on amd64-m5-4xlarge-linux at 2025-01-13 10:18:57Z
• Commit Run on arm64-t4g-2xlarge-linux at 2025-01-13 22:49:17Z
• Commit Run on test-mac-arm at 2025-01-14 08:49:27Z

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 amd64-c6a-4xlarge-linux at 2025-01-09 19:45:00Z

  • FilterFixedSizeBinaryFilterWithNulls (C++) with params=524288/6/8, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • FilterInt64FilterWithNulls (C++) with params=524288/3, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • FilterInt64FilterWithNulls (C++) with params=524288/0, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • FilterFSLInt64FilterWithNulls (C++) with params=524288/0, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • FilterFixedSizeBinaryFilterWithNulls (C++) with params=524288/0/8, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark
  • FilterFixedSizeBinaryFilterWithNulls (C++) with params=524288/3/8, source=cpp-micro, suite=arrow-compute-vector-selection-benchmark

• Commit Run on test-mac-arm at 2025-01-14 08:49:27Z

  • ShortVectorInsert (C++) with params=<STD_VECTOR(std::shared_ptr)>, source=cpp-micro, suite=arrow-small-vector-benchmark
  • SchemaEqualsWithMetadata (C++) with source=cpp-micro, suite=arrow-type-benchmark
  • SchemaEquals (C++) with source=cpp-micro, suite=arrow-type-benchmark