After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit ce012eb.

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 2025-02-20 15:14:47Z
• Commit Run on amd64-m5-4xlarge-linux at 2025-02-20 15:27:33Z
• Commit Run on arm64-t4g-2xlarge-linux at 2025-02-20 15:35:39Z
• Commit Run on test-mac-arm at 2025-02-20 15:35:56Z

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-m5-4xlarge-linux at 2025-02-20 15:27:33Z

  • BinaryMathOpAdd128 (C++) with source=cpp-micro, suite=arrow-decimal-benchmark

• Commit Run on arm64-t4g-2xlarge-linux at 2025-02-20 15:35:39Z

  • ExecuteScalarExpressionBaseline (C++) with params=/rows_per_batch:10000/real_time/threads:8, source=cpp-micro, suite=arrow-acero-expression-benchmark
  • FilterOverhead (C++) with params=complex_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time, source=cpp-micro, suite=arrow-acero-filter-benchmark
  • ArrayRangeEqualsStruct (C++) with params=65536/1, source=cpp-micro, suite=arrow-compare-benchmark