After merging your PR, Conbench analyzed the 2 benchmarking runs that have been run so far on merge-commit 33d1f32.

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 2025-10-08 09:42:17Z
• Commit Run on test-mac-arm at 2025-10-08 08:38:03Z

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 2025-10-08 08:38:03Z
  • ProjectionOverheadIsolated (C++) with params=ref_only_expression/batch_size:10000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=zero_copy_expression/batch_size:1000000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=ref_only_expression/batch_size:1000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=complex_expression/batch_size:100000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=simple_expression/batch_size:10000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=zero_copy_expression/batch_size:10000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark
  • ProjectionOverheadIsolated (C++) with params=zero_copy_expression/batch_size:100000/real_time, source=cpp-micro, suite=arrow-acero-project-benchmark