After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit 93377af.

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 2025-07-11 08:48:33Z
• Commit Run on amd64-c6a-4xlarge-linux at 2025-07-11 03:38:03Z
• Commit Run on amd64-m5-4xlarge-linux at 2025-07-11 03:54:54Z
• Commit Run on test-mac-arm at 2025-07-11 13:41:01Z

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-2xlarge-linux at 2025-07-11 08:48:33Z

  • BM_HashJoinBasic_KeyTypes (C++) with params="{int32}"/Hashtable krows:1, source=cpp-micro, suite=arrow-acero-hash-join-benchmark
  • BM_HashJoinBasic_ProbeParallelism (C++) with params=Threads:1/HashTable krows:1/process_time, source=cpp-micro, suite=arrow-acero-hash-join-benchmark
  • UniqueUInt8 (C++) with params=3, source=cpp-micro, suite=arrow-compute-vector-hash-benchmark

• Commit Run on amd64-c6a-4xlarge-linux at 2025-07-11 03:38:03Z

  • BenchmarkTemporalBinary (C++) with params=<WeeksBetween, zoned>/524288/100, source=cpp-micro, suite=arrow-compute-scalar-temporal-benchmark
  • BenchmarkTemporalBinary (C++) with params=<MonthDayNanoBetween, zoned>/524288/100, source=cpp-micro, suite=arrow-compute-scalar-temporal-benchmark