After merging your PR, Conbench analyzed the 6 benchmarking runs that have been run so far on merge-commit 7d84c1e.

Benchmarks with performance regressions

There were 3 possible performance regressions, according to the lookback z-score method.

Benchmarks with regressions:

• Commit Run on test-mac-arm at 2024-06-14 08:15:21Z
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-01, scale_factor=1
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-03, scale_factor=1
  • tpch (R) with engine=arrow, format=native, language=R, memory_map=False, query_id=TPCH-04, scale_factor=1

All benchmark runs analyzed:

• Commit Run on ec2-t3-xlarge-us-east-2 at 2024-06-14 22:43:09Z
• Commit Run on ursa-thinkcentre-m75q at 2024-06-14 01:10:40Z
• Commit Run on arm64-t4g-linux-compute at 2024-06-14 01:10:39Z
• Commit Run on ec2-m5-4xlarge-us-east-2 at 2024-06-13 22:36:52Z
• Commit Run on arm64-m6g-linux-compute at 2024-06-14 01:10:18Z
• Commit Run on test-mac-arm at 2024-06-14 08:15:21Z

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 ursa-thinkcentre-m75q at 2024-06-14 01:10:40Z

  • RoundArrayBenchmark (C++) with params=<Round, UInt32Type, RoundMode::HALF_TO_ODD>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • RoundDerivativesArrayBenchmark (C++) with params=<Ceil, FloatType>/size:524288/inverse_null_proportion:100, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • RoundArrayBenchmark (C++) with params=<Round, Int64Type, RoundMode::HALF_TOWARDS_ZERO>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • RoundArrayBenchmark (C++) with params=<Round, UInt32Type, RoundMode::DOWN>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • ArrayArrayKernel (C++) with params=<MultiplyChecked, UInt64Type>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • RoundDerivativesArrayBenchmark (C++) with params=<Ceil, FloatType>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • RoundDerivativesArrayBenchmark (C++) with params=<Trunc, DoubleType>/size:524288/inverse_null_proportion:100, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • RoundDerivativesArrayBenchmark (C++) with params=<Trunc, DoubleType>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-round-benchmark
  • ArrayArrayKernel (C++) with params=<MultiplyChecked, Int64Type>/size:524288/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark

• Commit Run on arm64-t4g-linux-compute at 2024-06-14 01:10:39Z

  • ErrorSchemeNoError (C++) with source=cpp-micro, suite=arrow-type-benchmark

• Commit Run on test-mac-arm at 2024-06-14 08:15:21Z

  • ArrayScalarKernel (C++) with params=<AddChecked, UInt64Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • BinaryCompareOp (C++) with source=cpp-micro, suite=arrow-decimal-benchmark
  • ArrayScalarKernel (C++) with params=<AddChecked, UInt16Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • ArrayScalarKernel (C++) with params=<AddChecked, UInt8Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • ArrayArrayKernel (C++) with params=<SubtractChecked, Int64Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • ArrayArrayKernel (C++) with params=<MultiplyChecked, Int64Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • ArrayArrayKernel (C++) with params=<SubtractChecked, Int32Type>/size:4194304/inverse_null_proportion:0, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark
  • ArrayArrayKernel (C++) with params=<SubtractChecked, Int64Type>/size:4194304/inverse_null_proportion:100, source=cpp-micro, suite=arrow-compute-scalar-arithmetic-benchmark