After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit 19f0652.

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-m5-4xlarge-linux at 2025-01-13 00:07:58Z
• Commit Run on test-mac-arm at 2025-01-13 18:55:55Z
• Commit Run on amd64-c6a-4xlarge-linux at 2025-01-09 17:45:08Z
• Commit Run on arm64-t4g-2xlarge-linux at 2025-01-13 13:53:21Z

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 test-mac-arm at 2025-01-13 18:55:55Z

  • BM_PlainDecodingSpacedByteArray (C++) with params=max-string-length:64/batch-size:2048, source=cpp-micro, suite=parquet-encoding-benchmark

• Commit Run on amd64-c6a-4xlarge-linux at 2025-01-09 17:45:08Z

  • ValueCountsDictionaryChunks (C++) with params=1, source=cpp-micro, suite=arrow-compute-vector-hash-benchmark
  • ValueCountsDictionaryChunks (C++) with params=0, source=cpp-micro, suite=arrow-compute-vector-hash-benchmark
  • BM_DeltaBitPackingEncode_Int64_Fixed (C++) with params=4096, source=cpp-micro, suite=parquet-encoding-benchmark

• Commit Run on arm64-t4g-2xlarge-linux at 2025-01-13 13:53:21Z

  • ExecuteScalarExpressionBaseline (C++) with params=/rows_per_batch:1000000/real_time/threads:1, source=cpp-micro, suite=arrow-acero-expression-benchmark