After merging your PR, Conbench analyzed the 4 benchmarking runs that have been run so far on merge-commit b852b72.

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-linux-compute at 2024-08-07 08:59:05Z
• Commit Run on ec2-c6a-4xlarge-us-east-2 at 2024-08-07 08:43:31Z
• Commit Run on test-mac-arm at 2024-08-07 09:27:55Z
• Commit Run on ec2-m5-4xlarge-us-east-2 at 2024-08-07 10:27:17Z

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-linux-compute at 2024-08-07 08:59:05Z

  • ExecuteScalarExpressionOverhead (C++) with params=ref_only_expression/rows_per_batch:1000/real_time/threads:1, source=cpp-micro, suite=arrow-acero-expression-benchmark

• Commit Run on ec2-m5-4xlarge-us-east-2 at 2024-08-07 10:27:17Z

  • ReadBuffer (C++) with params=num_cols:64/is_partial:0/real_time, source=cpp-micro, suite=arrow-ipc-read-write-benchmark
  • ThreadedTaskGroup (C++) with params=threads:8/task_cost:100000/real_time, source=cpp-micro, suite=arrow-thread-pool-benchmark