After merging your PR, Conbench analyzed the 6 benchmarking runs that have been run so far on merge-commit abf4fbf.

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-m6g-linux-compute at 2024-02-09 16:04:20Z
• Commit Run on ec2-m5-4xlarge-us-east-2 at 2024-02-09 14:29:49Z
• Commit Run on arm64-t4g-linux-compute at 2024-02-09 16:20:20Z
• Commit Run on ursa-i9-9960x at 2024-02-09 13:56:53Z
• Commit Run on ec2-t3-xlarge-us-east-2 at 2024-02-09 14:01:34Z
• Commit Run on ursa-thinkcentre-m75q at 2024-02-09 16:11:26Z

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 2024-02-09 16:20:20Z
  • ReadLevels_BitPack (C++) with params=MaxLevel:3/NumLevels:8096/BatchSize:2048/LevelRepeatCount:1, source=cpp-micro, suite=parquet-column-reader-benchmark
  • ReadLevels_BitPack (C++) with params=MaxLevel:1/NumLevels:8096/BatchSize:1024/LevelRepeatCount:7, source=cpp-micro, suite=parquet-column-reader-benchmark
  • ReadLevels_Rle (C++) with params=MaxLevel:3/NumLevels:8096/BatchSize:1024/LevelRepeatCount:1, source=cpp-micro, suite=parquet-column-reader-benchmark