[Perf] Windows/arm64: 2 Improvements on 3/26/2023 1:21:26 PM #14620
Description
Run Information
| Name | Value |
|---|---|
| Architecture | arm64 |
| OS | Windows 10.0.19041 |
| Queue | SurfaceWindows |
| Baseline | 95dbfc2f9cab66a9d5de834df5f25e8c3a71fc38 |
| Compare | 49a2919f66c79f33033659f9722eec24eacc3b22 |
| Diff | Diff |
| Configs | CompilationMode:tiered, RunKind:micro |
Improvements in Benchstone.BenchF.NewtR
| Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio | Baseline ETL | Compare ETL |
|---|---|---|---|---|---|---|---|---|---|---|
| Test - Duration of single invocation | 178.65 ms | 122.72 ms | 0.69 | 0.01 | True |
Repro
General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md
Payloads
git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'Benchstone.BenchF.NewtR*'Payloads
Histogram
Benchstone.BenchF.NewtR.Test
Description of detection logic
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 122.72265416666667 < 169.61595596153845.
IsChangePoint: Marked as a change because one of 3/26/2023 5:20:11 AM, 3/30/2023 4:24:24 AM falls between 3/21/2023 2:53:21 PM and 3/30/2023 4:24:24 AM.
IsImprovementStdDev: Marked as improvement because 433.85121901043794 (T) = (0 -122755875.14880954) / Math.Sqrt((4782110475855.205 / (299)) + (4644631258.355605 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.3114560983435548 = (178283294.43263245 - 122755875.14880954) / 178283294.43263245 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.
Docs
Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository
Run Information
| Name | Value |
|---|---|
| Architecture | arm64 |
| OS | Windows 10.0.19041 |
| Queue | SurfaceWindows |
| Baseline | 95dbfc2f9cab66a9d5de834df5f25e8c3a71fc38 |
| Compare | 49a2919f66c79f33033659f9722eec24eacc3b22 |
| Diff | Diff |
| Configs | CompilationMode:tiered, RunKind:micro |
Improvements in System.Collections.Sort<IntStruct>
| Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio | Baseline ETL | Compare ETL |
|---|---|---|---|---|---|---|---|---|---|---|
| Array_Comparison - Duration of single invocation | 23.16 μs | 21.19 μs | 0.92 | 0.29 | False |
Repro
General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md
Payloads
git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.Sort<IntStruct>*'Payloads
Histogram
System.Collections.Sort<IntStruct>.Array_Comparison(Size: 512)
Description of detection logic
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 21.193637142857135 < 22.10774841483516.
IsChangePoint: Marked as a change because one of 2/21/2023 5:46:41 AM, 2/25/2023 3:45:56 AM, 3/26/2023 5:20:11 AM, 3/30/2023 4:24:24 AM falls between 3/21/2023 2:53:21 PM and 3/30/2023 4:24:24 AM.
IsImprovementStdDev: Marked as improvement because 3.823669622865817 (T) = (0 -21619.211821733825) / Math.Sqrt((3398101.0223425794 / (299)) + (1598823.8475413516 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.06301685705634441 = (23073.21319977458 - 21619.211821733825) / 23073.21319977458 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.
Docs
Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository