[Perf] Windows/arm64: 2 Improvements on 9/1/2023 1:30:05 PM

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	Windows 10.0.19041
Queue	SurfaceWindows
Baseline	f0463a98d105f26037f9d3e63213421a3a7d4dff
Compare	c418a28cf5685685c5433a56f52156a702ec905c
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in System.IO.Tests.Perf_FileStream

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
WriteAsync - Duration of single invocation 📝 - Benchmark Source 📈 - ADX Test Multi Config Graph	7.30 ms	6.67 ms	0.91	0.17	False

Test Report

Repro

General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.IO.Tests.Perf_FileStream*'

Payloads

Baseline
Compare

System.IO.Tests.Perf_FileStream.WriteAsync(fileSize: 1048576, userBufferSize: 4096, options: Asynchronous)

ETL Files

Histogram

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 6.672725147928994 < 6.9372993645937395.
IsChangePoint: Marked as a change because one of 6/3/2023 12:39:03 AM, 8/31/2023 11:52:29 PM, 9/5/2023 8:21:55 PM falls between 8/27/2023 2:42:29 PM and 9/5/2023 8:21:55 PM.
IsImprovementStdDev: Marked as improvement because 9.863402446033994 (T) = (0 -6715323.898121896) / Math.Sqrt((85415121870.76007 / (299)) + (7200282977.440442 / (6))) is greater than 1.967824097752068 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (6) - 2, .975) and 0.053581045888821247 = (7095508.673987341 - 6715323.898121896) / 7095508.673987341 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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	f0463a98d105f26037f9d3e63213421a3a7d4dff
Compare	c418a28cf5685685c5433a56f52156a702ec905c
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in PerfLabTests.LowLevelPerf

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
GenericClassWithSTringGenericInstanceMethod - Duration of single invocation 📝 - Benchmark Source 📈 - ADX Test Multi Config Graph	222.80 μs	191.11 μs	0.86	0.11	False

Test Report

Repro

General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'PerfLabTests.LowLevelPerf*'

Payloads

Baseline
Compare

PerfLabTests.LowLevelPerf.GenericClassWithSTringGenericInstanceMethod

ETL Files

Histogram

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 191.11270520637896 < 211.70149672833193.
IsChangePoint: Marked as a change because one of 4/8/2023 1:46:22 AM, 4/8/2023 9:16:19 PM, 8/31/2023 11:52:29 PM, 9/5/2023 8:21:55 PM falls between 8/27/2023 2:42:29 PM and 9/5/2023 8:21:55 PM.
IsImprovementStdDev: Marked as improvement because 23.352331121359317 (T) = (0 -191035.8960515687) / Math.Sqrt((824885509.0559303 / (299)) + (1970.7758519515355 / (6))) is greater than 1.967824097752068 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (6) - 2, .975) and 0.1687791461613137 = (229825.67769966312 - 191035.8960515687) / 229825.67769966312 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository

[EgorBo]

Could be dotnet/runtime#91342 or noise