Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Perf] Linux/x64: 14 Improvements on 11/15/2022 10:02:59 PM #9976

Open
performanceautofiler bot opened this issue Nov 22, 2022 · 1 comment
Open

[Perf] Linux/x64: 14 Improvements on 11/15/2022 10:02:59 PM #9976

performanceautofiler bot opened this issue Nov 22, 2022 · 1 comment
Labels
arch-wasm arch-x64 branch-refs/heads/main kind-micro os-linux perf-improvement untriaged

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Nov 22, 2022 •
edited
Loading

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Numerics.Tests.Perf_Vector2

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CreateFromScalarXYBenchmark - Duration of single invocation 20.10 ns 16.80 ns 0.84 0.39 False
UnitXBenchmark - Duration of single invocation 19.60 ns 17.50 ns 0.89 0.34 False
UnitYBenchmark - Duration of single invocation 19.75 ns 16.97 ns 0.86 0.38 False
CreateFromScalar - Duration of single invocation 18.40 ns 16.43 ns 0.89 0.40 False

graph
graph
graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector2*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Vector2.CreateFromScalarXYBenchmark

Description of detection logic

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 16.796087664532187 < 18.995665350286984.
IsChangePoint: Marked as a change because one of 10/19/2022 7:27:15 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 7.324815433919425 (T) = (0 -17.253166749745326) / Math.Sqrt((4.899673487930778 / (299)) + (0.7263571383176524 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.08507239352561077 = (18.857411917243617 - 17.253166749745326) / 18.857411917243617 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Numerics.Tests.Perf_Vector2.UnitXBenchmark

```log

Description of detection logic

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 17.495542777847938 < 18.27586278886116.
IsChangePoint: Marked as a change because one of 10/19/2022 7:27:15 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 8.1031657090824 (T) = (0 -17.13316980526051) / Math.Sqrt((4.078549583868242 / (299)) + (0.8466399482118598 / (24))) is greater than 1.967381707010986 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (24) - 2, .975) and 0.09469817898754755 = (18.925367659262495 - 17.13316980526051) / 18.925367659262495 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Numerics.Tests.Perf_Vector2.UnitYBenchmark

```log

Description of detection logic

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 16.968592324750414 < 18.577099128279695.
IsChangePoint: Marked as a change because one of 10/19/2022 1:55:59 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 5.394290275247977 (T) = (0 -17.52596139529805) / Math.Sqrt((2.6351775207308745 / (299)) + (0.9652366169624401 / (24))) is greater than 1.967381707010986 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (24) - 2, .975) and 0.06380522482855557 = (18.720422138746223 - 17.52596139529805) / 18.720422138746223 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Numerics.Tests.Perf_Vector2.CreateFromScalar

```log

Description of detection logic

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 16.431868472630054 < 17.426093253660422.
IsChangePoint: Marked as a change because one of 10/19/2022 7:27:15 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 6.722647425123183 (T) = (0 -15.80891259469089) / Math.Sqrt((4.0197239104895495 / (299)) + (0.7764695074621232 / (25))) is greater than 1.9673585853226652 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (25) - 2, .975) and 0.08232300757193393 = (17.22709921370302 - 15.80891259469089) / 17.22709921370302 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

### Run Information
Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Memory.Constructors<Byte>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
MemoryMarshalCreateReadOnlySpan - Duration of single invocation 19.37 ns 17.19 ns 0.89 0.26 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Memory.Constructors&lt;Byte&gt;*'

Payloads

Baseline
Compare

Histogram

System.Memory.Constructors<Byte>.MemoryMarshalCreateReadOnlySpan

Description of detection logic

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 17.187669048821665 < 18.420203709343582.
IsChangePoint: Marked as a change because one of 9/25/2022 5:06:47 PM, 9/26/2022 2:03:14 AM, 10/19/2022 7:27:15 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 6.40471282139619 (T) = (0 -18.09276817673842) / Math.Sqrt((2.448851898637405 / (299)) + (0.40038981590790246 / (25))) is greater than 1.9673585853226652 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (25) - 2, .975) and 0.052199995587193924 = (19.0892256726117 - 18.09276817673842) / 19.0892256726117 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Numerics.Tests.Perf_Vector3

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CrossBenchmark - Duration of single invocation 77.93 ns 66.25 ns 0.85 0.13 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector3*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Vector3.CrossBenchmark

Description of detection logic

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 66.25127372136053 < 80.82137761135793.
IsChangePoint: Marked as a change because one of 10/11/2022 6:23:40 AM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 18.504300201810313 (T) = (0 -68.5521289666358) / Math.Sqrt((23.424422799884912 / (299)) + (13.372026632302505 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.17982260369291655 = (83.58207538429788 - 68.5521289666358) / 83.58207538429788 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Float

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsAllBenchmark - Duration of single invocation 108.28 ns 97.51 ns 0.90 0.18 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Float*'

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Float.EqualsAllBenchmark

Description of detection logic

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 97.50836621777867 < 99.62614871810203.
IsChangePoint: Marked as a change because one of 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 6.315694867295252 (T) = (0 -97.3722681647104) / Math.Sqrt((23.60279311329894 / (28)) + (0.5376297475073795 / (25))) is greater than 2.0075837703147283 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (28) + (25) - 2, .975) and 0.05687582457272504 = (103.24437725350076 - 97.3722681647104) / 103.24437725350076 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in StoreBlock.AnyLocation

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CopyBlock16 - Duration of single invocation 22.16 ns 20.75 ns 0.94 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'StoreBlock.AnyLocation*'

Payloads

Baseline
Compare

Histogram

StoreBlock.AnyLocation.CopyBlock16

Description of detection logic

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 20.751004197190074 < 21.055053578735063.
IsChangePoint: Marked as a change because one of 8/26/2022 11:36:04 AM, 9/19/2022 6:07:30 AM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 17.146808754701308 (T) = (0 -20.869912048690313) / Math.Sqrt((0.6103755790733023 / (299)) + (0.08059914867256285 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.0576567330762018 = (22.146825664514395 - 20.869912048690313) / 22.146825664514395 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Collections.IterateForEach<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
SortedList - Duration of single invocation 80.45 μs 73.33 μs 0.91 0.08 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.IterateForEach&lt;Int32&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.IterateForEach<Int32>.SortedList(Size: 512)

Description of detection logic

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 73.33027326458752 < 73.96191619649424.
IsChangePoint: Marked as a change because one of 10/11/2022 6:23:40 AM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 26.885345827945425 (T) = (0 -73204.33925408739) / Math.Sqrt((5626040.674998625 / (299)) + (644821.6982599699 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.07364133395482182 = (79023.7539058303 - 73204.33925408739) / 79023.7539058303 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Tests.Perf_TimeSpan

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
FromSeconds - Duration of single invocation 118.18 ns 109.87 ns 0.93 0.05 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_TimeSpan*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_TimeSpan.FromSeconds

Description of detection logic

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 109.86860621310265 < 112.05978406717105.
IsChangePoint: Marked as a change because one of 10/19/2022 4:51:05 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 14.585528693013456 (T) = (0 -111.15786617279109) / Math.Sqrt((15.886068244370602 / (299)) + (3.786083801896247 / (25))) is greater than 1.9673585853226652 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (25) - 2, .975) and 0.05602327528152603 = (117.75488024447017 - 111.15786617279109) / 117.75488024447017 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Numerics.Tests.Perf_Matrix3x2

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
IdentityBenchmark - Duration of single invocation 17.07 ns 13.77 ns 0.81 0.44 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix3x2*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Matrix3x2.IdentityBenchmark

Description of detection logic

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 13.772686446594504 < 15.979209914686788.
IsChangePoint: Marked as a change because one of 9/30/2022 9:31:11 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 5.0662140747869175 (T) = (0 -15.02044110081482) / Math.Sqrt((4.365517150633143 / (299)) + (1.1404008180166072 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.07872292850766899 = (16.30393457690633 - 15.02044110081482) / 16.30393457690633 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt64>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsAllBenchmark - Duration of single invocation 102.68 ns 95.31 ns 0.93 0.12 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt64&gt;*'

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt64>.EqualsAllBenchmark

Description of detection logic

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 95.30788485831138 < 95.35480723633012.
IsChangePoint: Marked as a change because one of 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 8.751926897805975 (T) = (0 -94.33082693281739) / Math.Sqrt((8.38886941085687 / (30)) + (1.0837501238881522 / (25))) is greater than 2.0057459953168344 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (30) + (25) - 2, .975) and 0.05008656003001263 = (99.30465552292615 - 94.33082693281739) / 99.30465552292615 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsAllBenchmark - Duration of single invocation 103.55 ns 93.67 ns 0.90 0.25 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&gt;*'

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>.EqualsAllBenchmark

Description of detection logic

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 93.67064518020932 < 95.75129900249874.
IsChangePoint: Marked as a change because one of 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 5.781342111031923 (T) = (0 -94.47989471436102) / Math.Sqrt((25.10425707959289 / (28)) + (1.2200662554437012 / (25))) is greater than 2.0075837703147283 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (28) + (25) - 2, .975) and 0.056155667665502965 = (100.10114112851132 - 94.47989471436102) / 100.10114112851132 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

Run Information

Architecture x64
OS ubuntu 18.04
Baseline dd871a350bff581b6d313e88b8f29043a9772d18
Compare 715fde2b08357dfb72248f18ab09dd27cf7f2888
Diff Diff

Improvements in PerfLabTests.LowLevelPerf

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EmptyStaticFunction5Arg - Duration of single invocation 36.31 ms 29.96 ms 0.82 0.10 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'PerfLabTests.LowLevelPerf*'

Payloads

Baseline
Compare

Histogram

PerfLabTests.LowLevelPerf.EmptyStaticFunction5Arg

Description of detection logic

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 29.955480798076923 < 35.085576437619046.
IsChangePoint: Marked as a change because one of 9/23/2022 2:49:56 PM, 9/28/2022 6:40:13 PM, 10/11/2022 6:23:40 AM, 11/11/2022 3:15:52 PM, 11/15/2022 8:48:03 PM, 11/22/2022 6:55:51 AM falls between 11/13/2022 1:28:42 PM and 11/22/2022 6:55:51 AM.
IsImprovementStdDev: Marked as improvement because 22.165386942399877 (T) = (0 -30154529.217974357) / Math.Sqrt((2492174807196.159 / (299)) + (105265122795.62495 / (25))) is greater than 1.9673585853226652 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (25) - 2, .975) and 0.07606915657778247 = (32637214.606109165 - 30154529.217974357) / 32637214.606109165 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
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
arch-wasm arch-x64 branch-refs/heads/main kind-micro os-linux perf-improvement untriaged
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@lewing