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] Changes at 2/20/2022 5:50:14 AM #3642

Closed
performanceautofiler bot opened this issue Feb 22, 2022 · 0 comments
Closed

[Perf] Changes at 2/20/2022 5:50:14 AM #3642

performanceautofiler bot opened this issue Feb 22, 2022 · 0 comments
Labels
arch-x64 branch-refs/heads/main kind-micro os-windows perf-improvement runtime-coreclr untriaged

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Feb 22, 2022 •
edited
Loading

Run Information

Architecture x64
OS Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in System.Globalization.Tests.StringSearch

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
LastIndexOf_Word_NotFound - Duration of single invocation 589.90 ns 521.21 ns 0.88 0.00 True
IndexOf_Word_NotFound - Duration of single invocation 591.05 ns 531.95 ns 0.90 0.01 False

graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Globalization.Tests.StringSearch*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (en-US, IgnoreNonSpace, False))

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 521.2109442224219 < 560.4205185745548.
IsChangePoint: Marked as a change because one of 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 1031.572389043489 (T) = (0 -521.2935797225682) / Math.Sqrt((0.968981687803321 / (299)) + (0.008838864561194204 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.11722955898760258 = (590.5199760933627 - 521.2935797225682) / 590.5199760933627 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

```#### System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, None, False))

```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 531.9495606527445 < 561.4635189556789.
IsChangePoint: Marked as a change because one of 12/22/2021 9:10:55 PM, 12/23/2021 8:24:04 AM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 131.00519454896286 (T) = (0 -530.7475699839523) / Math.Sqrt((7.463937452836511 / (299)) + (1.4442907313037727 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.10611151308787108 = (593.7514329302754 - 530.7475699839523) / 593.7514329302754 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in System.Linq.Tests.Perf_Enumerable

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Skip_One - Duration of single invocation 1.17 μs 1.05 μs 0.89 0.08 False
Sum - Duration of single invocation 492.34 ns 461.33 ns 0.94 0.09 False

graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Linq.Tests.Perf_Enumerable*'

Payloads

Baseline
Compare

Histogram

System.Linq.Tests.Perf_Enumerable.Skip_One(input: IEnumerable)

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 1.0486090006499473 < 1.1161060855172034.
IsChangePoint: Marked as a change because one of 1/21/2022 6:10:04 PM, 1/31/2022 7:28:13 PM, 2/11/2022 1:40:07 AM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 24.68452914561813 (T) = (0 -1048.087131141217) / Math.Sqrt((1921.4114003832087 / (299)) + (0.3475296555638418 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.05654502724501313 = (1110.9031818240285 - 1048.087131141217) / 1110.9031818240285 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Linq.Tests.Perf_Enumerable.Sum(input: IEnumerable)

```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 461.3294303757258 < 482.5597554101367.
IsChangePoint: Marked as a change because one of 1/25/2022 11:23:52 PM, 2/11/2022 3:03:06 PM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 35.79391553349497 (T) = (0 -460.395955718886) / Math.Sqrt((285.8125741193886 / (299)) + (0.20392824807903695 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.07163429270300616 = (495.9208985210831 - 460.395955718886) / 495.9208985210831 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in System.Buffers.Text.Tests.Utf8ParserTests

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TryParseInt32 - Duration of single invocation 5.35 ns 4.26 ns 0.80 0.11 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Buffers.Text.Tests.Utf8ParserTests*'

Payloads

Baseline
Compare

Histogram

System.Buffers.Text.Tests.Utf8ParserTests.TryParseInt32(value: 4)

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 4.261871706048507 < 5.088092937738445.
IsChangePoint: Marked as a change because one of 12/17/2021 7:05:08 PM, 1/4/2022 5:49:50 PM, 1/7/2022 4:24:34 PM, 1/12/2022 9:32:51 PM, 1/21/2022 2:09:07 PM, 1/25/2022 6:54:35 PM, 2/4/2022 6:00:25 PM, 2/7/2022 2:49:00 PM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 23.723488637600543 (T) = (0 -4.246168995670187) / Math.Sqrt((0.3439986994997729 / (299)) + (0.00010635394571398586 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.16019575970169275 = (5.05614141000515 - 4.246168995670187) / 5.05614141000515 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
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
InstanceDelegate - Duration of single invocation 218.40 μs 187.11 μs 0.86 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'PerfLabTests.LowLevelPerf*'

Payloads

Baseline
Compare

Histogram

PerfLabTests.LowLevelPerf.InstanceDelegate

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 187.1066432823129 < 207.37662977430557.
IsChangePoint: Marked as a change because one of 1/19/2022 7:01:24 PM, 1/21/2022 2:09:07 PM, 2/4/2022 6:00:25 PM, 2/8/2022 5:20:12 PM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 47.4080069377205 (T) = (0 -187104.03957897262) / Math.Sqrt((100840849.79379079 / (299)) + (33.552387700207824 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.12827287632686404 = (214636.01911408393 - 187104.03957897262) / 214636.01911408393 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in PerfLabTests.CastingPerf

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
FooObjCastIfIsa - Duration of single invocation 654.66 μs 617.21 μs 0.94 0.32 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'PerfLabTests.CastingPerf*'

Payloads

Baseline
Compare

Histogram

PerfLabTests.CastingPerf.FooObjCastIfIsa

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 617.2069711538462 < 622.1643504809558.
IsChangePoint: Marked as a change because one of 12/27/2021 6:47:12 AM, 2/7/2022 2:49:00 PM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 4.2461120499531475 (T) = (0 -547038.2178833071) / Math.Sqrt((1888015139.1503623 / (299)) + (1092198144.4029448 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.08999794414923393 = (601139.5406924416 - 547038.2178833071) / 601139.5406924416 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in System.Collections.Sort<String>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
LinqQuery - Duration of single invocation 268.01 μs 248.28 μs 0.93 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Collections.Sort&lt;String&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.Sort<String>.LinqQuery(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 248.27809066666669 < 256.24820462619044.
IsChangePoint: Marked as a change because one of 1/4/2022 5:49:50 PM, 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 20.779778682367326 (T) = (0 -250292.43899319725) / Math.Sqrt((141106325.69681528 / (299)) + (2094775.1736762496 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.06795308110955461 = (268540.60017832334 - 250292.43899319725) / 268540.60017832334 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 Windows 10.0.18362
Baseline 79d478ae01d9a3e7548bf41a76d047158ed8a62b
Compare 4017327955f1d8ddc43980eb1848c52fbb131dfc
Diff Diff

Improvements in System.Linq.Tests.Perf_OrderBy

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
OrderByString - Duration of single invocation 284.31 μs 262.63 μs 0.92 0.00 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Linq.Tests.Perf_OrderBy*'

Payloads

Baseline
Compare

Histogram

System.Linq.Tests.Perf_OrderBy.OrderByString(NumberOfPeople: 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 262.62982142857146 < 269.81428105070154.
IsChangePoint: Marked as a change because one of 2/19/2022 11:51:48 PM, 2/22/2022 12:12:47 AM falls between 2/13/2022 12:11:49 PM and 2/22/2022 12:12:47 AM.
IsImprovementStdDev: Marked as improvement because 19.850889687994858 (T) = (0 -265960.92481696076) / Math.Sqrt((11961077.976743821 / (299)) + (3682131.7046357393 / (7))) is greater than 1.9677981407170237 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (7) - 2, .975) and 0.053168181050104735 = (280895.63478330354 - 265960.92481696076) / 280895.63478330354 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-x64 branch-refs/heads/main kind-micro os-windows perf-improvement runtime-coreclr untriaged
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kunalspathak