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

Regressions in System.Text.Json.Tests #83398

Closed
performanceautofiler bot opened this issue Mar 14, 2023 · 6 comments
Closed

Regressions in System.Text.Json.Tests #83398

performanceautofiler bot opened this issue Mar 14, 2023 · 6 comments
Labels
arch-x64 area-System.Text.Json tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark
Milestone
8.0.0

Comments

@performanceautofiler
Copy link

Run Information

Architecture x64
OS Windows 10.0.19042
Baseline 0e1ede5588b9d464afc02e2c485da5507fe9071c
Compare 633dcab4b1d665dc355d0acd5ca6c6ba8a4ae059
Diff Diff

Regressions in System.Text.Json.Tests.Perf_Segment

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ReadMultiSegmentSequenceUsingSpan - Duration of single invocation 7.40 μs 8.24 μs 1.11 0.20 False
ReadMultiSegmentSequenceUsingSpan - Duration of single invocation 7.69 μs 8.15 μs 1.06 0.10 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Payloads

Baseline
Compare

Histogram

System.Text.Json.Tests.Perf_Segment.ReadMultiSegmentSequenceUsingSpan(segmentSize: 8192, TestCase: Json4KB)

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 8.238742101105846 > 7.844099232796462.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -21.176485353927564 (T) = (0 -8189.349245004544) / Math.Sqrt((27300.513827141483 / (36)) + (22024.615974794844 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.11839759960945816 = (7322.39522676394 - 8189.349245004544) / 7322.39522676394 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Segment.ReadMultiSegmentSequenceUsingSpan(segmentSize: 4096, TestCase: Json4KB)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 8.150794667893445 > 8.060214164494914.
IsChangePoint: Marked as a change because one of 2/9/2023 12:40:40 PM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -18.08701717994196 (T) = (0 -8150.621838382612) / Math.Sqrt((19684.88524199473 / (36)) + (1759.935394649776 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.05849338864052584 = (7700.210436695168 - 8150.621838382612) / 7700.210436695168 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
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 10.0.19042
Baseline 0e1ede5588b9d464afc02e2c485da5507fe9071c
Compare 633dcab4b1d665dc355d0acd5ca6c6ba8a4ae059
Diff Diff

Regressions in System.Text.Json.Tests.Perf_Reader

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ReadSingleSpanSequenceEmptyLoop - Duration of single invocation 4.69 μs 5.25 μs 1.12 0.04 False
ReadSingleSpanSequenceEmptyLoop - Duration of single invocation 98.79 ns 106.40 ns 1.08 0.04 False
ReadSpanEmptyLoop - Duration of single invocation 4.85 μs 5.26 μs 1.08 0.03 False

graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Payloads

Baseline
Compare

Histogram

System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: Json4KB)

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 5.252997863605898 > 4.906738088649842.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 2/24/2023 8:06:06 PM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -146.28746210083457 (T) = (0 -5255.148584600376) / Math.Sqrt((322.637021532014 / (36)) + (168.96872607262335 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.12531081614425058 = (4669.9529669558715 - 5255.148584600376) / 4669.9529669558715 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: HelloWorld)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 106.39897869024679 > 103.37795450954697.
IsChangePoint: Marked as a change because one of 2/6/2023 2:52:08 AM, 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -11.562173602895873 (T) = (0 -109.42132715238068) / Math.Sqrt((6.063515718708076 / (36)) + (16.017112606037244 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.10693328642763417 = (98.8508779110909 - 109.42132715238068) / 98.8508779110909 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: Json4KB)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 5.260348622311827 > 5.094311510211622.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -77.8848774479965 (T) = (0 -5236.486554493994) / Math.Sqrt((525.5217925025796 / (36)) + (223.46512762571095 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.07843047918728374 = (4855.6551910886865 - 5236.486554493994) / 4855.6551910886865 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository
@performanceautofiler performanceautofiler bot added the untriaged New issue has not been triaged by the area owner label Mar 14, 2023
@EgorBo EgorBo removed the untriaged New issue has not been triaged by the area owner label Mar 14, 2023
@EgorBo EgorBo changed the title [Perf] Windows/x64: 5 Regressions on 3/8/2023 4:46:51 PM Regressions in System.Text.Json.Tests Mar 14, 2023
@EgorBo EgorBo transferred this issue from dotnet/perf-autofiling-issues Mar 14, 2023
@dotnet-issue-labeler
Copy link

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

@ghost ghost added the untriaged New issue has not been triaged by the area owner label Mar 14, 2023
@EgorBo
Copy link
Member

EgorBo commented Mar 14, 2023

Likely regressed in #82789 cc @stephentoub

@EgorBo EgorBo added tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark arch-x64 labels Mar 14, 2023
@stephentoub
Copy link
Member

Are there corresponding improvements in other benchmarks? What we saw in local testing was a few things regressed but the majority improved.
cc: @MihaZupan

@EgorBo
Copy link
Member

EgorBo commented Mar 14, 2023

Are there corresponding improvements in other benchmarks? What we saw in local testing was a few things regressed but the majority improved. cc: @MihaZupan

From what I see: dotnet/perf-autofiling-issues#14008 and dotnet/perf-autofiling-issues#13938

@ghost
Copy link

ghost commented Mar 15, 2023

Tagging subscribers to this area: @dotnet/area-system-text-json, @gregsdennis
See info in area-owners.md if you want to be subscribed.

Issue Details

Run Information

Architecture x64
OS Windows 10.0.19042
Baseline 0e1ede5588b9d464afc02e2c485da5507fe9071c
Compare 633dcab4b1d665dc355d0acd5ca6c6ba8a4ae059
Diff Diff

Regressions in System.Text.Json.Tests.Perf_Segment

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ReadMultiSegmentSequenceUsingSpan - Duration of single invocation 7.40 μs 8.24 μs 1.11 0.20 False
ReadMultiSegmentSequenceUsingSpan - Duration of single invocation 7.69 μs 8.15 μs 1.06 0.10 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Payloads

Baseline
Compare

Histogram

System.Text.Json.Tests.Perf_Segment.ReadMultiSegmentSequenceUsingSpan(segmentSize: 8192, TestCase: Json4KB)

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 8.238742101105846 > 7.844099232796462.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -21.176485353927564 (T) = (0 -8189.349245004544) / Math.Sqrt((27300.513827141483 / (36)) + (22024.615974794844 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.11839759960945816 = (7322.39522676394 - 8189.349245004544) / 7322.39522676394 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Segment.ReadMultiSegmentSequenceUsingSpan(segmentSize: 4096, TestCase: Json4KB)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 8.150794667893445 > 8.060214164494914.
IsChangePoint: Marked as a change because one of 2/9/2023 12:40:40 PM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -18.08701717994196 (T) = (0 -8150.621838382612) / Math.Sqrt((19684.88524199473 / (36)) + (1759.935394649776 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.05849338864052584 = (7700.210436695168 - 8150.621838382612) / 7700.210436695168 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
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 10.0.19042
Baseline 0e1ede5588b9d464afc02e2c485da5507fe9071c
Compare 633dcab4b1d665dc355d0acd5ca6c6ba8a4ae059
Diff Diff

Regressions in System.Text.Json.Tests.Perf_Reader

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ReadSingleSpanSequenceEmptyLoop - Duration of single invocation 4.69 μs 5.25 μs 1.12 0.04 False
ReadSingleSpanSequenceEmptyLoop - Duration of single invocation 98.79 ns 106.40 ns 1.08 0.04 False
ReadSpanEmptyLoop - Duration of single invocation 4.85 μs 5.26 μs 1.08 0.03 False

graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Payloads

Baseline
Compare

Histogram

System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: Json4KB)

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 5.252997863605898 > 4.906738088649842.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 2/24/2023 8:06:06 PM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -146.28746210083457 (T) = (0 -5255.148584600376) / Math.Sqrt((322.637021532014 / (36)) + (168.96872607262335 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.12531081614425058 = (4669.9529669558715 - 5255.148584600376) / 4669.9529669558715 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: HelloWorld)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 106.39897869024679 > 103.37795450954697.
IsChangePoint: Marked as a change because one of 2/6/2023 2:52:08 AM, 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -11.562173602895873 (T) = (0 -109.42132715238068) / Math.Sqrt((6.063515718708076 / (36)) + (16.017112606037244 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.10693328642763417 = (98.8508779110909 - 109.42132715238068) / 98.8508779110909 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

```#### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: Json4KB)

```log

Description of detection logic

IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 5.260348622311827 > 5.094311510211622.
IsChangePoint: Marked as a change because one of 2/8/2023 6:12:15 AM, 3/8/2023 10:34:21 AM, 3/14/2023 5:57:45 AM falls between 3/5/2023 5:50:49 PM and 3/14/2023 5:57:45 AM.
IsRegressionStdDev: Marked as regression because -77.8848774479965 (T) = (0 -5236.486554493994) / Math.Sqrt((525.5217925025796 / (36)) + (223.46512762571095 / (24))) is less than -2.001717484144427 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (36) + (24) - 2, .025) and -0.07843047918728374 = (4855.6551910886865 - 5236.486554493994) / 4855.6551910886865 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

Docs

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

Author: performanceautofiler[bot]
Assignees: -
Labels:

area-System.Text.Json, tenet-performance, tenet-performance-benchmarks, arch-x64, untriaged
Milestone: -

@eiriktsarpalis eiriktsarpalis removed the untriaged New issue has not been triaged by the area owner label Mar 22, 2023
@eiriktsarpalis eiriktsarpalis added this to the 8.0.0 milestone Mar 22, 2023
@eiriktsarpalis
Copy link
Member

All reported Utf8JsonReader benchmarks have improved numbers since the issue was reported.

@eiriktsarpalis eiriktsarpalis mentioned this issue Jul 25, 2023
Closed
@ghost ghost locked as resolved and limited conversation to collaborators Aug 24, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
arch-x64 area-System.Text.Json tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark
Projects
None yet
Milestone
8.0.0
Development

No branches or pull requests
5 participants
@EgorBo @stephentoub @eiriktsarpalis @jeffschwMSFT @AndyAyersMS