andrewlock
commented
3 months ago Execution-Time Benchmarks Report :stopwatch:
Execution-time results for samples comparing the following branches/commits:
Execution-time benchmarks measure the whole time it takes to execute a program. And are intended to measure the one-off costs. Cases where the execution time results for the PR are worse than latest master results are shown in red. The following thresholds were used for comparing the execution times:
- Welch test with statistical test for significance of 5%
- Only results indicating a difference greater than 5% and 5 ms are considered.
Note that these results are based on a single point-in-time result for each branch. For full results, see the dashboard.
Graphs show the p99 interval based on the mean and StdDev of the test run, as well as the mean value of the run (shown as a diamond below the graph).
gantt
title Execution time (ms) FakeDbCommand (.NET Framework 4.6.2)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (74ms) : 65, 82
. : milestone, 74,
master - mean (73ms) : 64, 81
. : milestone, 73,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (902ms) : 874, 930
. : milestone, 902,
master - mean (896ms) : 868, 924
. : milestone, 896,
gantt
title Execution time (ms) FakeDbCommand (.NET Core 3.1)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (110ms) : 106, 114
. : milestone, 110,
master - mean (109ms) : 106, 112
. : milestone, 109,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (634ms) : 618, 651
. : milestone, 634,
master - mean (632ms) : 617, 648
. : milestone, 632,
gantt
title Execution time (ms) FakeDbCommand (.NET 6)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (95ms) : 92, 99
. : milestone, 95,
master - mean (93ms) : 89, 97
. : milestone, 93,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (593ms) : 577, 609
. : milestone, 593,
master - mean (592ms) : 578, 606
. : milestone, 592,
gantt
title Execution time (ms) HttpMessageHandler (.NET Framework 4.6.2)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (191ms) : 188, 194
. : milestone, 191,
master - mean (191ms) : 188, 195
. : milestone, 191,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (1,004ms) : 976, 1032
. : milestone, 1004,
master - mean (1,006ms) : 976, 1036
. : milestone, 1006,
gantt
title Execution time (ms) HttpMessageHandler (.NET Core 3.1)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (276ms) : 272, 281
. : milestone, 276,
master - mean (277ms) : 272, 281
. : milestone, 277,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (823ms) : 792, 855
. : milestone, 823,
master - mean (817ms) : 797, 836
. : milestone, 817,
gantt
title Execution time (ms) HttpMessageHandler (.NET 6)
dateFormat X
axisFormat %s
todayMarker off
section Baseline
This PR (5756) - mean (264ms) : 260, 268
. : milestone, 264,
master - mean (265ms) : 261, 269
. : milestone, 265,
section CallTarget+Inlining+NGEN
This PR (5756) - mean (811ms) : 786, 835
. : milestone, 811,
master - mean (805ms) : 778, 832
. : milestone, 805,
datadog-ddstaging[bot]
Summary of changes
Adds a new trait,
SkipInCIthat can be used to skip tests in CIReason for change
We currently use the built-in
[Fact(Skip="..."])or[Theory(Skip="..."])to skip tests that we don't want to run in CI. The downside to this is if you want to run a test locally, you have to explicitly change the code to remove theSkip, run the test, and remember to add it back later.Instead, we can change the default
Filterused by Nuke for each stage to beSkipInCI!=True. That way the tests are still excluded from CI (or if you run the "default" test suite locally), but you can still easily run the test without making code modifications if you want to run it directly from VS/Rider, or usingnuke RunWindowsIntegrationTests -Filter SomeSkippedTest(for example).Implementation details
Replaced existing usages of
with
and changed the default filters used by Nuke to exclude tests with this trait
Test coverage
This PR is the test - if it passes, we're good.
Other details
It would be nice to have the skip reason as the value of the trait, but the dotnet test syntax doesn't make it easy to select just based on the key, it's Key=Value, so we're stuck with comments.