MQE: improve runtime of mixed metrics tests #10678
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ime over multiple series at a time
…rs, and only run with two series rather than up to four
charleskorn
commented
Feb 18, 2025
Comment on lines
-2850
to
2852
| // Generate combinations of 2, 3, and 4 labels. (e.g., "a,b", "e,f", "c,d,e", "a,b,c,d", "c,d,e,f" etc) | ||
| // Generate combinations of 2 labels. (e.g., "a,b", "e,f" etc) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 2)...) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 3)...) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 4)...) | ||
|
|
There was a problem hiding this comment.
Rationale: these functions all operate a series at a time, and any issue working with multiple series is likely to be surfaced running with two series.
There was a problem hiding this comment.
that's fair for non aggregations
charleskorn
commented
Feb 18, 2025
| @@ -2981,8 +2977,6 @@ func TestCompareVariousMixedMetricsVectorSelectors(t *testing.T) { | |||
| for _, function := range []string{"rate", "increase", "changes", "resets", "deriv", "irate", "idelta", "delta", "deriv", "stddev_over_time", "stdvar_over_time"} { | |||
| expressions = append(expressions, fmt.Sprintf(`%s(series{label=~"(%s)"}[45s])`, function, labelRegex)) | |||
| expressions = append(expressions, fmt.Sprintf(`%s(series{label=~"(%s)"}[1m])`, function, labelRegex)) | |||
| expressions = append(expressions, fmt.Sprintf(`sum(%s(series{label=~"(%s)"}[2m15s]))`, function, labelRegex)) | |||
There was a problem hiding this comment.
Rationale: we already extensively test sum in TestCompareVariousMixedMetricsAggregations.
There was a problem hiding this comment.
I would like to keep these. They test reusing pools that have been used by the range-vector function.
There was a problem hiding this comment.
Do we need both sum test cases?
Don't forget we also have TestConcurrentQueries that should cover pooled slices being reused as well.
There was a problem hiding this comment.
one is probably sufficient. I'd suggest keeping this one (2m15s).
There was a problem hiding this comment.
Done, I've restored that in c4b2c57.
charleskorn
commented
Feb 18, 2025
Comment on lines
-2972
to
-2975
| // Generate combinations of 2, 3, and 4 labels. (e.g., "a,b", "e,f", "c,d,e", "a,b,c,d", "c,d,e,f" etc) | ||
| // Generate combinations of 2 labels. (e.g., "a,b", "e,f" etc) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 2)...) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 3)...) | ||
| labelCombinations = append(labelCombinations, testutils.Combinations(labelsToUse, 4)...) |
There was a problem hiding this comment.
Rationale: these functions all operate a series at a time, and any issue working with multiple series is likely to be surfaced running with two series.
jhesketh
reviewed
Feb 18, 2025
ying-jeanne
pushed a commit
that referenced
this pull request
Feb 19, 2025
* Make gauntlet tests ~3x faster by not using `t.Run()` * Don't bother running functions that operate over single series at a time over multiple series at a time * Don't run `sum` over functions that operate over range vector selectors, and only run with two series rather than up to four * Restore one `sum` over range vector function test case
ying-jeanne
pushed a commit
that referenced
this pull request
Feb 19, 2025
* Make gauntlet tests ~3x faster by not using `t.Run()` * Don't bother running functions that operate over single series at a time over multiple series at a time * Don't run `sum` over functions that operate over range vector selectors, and only run with two series rather than up to four * Restore one `sum` over range vector function test case
ying-jeanne
pushed a commit
that referenced
this pull request
Feb 20, 2025
* Make gauntlet tests ~3x faster by not using `t.Run()` * Don't bother running functions that operate over single series at a time over multiple series at a time * Don't run `sum` over functions that operate over range vector selectors, and only run with two series rather than up to four * Restore one `sum` over range vector function test case
ying-jeanne
pushed a commit
that referenced
this pull request
Feb 20, 2025
* Make gauntlet tests ~3x faster by not using `t.Run()` * Don't bother running functions that operate over single series at a time over multiple series at a time * Don't run `sum` over functions that operate over range vector selectors, and only run with two series rather than up to four * Restore one `sum` over range vector function test case
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What this PR does
This PR makes some improvements to the mixed metrics tests (aka the gauntlet), with the aim of preserving coverage while dramatically reducing how long they take to run.
On my machine, running all of the tests in
engine_test.gotook 3m35s before these changes, and now takes 45s.I've tried to build up these changes incrementally, I suggest reviewing each commit separately.
Which issue(s) this PR fixes or relates to
(none)
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX].about-versioning.mdupdated with experimental features.