*: Display CPU cores and CPU seconds for samples count profiles #4304
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
|
🤖 Meticulous spotted visual differences in 183 of 375 screens tested: view and approve differences detected. Last updated for commit fe9d3ff. This comment will update as new commits are pushed. |
metalmatze
reviewed
Feb 14, 2024
parca.yaml
Outdated
There was a problem hiding this comment.
This was probably pushed by accident.
Comment on lines
+366
to
+370
| preProjection := []logicalplan.Expr{ | ||
| logicalplan.Mul(logicalplan.Div(logicalplan.Col(profile.ColumnTimestamp), logicalplan.Literal(step.Milliseconds())), logicalplan.Literal(step.Milliseconds())).Alias(profile.ColumnTimestamp), | ||
| logicalplan.DynCol(profile.ColumnLabels), | ||
| logicalplan.Col(profile.ColumnDuration), | ||
| } |
parca.yaml
Outdated
There was a problem hiding this comment.
Looks like this got pushed by accident.
Comment on lines
-488
to
-500
| // Because we store the period with each sample yet query for the sum(period) we need to normalize by the amount of values (rows in a database). | ||
| period := periodSum / valueCount | ||
| // Because we store the duration with each sample yet query for the sum(duration) we need to normalize by the amount of values (rows in a database). | ||
| duration := durationSum / valueCount | ||
|
|
||
| // If we have a CPU samples value type we make sure we always do the next calculation with cpu nanoseconds. | ||
| // If we already have CPU nanoseconds we don't need to multiply by the period. | ||
| valuePerSecondSum := valueSum | ||
| if sampleTypeUnit != "nanoseconds" { | ||
| valuePerSecondSum = valueSum * period | ||
| } | ||
|
|
||
| valuePerSecond := float64(valuePerSecondSum) / float64(duration) |
This patch changes the UX of interacting with CPU profiling data that only reports samples count. It ensures that samples count is multiplied by their period (the time between samples), which allows comparing two profiles that were taken at different frequencies. As a side effect while changing the queries, this also fixed an issue where the average cpu time was previously summed up over a step period, which was incorrect, as the total cpu time over the step needs to be divided by the total duration over the step.
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.
This patch changes the UX of interacting with CPU profiling data that only reports samples count. It ensures that samples count is multiplied by their period (the time between samples), which allows comparing two profiles that were taken at different frequencies.
As a side effect while changing the queries, this also fixed an issue where the average cpu time was previously summed up over a step period, which was incorrect, as the total cpu time over the step needs to be divided by the total duration over the step.