Improve RSS measurement#515
Merged
Merged
Conversation
The harness took a single RSS snapshot after a forced full GC, so the RSS column carried no variance and small differences read as signal when they were only noise. Sample RSS once per iteration across the bench window (default, gc and warmup harnesses) and report it as mean +/- stddev% with a mean-based ratio. The raw per-iteration samples are stored under rss_samples in the JSON; the post-GC rss snapshot and lifetime maxrss are still recorded.
Member
Can we skip the diff in this PR? There's already a PR #514 for that problem. |
Member
|
The other commit looks good to me 👍 |
eightbitraptor
force-pushed
the
mvh-improve-rss
branch
from
a446486 to
55c1d3a
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Previously we would just capture the RSS (using either
smaps_rollupif available,ps -oif not), but we would only do this once at the end of each benchmark run.This change samples the RSS during each benchmark iteration (outside of the timed window), so that we can measure and report on the RSS variance during benchmark runs.