Fix #123: Use correct pooled standard deviation formula in multi-worker aggregation#142
Open
mcurrier2 wants to merge 1 commit into
Open
Fix #123: Use correct pooled standard deviation formula in multi-worker aggregation#142mcurrier2 wants to merge 1 commit into
mcurrier2 wants to merge 1 commit into
Conversation
…er aggregation The previous implementation computed std_dev as sqrt(sum(n_i * sigma_i^2) / N), which only accounts for within-group variance. When workers have different mean latencies, this significantly underestimates the true combined standard deviation. Apply the law of total variance: Var(X) = E[Var(X|G)] + Var(E[X|G]), which adds between-group variance (weighted squared deviation of each worker's mean from the combined mean) to the within-group variance. - Fix pooled std_dev formula in aggregate_latency_metrics() - Add test verifying between-group mean differences are captured - Add test verifying same-mean case still produces correct result - All existing tests passing, clippy clean AI-assisted-by: Claude Opus 4.6 Co-authored-by: Cursor <cursoragent@cursor.com>
📝 WalkthroughSummary by CodeRabbit
WalkthroughAdds three planning/documentation Markdown files (release-polish-plan.glow, release-polish-plan.md, tier1-pr-diffs.md) recording a release-polish review pass, filed issues, and Tier 1 PR diffs. Separately, corrects the pooled standard deviation formula in metrics.rs multi-worker latency aggregation and adds two unit tests. ChangesRelease Polish Planning Documentation
Pooled Standard Deviation Fix
Estimated code review effort: 3 (Moderate) | ~25 minutes Related issues: Suggested labels: documentation, bug, metrics Suggested reviewers: (none identified) 🐰 A pooled variance once ran astray, 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.cursor/plans/release-polish-plan.md:
- Around line 108-115: The release validation summary is inconsistent between
the plan documents, with mismatched test totals and modified-file counts. Update
the Validation summary in the release-polish plan using the same authoritative
numbers as the corresponding release-polish-plan.glow entry, and ensure both
summaries in the release-plan documents reflect identical counts. Keep the fix
localized to the Validation section and the release summary fields so the two
files stay synchronized.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Enterprise
Run ID: 692b7130-f254-4965-b23c-1b0a49e6dac7
⛔ Files ignored due to path filters (6)
.cursor/plans/benchmark-chore-release-polish.csvis excluded by!**/*.csv.cursor/plans/benchmark-fix-121-blocking-forward-duration.csvis excluded by!**/*.csv.cursor/plans/benchmark-fix-122-blocking-roundtrip-first-message.csvis excluded by!**/*.csv.cursor/plans/benchmark-fix-124-async-server-flag-forwarding.csvis excluded by!**/*.csv.cursor/plans/benchmark-fix-129-server-exit-status-check.csvis excluded by!**/*.csv.cursor/plans/benchmark-fix-131-shm-direct-mutex-init-checks.csvis excluded by!**/*.csv
📒 Files selected for processing (4)
.cursor/plans/release-polish-plan.glow.cursor/plans/release-polish-plan.md.cursor/plans/tier1-pr-diffs.mdsrc/metrics.rs
Comment on lines
+108
to
+115
| ## Validation | ||
|
|
||
| ``` | ||
| cargo fmt ✓ (no changes) | ||
| cargo clippy ✓ (0 warnings with -D warnings) | ||
| cargo test ✓ (42 passed, 0 failed) | ||
| cargo build ✓ | ||
| ``` |
There was a problem hiding this comment.
📐 Maintainability & Code Quality | 🟠 Major | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
printf '--- files ---\n'
git ls-files '.cursor/plans/release-polish-plan.md' '.cursor/plans/release-polish-plan.glow' || true
printf '\n--- md lines 100-150 ---\n'
sed -n '100,150p' .cursor/plans/release-polish-plan.md | cat -n
printf '\n--- glow lines 100-150 ---\n'
sed -n '100,150p' .cursor/plans/release-polish-plan.glow | cat -nRepository: redhat-performance/rusty-comms
Length of output: 259
🏁 Script executed:
#!/bin/bash
set -euo pipefail
printf '--- file stats ---\n'
wc -l .cursor/plans/release-polish-plan.md .cursor/plans/release-polish-plan.glow
printf '\n--- md validation/search ---\n'
rg -n "cargo (fmt|clippy|test|build)|passed|modified|files" .cursor/plans/release-polish-plan.md
printf '\n--- glow validation/search ---\n'
rg -n "cargo (fmt|clippy|test|build)|passed|modified|files" .cursor/plans/release-polish-plan.glowRepository: redhat-performance/rusty-comms
Length of output: 259
🏁 Script executed:
#!/bin/bash
set -euo pipefail
python3 - <<'PY'
from pathlib import Path
for p in [Path('.cursor/plans/release-polish-plan.md'), Path('.cursor/plans/release-polish-plan.glow')]:
print(f'--- {p} ---')
text = p.read_text()
for needle in ['42 passed', '472 passed', '17', '16', 'cargo fmt', 'cargo clippy', 'cargo test', 'cargo build']:
if needle in text:
print('contains', needle)
PYRepository: redhat-performance/rusty-comms
Length of output: 499
🏁 Script executed:
#!/bin/bash
set -euo pipefail
python3 - <<'PY'
from pathlib import Path
for p in [Path('.cursor/plans/release-polish-plan.md'), Path('.cursor/plans/release-polish-plan.glow')]:
print(f'FILE: {p}')
try:
text = p.read_text()
except Exception as e:
print('ERROR:', e)
continue
for i, line in enumerate(text.splitlines(), 1):
if 100 <= i <= 150:
print(f'{i:4d}: {line}')
PYRepository: redhat-performance/rusty-comms
Length of output: 9171
Reconcile the release summary across .cursor/plans/release-polish-plan.md and .cursor/plans/release-polish-plan.glow. The validation totals and modified-file counts diverge (42 passed vs 472 passed, 17 vs 16 files), so one authoritative summary needs to be synced in both places.
🧰 Tools
🪛 markdownlint-cli2 (0.22.1)
[warning] 110-110: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.cursor/plans/release-polish-plan.md around lines 108 - 115, The release
validation summary is inconsistent between the plan documents, with mismatched
test totals and modified-file counts. Update the Validation summary in the
release-polish plan using the same authoritative numbers as the corresponding
release-polish-plan.glow entry, and ensure both summaries in the release-plan
documents reflect identical counts. Keep the fix localized to the Validation
section and the release summary fields so the two files stay synchronized.
📈 Changed lines coverage: 100.00% (9/9)✅ All new and modified lines are covered by tests!📊 Code Coverage Summary
|
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
1 participant
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.
Summary
aggregate_latency_metrics()to use the law of total variance:Var(X) = E[Var(X|G)] + Var(E[X|G]), which accounts for between-group mean differencessqrt(sum(n_i * sigma_i^2) / N)only computed within-group variance, underestimating combined std_dev when workers have different mean latenciesTest plan
test_pooled_stddev_between_group_varianceverifies between-group differences are capturedtest_pooled_stddev_same_meansverifies same-mean case still worksFixes #123
Made with Cursor