Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix flaky spill stats check in aggregation test #9459

Closed
wants to merge 1 commit into from

Conversation

xiaoxmeng
Copy link
Contributor

@xiaoxmeng xiaoxmeng commented Apr 11, 2024

The flaky is due to spill check condition is fragile which is based on whether we have received
more than one input at partial aggregation. We shall change to final aggregation and given we
have four drivers so it is not guarantee one aggregation has received more than one input row.
Since we have recorded the spill injection count, then we just rely on this to check spill stats.

This PR also restrict the case that we trigger spill for output memory reservation by checking
if table is null or empty

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Apr 11, 2024
Copy link

netlify bot commented Apr 11, 2024

Deploy Preview for meta-velox canceled.

Name Link
🔨 Latest commit 02d77dc
🔍 Latest deploy log https://app.netlify.com/sites/meta-velox/deploys/6618c25f28371300083cf0b1

@xiaoxmeng xiaoxmeng marked this pull request as ready for review April 11, 2024 23:27
@xiaoxmeng xiaoxmeng force-pushed the flaky branch 2 times, most recently from 1fa1ca3 to e6971b9 Compare April 11, 2024 23:38
@xiaoxmeng xiaoxmeng requested a review from mbasmanova April 11, 2024 23:39
@facebook-github-bot
Copy link
Contributor

@xiaoxmeng has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

The flaky is due to spill check condition is fragile which is based on whether we have received
more than one input at partial aggregation. We shall change to final aggregation and given we
have four drivers so it is not guarantee one aggregation has received more than one input row.
Since we have recorded the spill injection count, then we just rely on this to check spill stats.

This PR also restrict the case that we trigger spill for output memory reservation by checking
if table is null or empty
@facebook-github-bot
Copy link
Contributor

@xiaoxmeng has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.

@facebook-github-bot
Copy link
Contributor

@xiaoxmeng merged this pull request in 2b2ef52.

@xiaoxmeng xiaoxmeng deleted the flaky branch April 12, 2024 16:10
Copy link

Conbench analyzed the 1 benchmark run on commit 2b2ef52f.

There were no benchmark performance regressions. 🎉

The full Conbench report has more details.

Joe-Abraham pushed a commit to Joe-Abraham/velox that referenced this pull request Jun 7, 2024
Summary:
The flaky is due to spill check condition is fragile which is based on whether we have received
more than one input at partial aggregation. We shall change to final aggregation and given we
have four drivers so it is not guarantee one aggregation has received more than one input row.
Since we have recorded the spill injection count, then we just rely on this to check spill stats.

This PR also restrict the case that we trigger spill for output memory reservation by checking
if table is null or empty

Pull Request resolved: facebookincubator#9459

Reviewed By: mbasmanova

Differential Revision: D56043610

Pulled By: xiaoxmeng

fbshipit-source-id: 4deaec31e7097a4aa13b695e515ac6baecc4f3ce
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Merged
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants