-
-
Notifications
You must be signed in to change notification settings - Fork 18.5k
CI: Try running single_cpu/not single_cpu tests together #52239
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
Conversation
This doesn't quite work yet (need to figure out how to propagate the PATTERN variable). |
I would recommend bumping up the |
It's already 3 hours which should be fine (all tests seem to take about 1ish hour). |
Looks like on the latest commit that the single cpu task isn't running |
…pu tests together
Nice cleanup! Thanks @lithomas1. Hoping to follow up with #52211 to reduce the number of queued workflows even further |
…ngle_cpu tests together) (#52292) Backport PR #52239: CI: Try running single_cpu/not single_cpu tests together Co-authored-by: Thomas Li <[email protected]>
doc/source/whatsnew/vX.X.X.rst
file if fixing a bug or adding a new feature.If I did this right, should get us from 20 jobs to around 14 jobs.
This'll probably slow down the current jobs by ~10 minutes (but they already take 1 hr), but we'll save at least the 5 minutes of compiling pandas on the single_cpu jobs + whatever it takes to setup the conda envs there.
Hopefully the shorter queue times during peak periods makes up for the extra time.