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.

Sign up for GitHub

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

Jump to bottom

Build and test with CUDA 12.5.1 #1405

Merged
merged 7 commits into from
Jul 19, 2024
Merged

Build and test with CUDA 12.5.1 #1405

merged 7 commits into from
Jul 19, 2024

Conversation

KyleFromNVIDIA
Copy link
Contributor

@KyleFromNVIDIA KyleFromNVIDIA commented Jul 11, 2024 •
edited
Loading

This PR updates the latest CUDA build/test version 12.2.2 to 12.5.1.

Contributes to rapidsai/build-planning#73

@KyleFromNVIDIA KyleFromNVIDIA added non-breaking Non-breaking change improvement Improvement / enhancement to an existing function labels Jul 11, 2024
@github-actions github-actions bot added the conda Related to conda and conda configuration label Jul 11, 2024
@KyleFromNVIDIA KyleFromNVIDIA marked this pull request as ready for review July 16, 2024 13:26
@KyleFromNVIDIA KyleFromNVIDIA requested review from a team as code owners July 16, 2024 13:26
@KyleFromNVIDIA KyleFromNVIDIA requested a review from msarahan July 16, 2024 13:26
@jameslamb jameslamb requested review from jameslamb and removed request for msarahan July 16, 2024 15:03
@jakirkham
Copy link
Member

/merge

@jakirkham
Copy link
Member

The notebook job appears to be hung. It has run for 3hrs without doing much. Before that it ran 6hrs and was finally terminated as it reached the CI time limit. Wondering if there is an issue with the notebook job that requires further independent investigation

Note: All other CI jobs have passed

@KyleFromNVIDIA
Copy link
Contributor Author

#1399 is also having problems with the notebook tests

@jakirkham
Copy link
Member

Think James is debugging PR: #1407

@jameslamb
Copy link
Member

The notebook job appears to be hung. It has run for 3hrs without doing much.

In logs here, it looks like it's stuck on the NYC taxi notebook.

[NbConvertApp] Writing 7305 bytes to /tmp/nyc_taxi_years_correlation-test.py

In my interactive testing (on a machine with V100s, CUDA 12.2, and using RAPIDS 24.08 nightly conda packages), I also observed that notebook hanging for a long time. I'd hoped it was just something with my local setup 😭

Specifically, this cell ran for 50 minutes before I killed it:

cuspatial/notebooks/nyc_taxi_years_correlation.ipynb

Lines 203 to 219 in c8616c1

"source": [
"%%time\n",
"taxi2016['PULocationID'] = 264\n",
"taxi2016['DOLocationID'] = 264\n",
"for i in range(len(pip_iterations)-1):\n",
" start = pip_iterations[i]\n",
" end = pip_iterations[i+1]\n",
" \n",
" zone = taxi_zone_rings[start:end]\n",
" pickups = cuspatial.point_in_polygon(pickup2016, zone)\n",
" dropoffs = cuspatial.point_in_polygon(dropoff2016, zone)\n",
"\n",
" for j in pickups.columns:\n",
" taxi2016['PULocationID'].loc[pickups[j]] = j\n",
" for j in dropoffs.columns:\n",
" taxi2016['DOLocationID'].loc[dropoffs[j]] = j"
]

Even when I reduced that for loop to for i in range(2) 😬

I'll keep investigating over on #1407. This is blocking CI for rapidsai/docker as well.

@jameslamb jameslamb mentioned this pull request Jul 18, 2024
Merged
@jakirkham
Copy link
Member

/merge

@rapids-bot rapids-bot bot merged commit 068f461 into rapidsai:branch-24.08 Jul 19, 2024
73 checks passed
@jakirkham
Copy link
Member

Thanks all! 🙏

@jakirkham
Copy link
Member

All packages uploaded

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jameslamb jameslamb jameslamb approved these changes

@jakirkham jakirkham jakirkham approved these changes

Assignees
No one assigned
Labels
conda Related to conda and conda configuration improvement Improvement / enhancement to an existing function non-breaking Non-breaking change
Projects
cuSpatial
Status: Todo
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@KyleFromNVIDIA @jakirkham @jameslamb