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

Dask-CUDA Excessive Memory Usage #7718

Closed
quasiben opened this issue Dec 1, 2020 · 4 comments
Closed

Dask-CUDA Excessive Memory Usage #7718

quasiben opened this issue Dec 1, 2020 · 4 comments

Comments

@quasiben
Copy link

quasiben commented Dec 1, 2020

Details

Expected Result

Successful build

Actual Result

Build Error:

Command killed due to excessive memory consumption

would it be possible to increase the memory limit ?

Total download should be around 50 MB+. I need both the conda-forge and the rapidsai-nightly channels to build the docs

@humitos
Copy link
Member

humitos commented Dec 1, 2020

The server instance where your build is run is using the maximum memory allowed 😓

This a known issue in conda when using conda-forge channel, and in particular together with another one. See #7239 for some context on this topic (it has links to other previous issues and PRs)

I'm not sure if there is something immediate that we can do here, but I think the medium term solution is to support mamba at some point since it will mitigate this issue and improve all the build times.

@quasiben
Copy link
Author

quasiben commented Dec 1, 2020

Thanks @humitos for the quick response. I'll see if I can move things on my end

@humitos
Copy link
Member

humitos commented Dec 1, 2020

OK, I did a small trick and the build passed: https://readthedocs.org/projects/dask-cuda/builds/12463653/ --I force the upgrade of conda before the build starts. I don't know how long it will work anyways.

@quasiben
Copy link
Author

quasiben commented Dec 1, 2020

Thanks @humitos. I'll close this for now. Thank you again for your time!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants