bbearce commented 2 months ago

Reviewers

@Didayolo, @ihsaan-ullah

Summary

Meant to transition us to poetry from pip dependency management. I only upped the python version to 3.9 and not even for all Dockerfiles. I ran into issues using python3.10. There are too many changes in a variety of packages. I think the only real way to do this is one package at a time and slowly over time bump up the python. For now I really just switched us to poetry from pip but kept the same packages versions. I think I changed one called ipdb and upgraded a bit because python 3.9 required it.

Dockerfiles updated:

Dockerfile

docker build --no-cache -f Dockerfile -t codabench-django:latest ./

Dockerfile.compute_worker

docker build --no-cache -f Dockerfile.compute_worker -t codabench-compute_worker:latest ./

Dockerfile.compute_worker_gpu

docker build --no-cache -f Dockerfile.compute_worker_gpu -t codabench-compute_worker_gpu:latest ./

Dockerfile.flower

docker build --no-cache -f Dockerfile.flower -t codabench-flower:latest ./

PS: Do we use Dockerfile.celery at all?

Issues this PR resolves

1413

Checklist

[x] Code review by me
[x] Hand tested by me
[x] I'm proud of my work
[x] Code review by reviewer
[x] Hand tested by reviewer
[x] CircleCi tests are passing
[x] Ready to merge

Didayolo commented 2 months ago

Nice progress. This is a fundamental change that needs to be tested deeply before merging.

ihsaan-ullah commented 2 months ago

Nice work done @bbearce. I see that there is poetry for compute worker. Will this affect the setting up of a compute worker? https://github.com/codalab/codabench/wiki/Compute-Worker-Management---Setup

Didayolo commented 2 months ago

Also, what is the file compute_worker/poetry.lock ?

Didayolo commented 3 weeks ago

This currently runs on https://codabench-test.lri.fr/
We need also to test compute workers
Building the GPU compute worker Dockerfile did NOT work
We are supposed to remove the requirements.txt files in this PR

bbearce commented 2 weeks ago

Was working on this over the weekend. A couple questions.

I had to delete basically every docker image to build the gpu one. Could we request maybe 50GB total for this VM (157.136.249.152)?
The compute worker and gpu worker both start from the same python3.9 and I believe that is for consistency. However installing cuda drivers and nvidia-toolkit is harder that way. I have a current Dockerfile I'm working on and it builds to about ~8GB but it starts from image ubuntu:20.04. Can we start from ubuntu or even better yet, nvidia-cuda images? Adding python after the fact is easy and I could even integrate pyenv, though don't want to go too crazy.
- The obvious con is that it wouldn't be the same base image as the compute worker, unless we switched that to come from ubuntu:20.04 as well (just skip cuda installation).

What do you think?

ihsaan-ullah commented 2 weeks ago

If you want to use a cuda image, you can check this one which we are using in another project

https://github.com/FAIR-Universe/HEP-Challenge/blob/master/docker/Dockerfile

bbearce commented 2 weeks ago

Exactly. The more I think about it, this is great but how much should the gpu worker and the cpu worker match (Dockerfile wise)? My take is that you'd want them to be as similar as possible and I'm worried using an nvidia image for the cpu worker doesn't make sense so then these would come from different bases. Does anyone else think having both workers start from ubuntu is too low level?

PS: Totally down to do an nvidia image for gpu worker and python or ubuntu for cpu worker if folks don't think the different base matters much. Would greatly simplify gpu setup, effectively allowing us to skip cuda install altogether.