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

[feature] Change the docker base image to nvidia/cuda, add docker-compose.yml #122

Closed
wants to merge 1 commit into from

Conversation

LucienShui
Copy link

Hi @utkuozdemir, thanks for your great work!

I noticed #110 #112 talking about how docker accessing the gpu, I thought using nvidia/cuda as the base image might be a good choice for less configuration to run in the docker, to I made this PR.

I added a file named docker-compose.yml and updated the document so that everyone can make everything fine and easy.

And also remote accessing from arm64 like rapberrypi-3b or later would still work because nvidia/cuda support arm64. (but not tested because I don't have such a device)

Feel free if there is any question or something wrong. :)

Best regards!

Types of changes

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to change)
  • I have read the CONTRIBUTING document.
  • My code follows the code style of this project.
  • My change requires a change to the documentation.
  • I have updated the documentation accordingly.
  • I have added tests to cover my changes.
  • All new and existing tests passed.

@utkuozdemir
Copy link
Owner

Thanks for the contribution - Just I am not sure if cuda's license allows us to use it as a base image - Nvidia's licenses are tricky. Need to look into that. If you have any findings, I'd appreciate if you share them here.

@felixsmli
Copy link

The plain Ubuntu 24.04 image can access nvidia-smi if you have installed NVIDIA Container Toolkit (which is the suggested way for using Docker and NVIDIA GPUs), as shown in this simple example found from NVIDIA's documentation

sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi

So I guess you can avoid nvidia/cuda and just update the example to:

 docker run -d -p 9835:9835 --runtime=nvidia --gpus all utkuozdemir/nvidia_gpu_exporter:1.2.0

Which is much simpler. But of course this won't work with machines that have Docker but cannot install the NVIDIA Container Toolkit.

@LucienShui LucienShui closed this Jun 3, 2024
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

Successfully merging this pull request may close these issues.

None yet

3 participants