Skip to content

A script to create a KVM attaching host graphic card for tensorflow nVidia CUDA computing

Notifications You must be signed in to change notification settings

eLvErDe/nvidia-docker-cuda-kvm-with-passthru

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

Edit top variables to suit your own setup

What's going to happen then

  1. Unbind graphic card id from current kernel driver

  2. Load PCI passthru drivers and attach the graphic card

  3. Create a KVM disk image using virt-builder

  4. Modify the image to deploy latest docker as well as nvidia-docker

  5. Create a simple start script with VNC console and host TCP proxy for SSH

You can now verify CUDA is working inside the KVM: docker pull nvidia/cuda nvidia-docker run --rm nvidia/cuda nvidia-smi

If you see your graphic card information, you're ready to go.

About

A script to create a KVM attaching host graphic card for tensorflow nVidia CUDA computing

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages