This content originally appeared on DEV Community and was authored by Open Sauce Projects
Why
- If you need gpu acceleration in a container or just have too much free time on your hands, you can run all kinds of things. For example Large Language Models or GUI apps.
- You can even create your own gpu accelerated docker image by using Nvidia ubuntu images
How
- We have to install the NVIDIA Container Toolkit
-
The package can be installed from the Extras repository
sudo pacman -S nvidia-container-toolkit
-
Configure the container runtime by using the nvidia-ctk command. The
nvidia-ctk
command modifies the/etc/docker/daemon.json
file on the host. The file is updated so that Docker can use the NVIDIA Container Runtime.
sudo nvidia-ctk runtime configure --runtime=docker
-
For the change to take effect we have to restart the docker daemon.
sudo systemctl restart docker
-
You can add the gpu using docker compose:
service_name: image: .... deploy: resources: reservations: devices: - driver: nvidia count: 1 capabilities: [gpu]
If you use the
docker run
command, you can do:
sudo docker run --runtime=nvidia --gpus all ....
Resources
This content originally appeared on DEV Community and was authored by Open Sauce Projects
Open Sauce Projects | Sciencx (2024-08-24T00:19:03+00:00) Access Nvidia gpu inside docker container in Arch linux. Retrieved from https://www.scien.cx/2024/08/24/access-nvidia-gpu-inside-docker-container-in-arch-linux/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.