# GPU on HUNT Cloud
We offer GPU machine types in our IaaS tier allowing you to run the AI/ML workloads of your choice.
See GPU specification for more details on the specific GPU models we offer.
# Software
To get you started, the GPU machines come pre-installed with the software below. If you need to install any other tools or versions you can do this as well.
- Ubuntu 20.04 LTS
- Docker
- NVIDIA 470 drivers or later (from the
graphics-drivers
repository (opens new window)) - NVIDIA Container Toolkit (also known as
nvidia-docker
)
# How to use
Run the following command to call the NVIDIA System Management Interface (nvidia-smi
) to manage and verify that things are working:
nvidia-smi
If you wish to use a specific version of CUDA, you can either install it manually or use the NVIDIA Container Toolkit to run a container of your choosing. Here is an example for CUDA version 11.0
:
sudo docker run --rm --runtime=nvidia nvidia/cuda:11.0-base nvidia-smi
# Jupyter in Nvidia Docker
If you want to use Jupyterlab on GPU machine we recommend iot-salzburg/gpu-jupyter (opens new window) docker image. This project uses the NVIDIA CUDA image as the base image and installs their toolstack on top of it to enable GPU calculations in the Jupyter notebooks. Python packages Tensorflow and Pytorch are preinstalled to match the Cuda version (GPU drivers).
docker pull cschranz/gpu-jupyter:v1.4_cuda-11.2_ubuntu-20.04
docker run --rm -d --runtime=nvidia -v ${HOME}:${HOME} --workdir ${HOME} -e HOME=${HOME} -e GRANT_SUDO=yes -e JUPYTER_ENABLE_LAB=yes -p 8888:8888 --user root --name gpu-jupyter cschranz/gpu-jupyter:v1.4_cuda-11.2_ubuntu-20.04
docker ps
This starts an instance of GPU-Jupyter with the tag v1.4_cuda-11.2_ubuntu-20.04
at http://localhost:8888
(port 8888). The default password is gpu-jupyter
.
More details about docker image can also be found on docker hub (opens new window).
# More info
- NVIDIA Drivers overview (opens new window) for an overview of drivers for NVIDIA GPUs.
- NVIDIA Container Toolkit documentation (opens new window) for more details on how to easily use and run containers.
- NVIDIA GPU Cloud catalog (opens new window) for an overview of CUDA containers from NVIDIA.