How to Install Folding@home with NVIDIA CUDA GPU Support

If you want to maximize your contribution to Folding@home, enabling CUDA GPU acceleration is one of the best ways to boost performance. NVIDIA GPUs can dramatically speed up protein folding calculations compared to CPU-only setups.

This tutorial walks you through installing Folding@home with NVIDIA CUDA support on Linux, using Docker and the NVIDIA Container Toolkit. By the end, you’ll have Folding@home running on your GPU, crunching numbers at full speed.

📖 Source: NVIDIA Container Toolkit Install Guide

Why Use Folding@home with CUDA?

Faster Work Units: NVIDIA GPUs can process Folding@home tasks much faster than CPUs.
Better PPD (Points Per Day): More points means more impact for your team and the global project.
Efficient Resource Use: If you already have an NVIDIA GPU, CUDA lets you unlock its full potential.

Prerequisites

Before starting, make sure you have:

A Linux system with an NVIDIA GPU.
Latest NVIDIA GPU drivers installed.
Docker or another container runtime installed.
Sudo/root access to configure system settings.

Step 1: Install the NVIDIA Container Toolkit

The NVIDIA Container Toolkit is required to give Docker access to your GPU. Installation steps vary by distribution.

Ubuntu / Debian

# Add NVIDIA GPG key and repository
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey \
  | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg

curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \
  | sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' \
  | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

# Install toolkit
sudo apt-get update
sudo apt-get install -y nvidia-container-toolkit

Fedora / RHEL / CentOS

# Add NVIDIA repo
curl -s -L https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo \
  | sudo tee /etc/yum.repos.d/nvidia-container-toolkit.repo

# Install toolkit
sudo dnf install -y nvidia-container-toolkit

Step 2: Configure Docker for NVIDIA Runtime

Once installed, configure Docker to use NVIDIA’s runtime:

sudo nvidia-ctk runtime configure --runtime=docker
sudo systemctl restart docker

This step ensures containers can access your GPU devices.

Step 3: Verify GPU Access in Containers

Run a test container to check if CUDA works:

docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

If you see your GPU details, CUDA is working correctly.

Step 4: Run Folding@home with NVIDIA CUDA

Now you can launch Folding@home with GPU acceleration:

docker run -d \
  --gpus all \
  --name foldingathome \
  foldingathome/fahclient:latest \
  --user YourUsername \
  --team YourTeamNumber \
  --gpu true

--gpus all enables CUDA GPU access.
Replace YourUsername and YourTeamNumber with your actual Folding@home details.
--gpu true ensures GPU folding is turned on.

Step 5: Monitor Folding@home Performance

You can monitor your Folding@home container logs:

docker logs -f foldingathome

You should see work units assigned to your GPU. You’ll also notice much higher PPD compared to CPU folding.

Troubleshooting

If CUDA isn’t detected inside the container, recheck the NVIDIA driver installation.
Ensure Docker is using the NVIDIA runtime (nvidia-smi inside container must work).
Check compatibility between CUDA version and GPU driver.

Conclusion

By following this guide, you’ve successfully set up Folding@home with NVIDIA CUDA GPU support. This setup unlocks the full power of your NVIDIA GPU, helping you contribute to groundbreaking research faster and more efficiently.

Whether you’re folding solo or with a team, enabling GPU acceleration ensures your contributions make a bigger impact.

Triii Technologies LLC