Folding@Home Docker

Folding@home is a project focused on disease research. The problems we’re solving require so many computer calcul­ations – and we need your help to find the cures!

If you have a question regarding the setup or found a bug feel free ping in the Gitter chat or open an issue on Github.

Image Flavors

Currently there are two types of image available:

• latest, cpu - lightweight image for CPU only workloads
• latest-nvidia, nvidia - image with Nvidia GPU support. More information here

Usage

docker cli

CPU Instance

docker run \
  --name folding-at-home \
  -p 7396:7396 \
  -p 36330:36330 \
  -e USER=Anonymous \
  -e TEAM=0 \
  -e ENABLE_SMP=true \
  -e PUID=$(id -u) \
  -e PGID=$(id -g) \
  # Required for persistent data \
  -v /path/to/fahdata:/opt/fahclient/work \
  --restart unless-stopped \
  yurinnick/folding-at-home:latest

GPU Instance

docker run \
  --name folding-at-home \
  -p 7396:7396 \
  -p 36330:36330 \
  -e USER=Anonymous \
  -e TEAM=0 \
  -e ENABLE_SMP=true \
  -e ENABLE_GPU=true \
  -e PUID=$(id -u) \
  -e PGID=$(id -g) \
  # Required for persistent data \
  -v /path/to/fahdata:/opt/fahclient/work \
  --gpus all \
  --restart unless-stopped \
  yurinnick/folding-at-home:latest-nvidia

docker-compose

Note: Requires docker-compose version 1.28+

CPU Instance

docker-compose up -d folding-at-home-cpu

GPU Instance

docker-compose up -d folding-at-home-gpu

Building Locally

While providing pre-build images, we encourage everyone to read the Dockerfile and to build it yourself.

Based on your prefered flavor use the command below:

# CPU-only image
docker build -f Dockerfile -t folding-at-home:cpu .

# Nvidia CUDA image
docker build -f Dockerfile.nvidia -t folding-at-home:nvidia .

Parameters

• USER - Folding@home username (default: Anonymous)
• TEAM - Foldinghome team number (default: 0)
• PASSKEY - [optional] Folding@home passkey
• ENABLE_GPU - Enable GPU compute (default: false)
• ENABLE_SMP - Enable auto-configuration of SMP slots (default: true)
• POWER - "full" by default, but you can switch to "medium" or "light" (if your laptop runs too hot, or if your computer ventilates too much)
• PUID - User ID for data volume. See details: persistent-storage (default: 1000)
• PGID - Group ID for data volume. See details: persistent-storage (default: 1000)
• ALLOWED_HOSTS - Allowed remote access to specified subnet (default: 127.0.0.1/32)
• EXTRA_OPTIONS - Additional FAHClient command-line options (default: "")

Persistent Storage

By default Docker doesn't store any data outside of a continer, so upon stop/restart/recreate all temporary FAH data will be lost. To persistently store working data mount /opt/fahclient/work onto some directory on the disk.

While mounting local volume there may be permissions issues between the host and the container. Specify current users PUID/PGID as parameters to ensure that data volume owned by the same user inside the container.

docker run \
...
-v /path/to/fahdata:/opt/fahclient/work
-e PUID=$(id -u)
-e PGID=$(id -g)
...

GPU Support

This image currenly supports only Nvidia GPUs.

To enable Nvidia support in Docker follow the instructions on Nvidia Container Toolkit installation guide for Docker.