智慧物流>> fah>> 返回
项目作者： sthysel-freight

项目描述 ：
Folding@Home
高级语言： Shell
项目地址: git://github.com/sthysel-freight/fah.git
创建时间： 2020-04-16T13:09:21Z
项目社区：https://github.com/sthysel-freight/fah
开源协议：MIT License
下载


Folding@home docker containers

No Magic

This repo provides docker containers to run the folding@home client.

So this is one way of getting Folding@Home done. It has all the usual benefits
of Docker containerization. Assuming you are already a docker user the install
should take a few minutes, most of which is building the containers. Perhaps
I’ll push this to dockerhub later, idk.

Install

Prepare the docker environment if you have a Nvidia GPU

Be sure docker 19.03 or higher as well as the nvidia-container-toolkit is
installed. If docker is already installed be sure to restart docker after
installing nvidia-container-toolkit

For arch this would be:

  1. $ yay -S docker nvidia-container-toolkit
  2. $ sudo systemctl restart docker

Ubuntu or whatever will be very similar.

Once this is done run the test-gpu.sh script to verify your GPU’s are available:

  1. ❯ ./test-gpu.sh
  2. Thu Apr 16 11:14:52 2020       
  3. +-----------------------------------------------------------------------------+
  4. | NVIDIA-SMI 440.82       Driver Version: 440.82       CUDA Version: 10.2     |
  5. |-------------------------------|----------------------|----------------------+
  6. | GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
  7. | Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
  8. |===============================+======================+======================|
  9. |   0  GeForce RTX 2080    Off  | 00000000:01:00.0  On |                  N/A |
  10. |  0%   43C    P8     4W / 260W |    952MiB /  7979MiB |      6%      Default |
  11. +-------------------------------|----------------------|----------------------+
  13. +-----------------------------------------------------------------------------+
  14. | Processes:                                                       GPU Memory |
  15. |  GPU       PID   Type   Process name                             Usage      |
  16. |=============================================================================|
  17. +-----------------------------------------------------------------------------+

Run

So docker-compose does not support the —gpus flag at this time, so you have to
slum it with the run-gpu.sh script.

  1. $ ./run-gpu.sh -u sthysel -t 248479

If you have FAH passkey make sure its in your environment called FAH_PASSKEY

Resources