peteroyle.folding_at_home
Folding At Home
This guide explains how to set up a simple instance of the Folding@home client on several servers. We're using it at Spruce^ki to donate extra processing power from our staging system.
Important: Please do not use this setup in a production environment, even if you want to make good use of those powerful servers! It's not guaranteed to work properly—there are even spelling mistakes in the instructions. It might also cause your monitoring system to go crazy.
Requirements
Currently, this setup works only with recent versions of CentOS/RHEL and Debian/Ubuntu. We would welcome pull requests for other operating systems!
Impact
Our servers use only CPUs, and this setup would need some changes to work with GPUs. Therefore, we won’t earn a lot of points—my single laptop is earning 10 times more points than our entire staging setup.
However, in terms of the number of Work Units (WUs) solved, the staging system is 40 times ahead of that laptop. Some tasks can’t be done on GPUs, so adding more CPUs is still useful.
Role Variables
The following default settings are used in the config.xml file:
- gpu: false
- power: light (other options are 'medium' and 'full')
- user: ansible.spruce.ki (set this to your preferred username for stats)
- team: 257758 (this is team "spruce.ki"; you can use it or change it. Set to 0 for no team)
- state: present (use 'absent' to stop and remove the Folding@home package)
Dependencies
N/A
Example Playbook
Here’s the configuration we are currently using on our staging servers. So far, with full power, we haven’t noticed any impact on our productivity, but your experience may vary.
- hosts: servers
roles:
- { role: peteroyle.folding_at_home, user: ansible.spruce.ki, team: 257758, state: present, power: full }
License
Apache
Author Information
Spruce^ki develops a fully-featured, cloud-based telecommunications platform used by many Australian telecom companies. Visit [https://www.telcoworks.com.au] to see it in action.
ansible-galaxy install peteroyle.folding_at_home