Stephen Smith's Blog

Musings on Machine Learning…

Solving COVID-19 with Folding@Home

with 3 comments


Pretty much all of us have several computers lying around the house, most of them not doing anything for 90% of the time. The people looking to discover extraterrestrial life came up with the idea of analysing everything ever received via radio telescopes to look for patterns or messages. They devised a distributed system where any computer could run a client that would run when the computer is idle and work on a small piece of the problem. All the pieces are uploaded to a central set of servers where they are combined. This project is called seti@home and has a vast network of computers looking for ET.

Proteins are extremely complicated molecules that our body’s DNA manufactures to perform all the functions within our cells. Our DNA specifies the individual elements that make up the protein, but the true wonder of proteins comes from how this string of building blocks folds together. Studying this process is vastly complicated and until recently far beyond our computational power to model. With new algorithms and a vast network of distributed computers, the folding@home project was born and today forms the largest supercomputer the world has ever seen.

There are hundreds of problems being worked on. Each computer is given a series of work units to calculate. You get points for work accomplished that you can use for bragging rights. Folding@home has been running for twenty years, since 2000, but with problems associated with COVID-19 being added, usage has soared. I have all my household computers dedicated to this task, since as it stands it seems like solving this problem is the only way that life will return to normal.

Installation and Configuration

Folding@home will run on most Intel/AMD based computers, usually as long as they are at least 64-bit and dual-core. I’m running it on three 2008 MacBooks with core2duo processors, an 2015 HP laptop with an i3 processor and my new gaming laptop with an i7 and nVidia GPU. The installation is straightforward. You download the correct install image, for either Windows, MacOS or Linux and run the installer. This installs and configures a process that is always running and you can see in your task tray. You can access it from your browser at or via one of the configuration GUIs. Here is it running on my gaming laptop:

To get the GPU going, I needed to reboot. I suspect Tensorflow of X-Plane still owned the GPU. Then I could add a GPU slot in the advanced control configuration screen.

Joining a team is fun, you can combine your resources, note that I’m part of team 253800 which is the SunshineCoastBC_Team.

Notice the lower right where it describes the project your computer is currently working on. You can also use the protein viewer to see the associated protein.

Only Partly Open Source

I was planning on running this on my nVidia Jetson Nano and two Raspberry Pis. Sadly, I discovered that folding@home only runs on Intel/AMD processors and not ARM processors. I went to have a look to see if I could compile for ARM, but found the source code isn’t open source. Folding@home uses several open source libraries like gromacs, but they claim they don’t want to go fully open source to prevent people cheating on the points system or corrupting the results.

I think it would benefit them to support ARM processors; the millions of Raspberry Pis and other SBCs could really add to the effort. I looked at a couple of their dependent projects like gromacs and found these do have ARM support, so I don’t think it would be hard to add. Especially since they keep promising a mobile version.



There isn’t a giant claim that folding@home has cured a major disease yet. However, it has contributed quite a number of results leading it that direction. You can see all the scientific papers that have been generated from the results of all this computation here.


Watching the folding@home web page is fascinating. It enables you to contribute to solving a number of major problems in medicine. If you configure folding@home for a light load, you won’t even notice it is there. Just leave your computer turned on and let it solve the world’s problems.

Written by smist08

April 25, 2020 at 4:47 pm

Posted in Life

Tagged with , , ,

3 Responses

Subscribe to comments with RSS.

  1. […] watching the performance of the various computers around the house running Folding@Home which I blogged about last time. Folding@Home provides all sorts of interesting statistics that I’ll talk about as well. The main […]

  2. […] If you are doing something more general or completely new, you can consider a general parallel processing library like OpenCL which has support for all sorts of devices, including the limited SIMD coprocessors included with most modern CPUs. A good example of a program that uses OpenCL is Folding@Home which I blogged on here. […]

  3. […] is being used to solve protein folding problems around developing a cure for COVID-19, similar to folding@home. This is a truly impressive warehouse of technology and shows where you can go with the ARM CPU and […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: