Crowdsourced Supercomputing in the Fight against COVID-19

folding at home 4 28 21 post
Dell Technologies

To overcome a pandemic like COVID-19, even the smartest scientists won’t win the battle alone. We need a global village of scientists working in parallel and sharing their learnings to uncover the mysteries of the SARS-Cov-2 virus, which causes the disease, and to help discover treatments or cures to help end the pandemic.

This same spirit of shared purpose underlies the Folding@home distributed computing project. This project brings together a global village of computer users who lend their idle processing power to scientists who require massive amounts of computational power to run complex simulations, including those involved in the fight against COVID-19. While the processing power of one CPU can’t do much for the cause, the hundreds of thousands of CPUs lent to the Folding@home project could help the world conquer COVID-19.

While it has a broader scientific mission, the Folding@home project is currently focused sharply on the global effort to understand and find a cure for COVID-19. In response to the pandemic, a huge wave of citizen scientists joined Folding@home in 2020 to make their computers available to researchers who are working to understand the SARS-CoV-2 virus and to identify new therapeutic opportunities.

To join the project, volunteers download a software app that runs small parts of much larger simulations for the Folding@home project. On the back-end servers, the project’s algorithms piece together the parts to create a composite simulation that allows scientists to better understand the biology of processes they can’t observe directly — like the tricks the SARS-CoV-2 virus uses to invade our cells.

In just three months in 2020, more than a million new user devices joined the Folding@home project, according to the project’s director, Dr. Greg Bowman, Associate Professor of Biochemistry & Molecular Biophysics at Washington University of St. Louis. With this increased computational power, scientists were able to gain an understanding of the moving parts of every possible protein from the virus and capture large structural changes.

With the surge of new participants, the Folding@home distributed computing network grew to encompass approximately 2.4 exaflops of computational power. That’s more raw computing power than the world’s largest 500 traditional supercomputers combined.1 This level of theoretical performance earned the system the distinction of being the world’s first exascale computer.2

“Within a couple of months, we had something like a hundredfold growth in participation in Folding@home,” Dr. Bowman says in a Dell Technologies case study. “And we had five times the performance of the world’s fastest supercomputer, and that was with a very conservative estimate of what we had available on the Folding@home network. This scale allows us to do calculations that are inconceivable by any other means.”

Support from the tech industry

When it experienced the huge surge in volunteer participation in 2020, Folding@home came face to face with some backend challenges, including network limitations. At that point, many tech companies, among others, stepped up and offered additional resources to the project, which has servers scattered around the world, including some in cloud installations.

“A lot of tech organizations, like Dell Technologies, reached out to us and asked, ‘How can we help?’” Dr. Bowman says. “And that was really instrumental in our ability to make progress and harness this opportunity.”

Today, teams from Dell Technologies are part of the legions of volunteers who make their unused processing cycles available to Folding@home. The project tracks the contributions of each of its volunteer teams, in terms of active CPUs, work unit counts and more.

VMware is also actively involved in the project. Its contributions include a VMware vSphere Appliance that contains the Folding@home client software. When a user configures the software and deploys the appliance, the project client is up and running and ready to go to work.

“We’ve got an incredible community — here in the United States and all around the world — contributing to the scientific process of understanding the COVID virus and our other research pursuits,” Dr. Bowman says. “In terms of our volunteers, they are literally everywhere that is inhabited by human beings. This outpouring of support has been extraordinary.”

To learn more

For the full story, see the Dell Technologies case study “Crowdsourced Supercomputing.” And to learn how you can lend your extra process power to the project, visit Folding@home.

Explore more HPC solutions from Dell Technologies and Intel.

  1. Folding@home via Twitter, @foldingathome, April 13, 2020.
  2. Folding@home, “Citizen Scientists Create An Exascale Computer To Combat Covid-19,” July 26, 2020.
Related:

Copyright © 2021 IDG Communications, Inc.