Folding@home Flexes 470 PetaFLOPS Compute, Surpassing Top Supercomputers In COVID-19 Research Push

pande lab stanford folding at home
The Folding@home network has been working hard to defeat COVID-19 (coronavirus). This battle has caused the network to pass some interesting milestones as result of renewed interest in folding for a cure or vaccine. The Folding@home network now boasts 470 PetaFLOPS of compute power and is more powerful than the world’s top 7 supercomputers. Folding@home is a “distributed computing” project. It was started by Dr. Vijay Pande at Stanford University and is now based at the Washington University in St. Louis School of Medicine. The project relies on the processing power of volunteer personal devices that process “different portions of data” simultaneously. The Folding@home team contends that this approach can “significantly accelerate our research” and it has allowed them to publish over 223 research papers as a result.

At the beginning of March, Folding@home was pushing out about 98.7 PetaFLOPS of compute power. This was an impressive number, but it paled in comparison to supercomputers like the Summit supercomputer at the Oak Ridge National Laboratory (ORNL). The Folding@home output is now at 470 PetaFLOPS. Folding@home is now not only more powerful than Summit, but is more powerful than the top seven supercomputers combined. Summit boasts 149 PetaFLOPS while the six next fastest supercomputers do not break 100 PetaFLOPS.

folding at home covid 19 protein spike
Image from the Washington University in St. Louis School of Medicine via Twitter

Why has Folding@home’s compute power increased so quickly over such a short period of time? The project has long been focused on disease research, but is now paying particular attention to the “coronavirus” or COVID-19. As a result, Folding@home has witnessed an influx of volunteer compute resource donors since the outbreak - a sure sign the humanity has it's priorities in order, at least in this area. The project was recently able to release a simulation of COVID-19’s spike protein. This simulation will help researchers to better understand what occurs during infection.

Folding@Home Coronavirus GPU Workload
Folding@Home Coronavirus Project GPU Workload

Folding@home is still looking for volunteers with spare CPUs. or GPUs. You can find out more here about how to download the installer from the Folding@home websites. The software uses idle resources so it should not interfere with your device’s performance.

NVIDIA is also calling upon users to put their idle GPU resources to use. There are several GPU projects currently available that are focused on COVID-19. You can click here to learn more about NVIDIA’s call to action, and how you can also join HotHardware’s F@H team (Team # 32407).

Image credit: Folding@Home