Ever since users starting donating their compute prowess to Folding@home which would be put in use by researchers and scientists to fight against COVID-19, we have seen some spectacular feats being achieved. Last month, Folding@home announced that the total processing power added to its network has hit 1.5 Exaflops but just three weeks later, they have far exceeded the number with far more processing power that is faster than the top 500 supercomputers of the world.
Folding@Home's Total Processing Power To Fight COVID-19 Now At Almost 2.5 Exaflops
The announcement was posted on Twitter by Folding@home where they confirmed that their total processing power is now at ~2.4 Exaflops which is almost 2.5 Exaflops of peak processing power to crunch solutions to fight against COVID-19.
With our collective power, we are now at ~2.4 exaFLOPS (faster than the top 500 supercomputers combined)! We complement supercomputers like IBM Summit, which runs short calculations using 1000s of GPUs at once, by spreading longer calculations around the world in smaller chunks! pic.twitter.com/fdUaXOcdFJ
— Folding@home (@foldingathome) April 13, 2020
According to Folding@home, the total processing power exceeds that of the world's top 500 super-computers. it's great to see the tech community come together in an effort to fight against COVID-19 which has resulted in over 100,000 fatalities and over 1.8 million cases across the globe. The 3 big hardware companies, NVIDIA, Intel & AMD have pledged support to Folding@home. NVIDIA also announced that it would bring it AI and HPC expertise to fight against COVID-19.
We are here to help! pic.twitter.com/LVNuN3npiW
— Intel Gaming (@IntelGaming) March 14, 2020
We are honored to join the @WhiteHouse supercomputing partnership alongside other industry leaders working together to fight #COVID19. @AMD is dedicated to bring high performance compute compute to the brightest minds in the scientific community to fight this pandemic. https://t.co/jBMkeNGyTd
— Lisa Su (@LisaSu) April 6, 2020
The NVIDIA task force will help the consortium by:
- Applying AI: NVIDIA brings expertise in AI to help the consortium’s scientific teams ingest and process more data, faster to deliver better insights in less time.
- Accelerating Science: NVIDIA brings more than a decade of expertise in molecular biology, medical imaging, genomics, and computational fluid dynamics and visualization.
- Optimizing computing at scale: NVIDIA brings expertise in optimizing throughput of supercomputers. Optimizing performance is as important as adding more computing resources — a 20 percent optimization, for example, on a 330 petaflops system is equal to 60 petaflops, equivalent to the fourth fastest supercomputer in the world.
Folding@Home ~2.5 Exaflops Compute Power Statistics
| OS | AMD GPUs | NVIDIA GPUs | CPUs | CPU cores | TFLOPS | x86 TFLOPS |
|---|---|---|---|---|---|---|
| Windows | 112,816 | 430,181 | 847,028 | 5,719,659 | 952,891 | 1,929,969 |
| Linux | 7,506 | 115,643 | 558,752 | 4,877,640 | 270,402 | 496,915 |
| macOSX | 19 | 0 | 81,449 | 417,912 | 4,708 | 4,738 |
| Totals | 120,341 | 545,824 | 1,487,229 | 11,015,211 | 1,228,001 | 2,431,622 |
At the moment, there are a total of 545,824 NVIDIA GPUs, 102,341 AMD GPUs and a total of 1,487,229 CPUs available to feed the massive 2.4-2.5 Exaflops network. The CPUs amount a total of 11,015,211 processing cores which is simply mind-blowing. The cumulative processing power is listed at 2.431 Exaflops. There are several people still joining the network to add even more processing power to fight against COVID-19 which is simply amazing!
