Graphics processing units (GPUs) from NVIDIA are delivering over 1 petaflop (1000 teraflops) of processing power in a distributed computing application at Stanford University. According to statistics published by the university, 11,370 active NVIDIA GPUs provided 1.251 petaflops, or 42% of the total processing power for the Folding@home application. In comparison, NVIDIA said that 208,268 CPUs running Windows were active, contributing 198 teraflops, or 6% of the total processing power.
Stanford released a Folding@home client specifically for NVIDIA GPUs in June. Developed using NVIDIA CUDA, a C language programming environment for many-core parallel architectures, the CUDA port of the Folding@home client has delivered more processing power than any other architecture in the history of the project, according to NVIDIA.
The Folding@home application has become a major force in researching cures for life-threatening diseases such as cancer, cystic fibrosis, and Parkinson’s disease by combining the computing horsepower of millions of processors to simulate protein folding. The Folding@home project is the latest example in the expanding list of non-gaming applications for GPUs. By running the Folding@home client on NVIDIA GPUs, protein-folding simulations can be done 140 times faster than on some of today’s traditional CPUs.
“As these statistics show, the impact of NVIDIA GPUs on protein folding simulations has been extraordinary,” said Vijay Pande, associate professor of chemistry at Stanford University and director of the Folding@home project. “Teams that are folding with NVIDIA GPUs are seeing huge boosts to their production and this is helping to accelerate the project significantly.”