UC students help create world's biggest supercomputer

Daniel Corcoran is looking for a few good University of Cincinnati students — or their computers, at least — to help fight COVID-19.

Corcoran, a computer science major in UC’s College of Engineering and Applied Science, is part of an international effort to turn everyday computers into a networked supercomputer to tackle public research projects.

Called Folding@Home, the project dedicates a fraction of the processing power in ordinary off-the-shelf laptop and desktop computers to help with research projects. Medical researchers around the world are using the power of this network to help understand the way proteins in pathogens like coronavirus form and bind with human cells.

This research needs to process huge volumes of data quickly, which is a challenge, Corcoran said.

Proteins develop or “fold” in unique and sometimes unpredictable ways. Researchers can capture the static structure of a protein, but it doesn’t provide much information since it’s always changing.

“It’s like if you had one frame of a movie and tried to infer what the movie was about,” Corcoran said.

Computer simulations allow researchers to understand how drugs might bind to proteins and how they could be used therapeutically.

The project is a collaboration of more than 700,000 computers and growing around the world.

Folding@Home has contributed to research on breast cancer, Alzheimer’s disease, Parkinson’s disease and other viral infections such as Zika and hepatitis C. This year, organizers of Folding@Home are prioritizing COVID-19 research projects so they have the processing power needed to run simulations in search of vaccines and treatments.

“In the past 20 years, they’ve facilitated the research of 220 or 230 published papers,” Corcoran said. “Now the focus is on COVID-19. That’s why we’re encouraging students to get involved now.”

UC physics professor Michael Sokoloff is the faculty adviser for UC’s Folding@Home effort.

“The amount of processing power committed to this means they are running more cycles throughout the world in real time than otherwise would be available on the top supercomputers,” Sokoloff said.

How much processing power?

The typical gaming console like the PlayStation 4 has 1.84 teraflops, or the capability of a processor to calculate one trillion floating-point operations (flops). Folding@Home has a combined 2.4 exaflops, allowing for 2.4 billion billion (i.e., a quintillion) calculations per second. And international labs like CERN and the Oak Ridge National Laboratory in April dedicated more computer cores to the project, boosting its capacity.

“There are gazillions of these possible folding configurations,” Sokoloff said. “And the simulation sends out similar but slightly different configurations and collects the results. And then there’s another set of programs to digest all the calculations.”

Featured image at top: UC computer science student Daniel Corcoran studies from home during the pandemic. Photo/Provided