RealTime IT News

Stanford Disease Research Effort Calls On PS3 Users

Sony announced it's going to give its PlayStation 3 (PS3) users the ability to help medical research. In cooperation with Stanford University's Folding@Home program, the entertainment giant said it will participate in a distributed computing system that already borrows compute cycles from PCs and other computer architectures.

The aim is to help support Folding@Home, a research effort trying to unlock the the causes of Parkinson's, Alzheimer's, cystic fibrosis and many cancers. The program was originally launched in 2000.

"We're applying the technology to study biomedical questions such as folding, misfolding and Alzheimer's," Vijay Pande, associate professor of chemistry at Stanford, told "We have a long track record of success with peer computing. About 2 million people have participated since we started in 2000."

Pande said about 200,000 computers contribute to its distributed-computing effort. He hopes to attract at least 10,000 PS3 users in the first month the program goes live, which is expected at the end of this month as part of a software update to PS3 customers. Participation is voluntary. PS3 users will just have to click a Folding@Home icon to participate by sharing their PS3 computer during times it's idle.

At its Web site, Folding@Home explains that proteins are "biology's workhorses -- its 'nanomachines.' Before proteins can carry out these important functions, they assemble themselves, or 'fold.' The process of protein folding, while critical and fundamental to virtually all of biology, in many ways remains a mystery."

When proteins do not fold correctly (i.e. "misfold"), there can be serious consequences, including diseases, such as Alzheimer's, Mad Cow (BSE), CJD, ALS, Huntington's, Parkinson's and many cancers and cancer-related syndromes.

Sony wanted to participate and see what the PS3 could do beyond its bread-and-butter entertainment applications.

"This is a donation," explained Noam Rimon, software development manager at Sony. "The user has to be running the machine, but otherwise there is no performance hit. If you're playing a game, Folding@Home doesn't run in the background."

Folding@Home originally started out using spare CPU cycles but has since embraced video graphics processors as a means of crunching through work units. The protein fold simulations are floating point calculations, and GPUs are one big floating point calculator.

Pande said he was particularly excited about getting PS3 users because of the power of the Cell processor that runs it. "The PS3 with the Cell is about 20 times faster than a typical PC," said Pande. "If we can get 10,000 PS3 users, that's like getting 200,000 PC users. If we reach 50,000 users we're getting to roughly the level of a petaflop , it's a huge deal."

The Folding@Home project includes a server farm of 50 servers and about 100 terabytes  of storage. Pande also said he has access to the second largest academic computer that's housed on the Stanford campus just up the hall from his office. But he describes the distributed computing system as far more cost-effective than a supercomputer.

While the number or participants varies, Pande said Folding@Home has developed software to identify reliable users and account for fluctuations in available processing.