Talkibie has covered the concept of cloud computing in relation to its effects on software and as a technology base. In this regard, cloud computing, or using a remote network of decentralized machines to complete an individual user’s request, is inherently self-serving. Recently, educational and medical groups are finding ways of utilizing the cloud in a more selfless fashion. It involves sending out from the cloud modular requests for extremely complex computations to individual user’s machines. Stanford University, in particular, has lead the charge with a program called Folding@home. The results, thus far, have been nothing short of miraculous.
The program is tackling complex computations related to the study of degenerative diseases. As Stanford’s Folding@home website explains, “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. Moreover, when proteins do not fold correctly (i.e. ‘misfold’), there can be serious consequences, including many well known diseases, such as Alzheimer’s, Mad Cow (BSE), CJD, ALS, Huntington’s, Parkinson’s disease, and many Cancers and cancer-related syndromes.” The computations required to study a single protein strand fold would normally require the processing power of a supercomputer. However, the cost of purchasing supercomputers is prohibitive for most research facilities. The solution came from a creative use of cloud computing and the belief that given an opportunity, people will help out.
In December of 2000, Vijay Pande and Michael Shirts of Stanford wrote an article for Science Magazine called, “Screensavers of the World, Unite!”. This groundbreaking article proposed that distributed computing could be “used to tackle problems which make even supercomputers quake.” Pande continued his work and with the help of Sony, ATI, Dell, and other technology companies created the Folding@home project. But Pande’s vision of distributed computing would only become reality when thousands of users volunteered their graphic processing units, PS3 game stations, and other CPUs to help in the search for cures. The Folding@home project would need hundreds of thousands of people to make this work. People like 26 year old, Matt Abdou.
An IT and application specialist from Salem, NH, Matt’s enthusiasm is evident as he describes his interest in the program. “Folding is a really cool idea,” he explains. “I first got involved in folding when SETI was still in operation. So when I got my PS3 upgrade and saw a blurb about Folding@home, I thought ‘Why not?’ I use my PS3 as a DVD player, for media like music, and for games. I really don’t ever shut it off so when I’m not home, why not put it to good use? The Folding@home project works like a screensaver. It’ll show the helix of the protein that is currently being worked on as it’s crunching the numbers. Very cool.”
The Folding@home group even has a way of tracking progress for how many computations have been completed by each user. This has created a bit of good-natured rivalry among individual participants, as each user can check and see where they rank compared to others. Playing on gamers’ inherent desire to compete and claim the number one spot in stack rankings, the Folding@home project even sees teams of users banding together to create machines that are capable of solving equations faster for better scores. Matt chuckles as he says, “I remember one time a friend of mine and I got this idea, ‘Mineral oil won’t hurt electronics, so what if we…?’ and then we built this board, immersed it in a six gallon fish tank filled with mineral oil, got a circulation pump to keep the temperature right and overclocked the heck out of it just to see how fast we could run the folding models. Geeks can be competitive.”
As a result, the Folding@home project is making history. On September 16, 2007, Folding@home passed a major milestone in computations when it was the first distributed computing network to surpass one petaFLOP. As Sony’s officials explain, “In other words, if every person on the planet were to perform a simple mathematical calculation, such as calculating a percentage, each person would have to perform 75,000 calculations every second for the world’s population to achieve a petaflop.” In October 2007, Folding@home made the Guiness World Book of Records as the world’s most powerful distributed computing network. As for the research that has been completed, the results (though extremely technical) are overwhelmingly positive.
For thousands of years, people saw the power of clouds in the form of lightning that surged from heaven to earth. It wasn’t until much later, with the advance of science, that people came to understand that the same power that came from the cloud to the ground could also go the other way. Cloud computing hurled the first bolt that changed the way people thought about the power of processing. Perhaps we are seeing the same lesson of natural phenomenon being echoed here in the realm of distributed computing with a little help from some earthbound angels.