“I can’t get this graphics card to fold!” – Snark
“I usually score cards where I want them to bend to make it easier.” –Shwaip
“…Thanks.” -Snark
Folding@home (often shortened to F@H) is a distributed computing project run by Pande Lab at Stanford University. The entire purpose of the project, very simply stated, is to study protein folding and misfolding.
Proteins are long chains of amino acids. Amino acids are the fundamental elements of biological process. Amino acids arrange in certain patterns to perform specific tasks. In order for them to carry out their specific tasks, they need to fold into a specific shape. If they do this incorrectly, or not at all, serious problems can occur. Alzheimer’s disease, cystic fibrosis, Mad Cow disease, inherited emphysema, and some cancers are attributed to misfolded proteins.
Now for the really tricky part: proteins fold incredibly fast. How fast? 10,000 nanoseconds. This is so fast that it would take several decades for a typical computer to calculate the predicted way that a single protein would fold. That’s why more firepower was needed to tackle these kinds of calculations. By running this project through distributed computing it allows a giant pool of computers to run small portions of an extremely large project, and then report its results back to a central location where all of the results can be combined into a complete set. The project has been running for roughly a decade, producing 73 peer-reviewed articles, which works out to one full article every 1.6 months.
The way the project works on an individual level (and that individual that can be you), is that one can download one of the many clients (small programs that run on your PC) from the F@H download page, and run it all the time to contribute small chunks to the research. There are clients for Windows, Linux, Mac, and PlayStation 3; there is even a different download page for high performance clients. Through the years there have been single core CPU clients, multi core CPU clients(SMP, SMP2), and graphics card clients(GPU, GPU2, GPU3). On top of that, Pande Lab has produced numerous computational engines that run within the clients that have evolved in the manner in which the math is done; burst computation is one of the last big changes to the way computations can be done. Pande Lab is always trying to improve the way that clients work with specific hardware, so that results are generated faster.
The real-time payback for participating in Folding@home is the accrual of points. When you run the client, it asks you for a username and team number in which to attribute the completed work to. This allows people to compete against each other in a points ranking system—and when people form teams, it allows for groups to participate in the same manner globally.
Many computer hardware companies have their own teams, and so do a large number of online communities. There is a lot of high drama in this friendly competition as well: teams passing each other can take years of dedication and still longer to make sure that position is not lost. The same time scale can be applied to individual users when they approach the top 100, mostly because the majority of these users have such extremely high daily production numbers—and yes, it is partly an ego and pride thing.
There has always been some question about the validity of the F@H program because of the teraflops of computations being done, along with the massive amount of total electrical consumption. Some claim that money spent by individuals on equipment and electricity for Folding could be used for funding other research groups and causes. This is something that, of course, cannot be answered with a resounding “Yes, this is the best thing”. People who contribute to this endeavor are called donors because they are donating computer power, and thus some feel it is equal to giving to a charity—and giving to any charity has always been a personal choice based on feelings and your own personal interest. The people who have been doing this for a long time believe in it wholeheartedly and what it is attempting to accomplish.
Folding@home distributed computing is not the only project like this out there. If the concept intrigues you, look around. SETI@home (Search for Extraterrestrial Intelligence), for example, is another very popular program. However, if you are interested in the F@H project, make sure to check out Icrontic’s dedicated folding discussion forum where we have strung some information together for you. Icrontic is the home of one of the longest standing teams on the internet, and for quite some time we were one of top 10 teams in the world.
“Foldin’, foldin’, foldin’… get those rigs a foldin’!”
-_k_, F@H Team 93 leader
Articles RSS