Distributed computing is the Hot New Thing, and you may be participating without really knowing anything about it. If you’re one of the hundreds of thousands of people running SETI@Home, guess what? Yep…your computer is part of a giant distributed computer.
How does it work? Basically a whole lot of small computers (like the desktop computers you and I use) take chunks of data and process them, then send the results to a “master” computer that puts all of the processed pieces together and looks at the results. It’s sort of like having a single computer with hundreds, or thousands of CPUs in it.
SETI@Home is one of the most famous examples of distributed computing, but others are springing up. SETI@Home uses its computer resources to search data from the Arecibo radio telescope in Puerto Rico for evidence of extra-terrestrial life. But I think that there is a distributed computing project that has an even better real-world application.
Folding@Home (web site) is a new project that analyzes data from the Human Genome Project and attempts to identify the thousands of proteins that combine together to make us what we are. The folds and twists that these proteins make as they form genes make for hideously complicated mathematical problems, but that’s just what computers were made for.
To get an idea of the scope of protein folding (and without turning us all into scientists), IBM is building a supercomputer based on Deep Blue called Blue Gene. It will model protein folding, but at a cost of over $100 million. The National Institutes of Health’s Protein Structure Initiative will spend about $150 million over five years to determine the structure of some 5,000 proteins.
But perhaps the biggest effect, at least dollar-wise, is on the genomics industry. It has seen its value grow four-fold over the past three years to over $50 billion. And that’s why companies like Structural GenomiX and Syrrx are raising bucketloads of venture capital money to do protein research.
While Structural GenomiX and Syrrx are commercial companies, more philanthropic organizations are getting into the action. The Cystic Fibrosis Foundation has hired Structural GenomiX to solve the CFTR protein which, when defective, causes CF. Stanford University is also jumping into the fray. And that’s what this article is all about.
The Folding@Home software is a screensaver that graphically shows the protein modeling that the software is performing. It’s a mesmerizing display of spheres (representing atoms) and tubes (representing the atomic bonds) that plays out over and over again on your display. And while looking kind of cool, it’s also doing real science, sending its results back to Stanford for their scientists to study.
The web site allows you to download the client, gives background on the project and provides information on applications of the work that you are helping to complete.
The software runs graphically on Windows 95, 98, NT, 2000 and ME. There is also a command line version without the graphics for those versions of Windows, as well as Linux and Solaris. According to the web site, other operating systems will be supported soon.
Since this is a relatively new project, the web page isn’t as highly developed as, say, SETI@home’s site, but there are some interesting features well worth checking out, even if you don’t choose to run the software.
So if you have extra processor cycles that you aren’t using and you’d like to contribute to a project that can potentially make a significant difference to society, Folding@home may be just what you’re looking for.