Here are some snippets. Sorry for the length...but at least its not code.
Supercomputing '@Home' Paying Off for Other Research
SETI, will reach a milestone: it will have spent a million years of computer time sifting the electromagnetic noise emanating from the sky for a sign that someone or something is trying to get in touch.
An impromptu grass-roots supercomputer is inspiring other researchers to turn to the masses
Talks about Folding@home and Genome@home, Evolution@home and FightAIDS @home and “other projects (without the obligatory @'s in their names)”
Last year Dr. Pande's research group set a record by using its volunteer network to simulate 38 microseconds of the folding of a snippet of protein called the beta-hairpin.
In November, a different kind of milestone was reached when Gimps, for Great Internet Mersenne Prime Search, found the largest known number that has no factors.
When you lean back from your computer for a moment, the Pentium [yeah, oh well
] inside continues churning at a rate of hundreds of millions of times a second. To computer scientists, that is as wasteful as leaving your car idling while you run into the store. Hence the appeals to donate these "spare processing cycles" to the scientific charity of your choice.
"People have been locked into this supercomputing mentality," Dr. David P. Anderson, the SETI@home project director, said. "They want to have some gigantic thing in a box somewhere. I think that approach will ultimately go the way of the dinosaurs." Distributed computing has another advantage: unlike an ordinary computer, the grass-roots networks automatically upgrade themselves as people succumb to the desire for faster machines.
"A lot of people never believed the kind of calculations we do could be chopped up this way," Dr. Pande said. Since then the team has been working to simulate longer folding times. The ultimate goal — nowhere in sight — is to start with the description of an amino acid sequence and predict the final shape of the protein.
So many laboratories are asking for spare cycles that computer users must pick and choose lest they bog down their processors combating H.I.V., anthrax, smallpox, Ebola, multiple sclerosis, tuberous sclerosis complex and various cancers and neuromuscular diseases. (A list is available at
www.aspenleaf.com /distributed.)
Another of the earliest projects, called distributed.net, also cracks cryptographic codes and is trying to find something called an optimal golomb ruler, a mathematical artifact important to coding and communications theory.
One vital ingredient is arresting animation — gyrating proteins, mountain ranges of SETI data being analyzed. These sophisticated cartoons (known in the trade as cool graphics) sap processor power that could go toward analyzing the data. But they keep computer owners motivated by giving them a sense that something interesting is going on.
Top
