Distributed computing is all the rage and some of the scientific projects underway look like they might well enable profound advances. The big question is how do we make it worthwhile for people and organizations to donate computer time?
We’ll get to that in a second but first let’s look at some of the leading distributed computing projects such as Seti@home, Folding@home, seventeenorbust.com, and Einstein@home …
Seti@home is an attempt to find intelligent life on other planets by crunching radio telescope data (so far the project has consumed 2,241,197.632 years of processing time). Intriguing but so far no bug eyed aliens.
Next, Folding@home. This is “a distributed computing project which studies protein folding, misfolding, aggregation, and related diseases. We use novel computational methods and large scale distributed computing, to simulate timescales thousands to millions of times longer than previously achieved. This has allowed us to simulate folding for the first time, and to now direct our approach to examine folding related disease.” To date this has been a hugely successful project.
And there’s seventeenorbust.com (one of our favorites), a distributed attack on the Sierpinski problem. This problem deals with Proth numbers which are numbers of the form N = k * 2^n + 1, for any odd k and n > 1. As the site notes: “If, for some specific value of k, every possible choice of n results in a composite (non-prime) Proth number N, then that k is called a Sierpinski number. The Sierpinski problem itself is: ‘What is the smallest Sierpinski number?’”
In January the seventeenorbust team announced “28433 * 2^7830457 + 1 is prime!$#!@!! It weighs in at 2,357,207 digits, sliding in as the fourth-largest known.” This project will create somewhat less practical results than folding@home but its importance in furthering mathematics is not trivial.
One of the latest distributed computing projects is “Einstein@home, a search for spinning neutron stars otherwise called pulsars. The Einstein@home project will use data from the LIGO and GEO gravitational wave detectors. According to the home page “March 5, 2005 … In the two weeks since our public launch, Einstein@Home has gone from about 6,000 users (with credit) to over 20,000, and is currently adding users at the rate of about 1,000/day.“
All of these projects are wonderful examples of community effort and definitely worth the IT industry giving up some of the spare computing capacity that we have remarkable amounts of.
Here’s an idea: Perhaps we should encourage this kind of altruism by granting tax credits to entities donating computer cycles. This would create a justification for all the currently wasted energy expended running idle computers. Such a scheme would be based upon a notional End User Work Unit (EWU) that would be some number of floating point operations, say, 10 billion FLOPs, worth a total of something like $0.50 or $1.
A project bidding for computing resources from the National Voluntary Distributed Computing Infrastructure (NVDCI) would have their computing requirements reviewed to determine how many EWUs were required to process a Project Work Unit (PWU). A system of key exchanges would be used to validate the projects serviced and the EWUs expended and the donator of the computer time would receive an annual statement that would summarize the contribution which would then be multiplied by the going rate for an EWU and included on tax returns.
Voila! Everyone wins. Science gets a free resource, business gets a tax break, and the world gets fundamental science done and maybe even cures for some of our most dangerous diseases.
