Salon Come together, right now, over P2P
"Nelson Minar built a moral challenge into the distributed computing software developed by his company Popular Power. Users who agree to let their computers join the start-up's peer-to-peer network have a choice: They can earn about $5 per month by selling their CPU power to businesses; or they can donate their shiny new Apple G4 Cube to nonprofit tasks, receiving nothing but the warm fuzzies that result from doing good deeds.

But there's a twist. No matter whether users choose to be greedy, benevolent or something in between, they'll still end up contributing processing power to both for-profit and nonprofit clients. Minar designed the system so that even libertarian Ayn Rand fans who click the "maximize payment" option and utterly disdain charity will still end up donating at least some time and effort to helping scientists find a better flu vaccine. And vice versa."

redux [10.09.00]
FightAids@Home Overview
"Now more than ever, you are needed to help in the fight against AIDS. In the mid 1980's HIV infections reached epic proportions and have since continued to rise at alarming rates. Nearly twenty years later, technology has reached a point where you can make a difference by contributing the idle processing time of your PC.

Here is how it works:
To support FightAIDS@Home, you simply download a free software program from Entropia that runs "in the background" on your computer. FightAIDS@Home processes information (in this case using AutoDock) and calculates prospective targets for drug discovery. Basically, what this means is that FightAIDS@Home uses idle processor cycles that would normally be wasted. FightAIDS@Home captures the otherwise wasted cycles of your PC and applies them to model the evolution of drug resistance and to design the drugs necessary to fight AIDS. When your computer has finished a FightAIDS@Home computation, the FightAIDS@Home results are packed up and sent back to Entropia, ready for Scripps researchers to collect and analyze them.

Then when you are using your computer and it needs cycles, FightAIDS@Home simply and automatically turns those resources back over to the program you are using.

Download FightAIDS@Home now so your PC can start making a difference!"

ACM CrossRoads The SETI@Home Problem
"The SETI@Home problem can be thought of as a special case of the distributed computation verification problem: "given a large amount of computation divided among many computers, how can malicious participating computers be prevented from doing damage?" This is not a new problem. Distributed computation is a venerable research topic, and the idea of "selling spare CPU cycles" has been a science fiction fixture for years.

In real life, distributed computation has been used since at least the late 1980's to create "farms" of machines for rendering 3-D images. Farms allow graphic artists to create large images without needing to buy a supercomputer. More recently, the needs of scientific computation have led to the creation of frameworks such as Parallel Virtual Machine (PVM) and Beowulf, which make it easier to distribute computations across many machines. The machines involved are usually owned by the same entity and a machine is either "good" or "bad" if it is operating or malfunctioning. There are no blatantly malicious machines.

The Internet makes it possible for computation to be distributed to many more machines. However, distributing computing around the internet requires developers to consider the possibility of malicious clients."

"The general study of secure multiparty computation has produced much interesting work over the last two decades. Less well studied, unfortunately, are the tools and techniques required to move the theoretical results to the real world. The old dream of massively distributed computations is finally coming true, and yet our tools for building and analysing real systems still seem primitive. The challenge of the next few years will be to bridge this gap."

redux [08.09.00]
BBC Screensavers could save lives
"Your computer could be helping to save lives when you are not using it to play games or surf the internet.

Instead of it sitting idle, it could be taking part in scientific experiments being distributed across thousands of computers on the internet.

Drugs to beat cancer and flu are starting to be tested in simulations split up and run on personal computers that would otherwise be doing nothing useful." [via slashdot.org]

PC Magazine New Apps Exploit Connectivity
"A natural complement to distributed file-sharing capabilities is distributed computation. The idea behind distributed computation is that a really big problem gets split into discrete, independent chunks, which are then parceled out to individual computers whose owners have volunteered their idle processor time to the cause. In aggregate, the users' computers form a sort of distributed supercomputer. The concept was first popularized by U.C. Berkeley's SETI@Home project, a 1999 PC Magazine Technical Excellence finalist that's now been downloaded by more than 2 million users. Though SETI@Home is a single-purpose tool designed solely to scour radio-telescope signals for signs of extraterrestrial transmissions, you can expect to see general-purpose mechanisms for distributing all kinds of massive computations. United Devices, for example, is a company that will use distributed computing for projects in areas such as bioinformatics research, drug design, and climate studies."

redux [07.22.00]
The Standard Distributed Computing Goes Commercial
"The distributed-computing model could be one of those rare cases where capitalism and pure scientific research mesh. Not every lab can afford to pay $200,000 for an eight-processor Origin 2000 SGI supercomputer, much less $1 million for a 40-processor machine, says David Fenstermacher, director of scientific computing for the medical school at the University of North Carolina at Chapel Hill. (Fenstermacher is also acting director of the campus' Center for Bioinformatics and a United Devices adviser.) And even the most powerful supercomputers need time to process data.

A project that would take several months on a supercomputer – creating a 3D model of a protein's linear be accomplished in much less time using thousands of distributed computers"

redux [04.05.00]
Wired Researcher Borrows from Napster
"A researcher working on the Human Genome Project is using Napster technology, and he's not looking for T3 connections to download Moby.

Dr. Lincoln Stein, an associate professor of bioinformatics at the Cold Spring Harbor Lab in New York, is investigating ways to use Napster-type technology to allow scientists to share their discoveries of the genome.

"I was very interested when I saw Napster," Stein said. "It has a similar architecture (to what we use now), but it allows for 'peer-to-peer' data exchange and it dawned on me that it would be marvelous for our annotation system.""