Distributed computing -- sometimes called grid computing -- is very worldchanging. A bunch of PCs can chew on pieces of a big computing problem in their spare time, coming up with answers in the aggregate faster than most supercomputers. SETI@Home is the most famous of these projects, but now IBM has launched World Community Grid as a mechanism for undertaking large-scale computing projects in the public interest:
World Community Grid's mission is to create the world's largest public computing grid to tackle projects that benefit humanity.Our work has developed the technical infrastructure that serves as the grid's foundation for scientific research. Our success depends upon individuals collectively contributing their unused computer time to change the world for the better.
IBM and United Devices, the company providing the grid technology, have already had one successful project come from their efforts: smallpox drug discovery. Distributed contributors ran through 35 million potential drug molecules and found 44 strong candidates, doing in months what would have normally taken years, even decades.
The World Community Grid is now focusing on the Human Proteome Folding project, seeking to understand how protein shape relates to protein activity. Figuring out protein folding -- which is a really hard problem -- would lead to better cures for a wide array of diseases and disorders. The Institute for Systems Biology, which is coordinating the project, has a page full of useful information here.
The main downside with WCG is that the United Devices software they use is closed source and Windows-only. BOINC (Berkeley Open Infrastructure for Network Computing), the engine running SETI@Home, is free/open source software and runs on a wide array of platforms, including Linux and MacOS X. The projects now underway for BOINC users include climate prediction, astronomical searches for pulsars, and (lo and behold!) an effort to understand protein structure.
Comments (8)
I really do suggest supporting an open source architecture over a proprietary one, especially since it's not multi-platform yet (if ever).
BOINC it is, then.
Posted by Mikhail Capone | November 16, 2004 2:48 PM
Posted on November 16, 2004 14:48
There is also no indication of whom will profit from the information. IBM it looks like was contracted by the Institute for Systems Biology, there is no specific indication on their website whether or not they will sell this information or "open source" it. Though ISB is a non-profit i don't think that prevents them from not selling the information.
Posted by Alex Wenckus | November 16, 2004 4:55 PM
Posted on November 16, 2004 16:55
Excellent point, Alex.
Posted by Jamais Cascio | November 16, 2004 5:18 PM
Posted on November 16, 2004 17:18
IBM, Sun, and other members of the Global Grid Forum look to profit off of this through hardware and eventually software sales. A lot of the current stuff they are behind uses tools like the Globus toolkit to provide essentially a somewhat better distributed computing paradigm for existing clisters and groups of clustered machines. Not very worldchanging at all really; sort of like distributed web services on steroids.
BOINC and other distributed computing projects are fundamentally different.
Do not confuse "grid computing" with general distributed computing or decentralized peer to peer computing; the words "grid computing" are now taking on a specific meaning which I feel is going to end up totally eclipsed by the other forms of distributed computing.
Of course, I am a bit biased, as this is the main area of my research.
Posted by Howard | November 16, 2004 9:32 PM
Posted on November 16, 2004 21:32
I should add that I don't mean to imply that the WCG is not a worthy project or one that will reach a lot of people. I just wanted to point out that grid computing is a specific term relating to a subset of distributed computing, and there is a lot more exciting research going on in distributed computing than that represented by the Global Grid Forum, which is essentially the nexus of grid computing research.
Posted by Howard | November 16, 2004 9:49 PM
Posted on November 16, 2004 21:49
Thanks for the insights, Howard.
Posted by Jamais Cascio | November 16, 2004 11:10 PM
Posted on November 16, 2004 23:10
Like Alex pointed out, I for one would like to know exactly who reaps the rewards of this?
The information in the smallpox drug case was given to US Dept. of Defense for further evaluation. Shouldn't the information obtained via WCG be free for anyone, without restrictions, since the project should "benefit humanity"? Why were the results not given for a larger number of institutions for evaluation? Why can't _I_, for example, download the potential molecules they found?
That's quite a lot of unanswered questions.
There must be some kind of system in place which guarantee that if the masses donate their efforts, the masses shall get the rewards as well (and not have to pay for them again by buying them).
Posted by Jani N | November 21, 2004 6:43 AM
Posted on November 21, 2004 06:43
It's a bit shady how they don't really make any mention of where the information ends up.
I want to know that before I'd sign up.
Posted by Uncle B | November 24, 2004 12:53 PM
Posted on November 24, 2004 12:53