Slashdot Mirror
IBM Sponsors Humanitarian Grid Computing Project
BrianWCarver writes "Reuters reports that IBM and top scientific research organizations are joining forces in a humanitarian effort to tap the unused power of millions of computers and help solve complex social problems. Following the example of SETI@home, the project, dubbed The World Community Grid, will seek to tap the vast underutilized power of computers belonging to individuals and businesses worldwide and channel it into selected medical and environmental research programs. The first project to benefit will be Human Proteome Folding, an effort to identify the genetic structure of proteins that can cause diseases. The client is currently available for Windows XP, 2000, ME, and 98."
But isn't the Stanford Folding project already doing part of this?
I like you, Stuart. You're not like everyone else, here, at Slashdot.
would you seriously consider running a closed-source application, that is
..
a) cosuming your entire cpu resources
b) recieves instructions from the internet
c) sends back information gathered at your computer
d) has not provided any scientific value (a la seti@home)
this program could do anything! this looks like a perfect and cheap way for intelligence services to crack all those rsa keys they ever wanted.
Only morons moderate based on a sig.
Where's distributed.net? Oh yeah, and some Linux clients might be nice.
This way to the egress...
All my Windows boxes are 5+ year old crap with the cream of the crop being a PIII 600.
I have plenty of unused cycles on 4-way Sun boxes with gigs of spare RAM, though.
It would be nice if they released a client in portable C.
Do daemons dream of electric sleep()?
Suppose this effort discovers something. Just exactly who will own the patent?
Suppose it leads to the creation of a new revolutionary drug. Just exactly who will get the profits from the drug? (And who will have to travel to Canada to buy it?)
Is the cancer research they mention part of the United Devices effort or is this something different? The article confused me a bit on that count. It would be a shame to duplicate efforts.
Blaze a trail to the New World
If we were to use these millions of particularly unspecialized (in terms of computational ability) home PCs, wouldn't the cost be in pollution? You're consuming lots energy to crunch some numbers... you'd be plenty more efficient if you used some supercomputers. I think it's a good idea, but I wonder if this wouldn't cause more problems.
Our center has huge availability on Solaris and Linux platforms. At home I have Mac OS X. How can I help?
Some distributed computing projects appear benevolent, but the actual results remain the property of commercial organisations/universities and trusts and there's no guarantees that the results won't be used purely from a commercial and non-humanitarian point of view. I haven't looked into this new IBM project, but I'd like to advise people to always read the fine print in who own what when the project is completed.
In the past, I've investigated a couple of projects, that upon closer scrutiny look quite troubling. They often fail to address what the actual project is specifically, and who will profit from the results financially. Instead, their websites are full of feel good graphics, but the bucks stop at a pharmaceutical company's coffers when you look at the fine details, and there's no discussion of what the findings will be specifically used for, and by whom. In some cases, the whole issue of profit and ownership is quite smoothly whitewashed.
Si tacuisses philosophus mansisses. If you had kept quiet, you would have remained a philosopher.
Now how is this really different from IBM's project?
A skeptic might think that IBM simply want to have a foot in the door of these big anarchic distributed projects.
Despite the stunning power available to this kind of distributed computing, it is less useful than it appears. In my research area (computational biology), the effort of parallelizing an algorithm and collating the results is seldom worth the dividend in speedup. Supercomputers generally run idle at most universities, for this very reason.
Folding@home was a nice success story, and there are further applications of those models, e.g. simulations of prion aggregation (mad cow disease, Alzheimer's, etc). But (IMO) this is the exception, rather than the rule. Anyone who thinks that parallelization is a quick & easy panacea to difficult computational problems in general is living in a dream world (and I say that as a proud owner of several Macs with parallelized RISC CPUs *and* go-faster stripes).
I've lost count of the number of times I've heard these cheap parallelization ideas floated (another example is building cheap clusters out of console hardware which I reckon I first heard in 1996!). And every other month someone offers me supercomputer time... the problem is in redesigning the algorithm to work in parallel. Certain algorithms, such as MCMC, are better suited to this treatment than others.
Of course, then you have to persuade a bunch of other scientists that Your Algorithm is the most deserving, which is a political issue (but hey, if it saves those CPUs from being used for the eminently futile task of looking for bug-eyed aliens, maybe it's a good thing...)
The majority of people behind "corporate" proxies may well not be authorized to install anything of this sort on their work computers.
If you can't see the value in jet powered ants you should turn in your nerd card. - Dunbal (464142)