Yes, as a matter of fact, I think the topic has come up before. I
believe some folks in Physics raised the question a few years ago.
As you may have noticed, the MSU microlabs differ from many other
public labs, in that they reboot with a clean system every time.
Personally, I used to grouse about the time it takes to boot an
MSU microlab computer. I officially stopped grousing last fall,
when Blaster and Welchia hit.
In fact, when I suspect an e-mail has a virus in it, I now go to
a microlab to confirm or deny.
We encourage users of microlab computers to log out or power down
the computer when done, so that their identity is not compromised.
Bottom line is you'd have to figure out a way to do this that works
within a framework of highly secure computing the labs offer.
/rich
PS -- you may have seen the news item that an MSU grad student
participated in finding the largest known Mersenne prime number using
a grid computing tool such as you propose.
>The campus has several computer labs with a few hundred machines loaded with
> software images from ACNS. Has anyone persued the question of using idle
>cycles
> for distributed computing applications? The way they are setup the machines
>are
> on most of the time, but not really doing much of anything. Reading web pages
> or typing Word documents leaves a lot of processors doing nothing useful.
>
>Just wondering.
>
>Don Bosman
>Information Technologist
>Michigan State University, Libraries
>100 Library
>East Lansing, MI 48824-1048
>517 432-6123 ext 233
>[log in to unmask]
|