From: A Davidson (ajd@gpu.srv.ualberta.ca)
Date: Tue Jan 12 1999 - 20:12:08 MST
On Tue, 12 Jan 1999 CALYK@aol.com wrote:
> In a message dated 1/12/99 3:43:37 PM Central Standard Time,
> Doug.Bailey@ey.com writes:
>
> There already is something like that, caled Distributed.net, I believe there
> are 800,000 computers signed on, and its basically a small program that
> continuously runs using the cpu you're not using, so its always out of the
> way, and while you're online it trades informtation packets. I believe the
There is a huge difference between the distributed computing which
organizations such as distributed.net and seti@home do and parallel
processing, which is what the group in the press release appear to be
claiming.
Distributed Computing is useful for solving a limited subset of problems
which can be decomposed into small chunks, processed elsewhere, and the
results reasembled. It is a 'loosely coupled' system where each machine is
distinct from one another.
Parallel processing is a 'tightly coupled' arcitechture, where there is one
machine, and many processors. For network parallel processing to be a
reality, extremely high network speeds are needed, as the network acts like
a system bus in a single computer.
+-----------------------------------------------------------------+
| Aaron Davidson | ajd@gpu.srv.ualberta.ca |
| Silicon Creek Software | http://ugweb.cs.ualberta.ca/~davidson/ |
+-----------------------------------------------------------------+
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:02:48 MST