Newsgroups: comp.parallel.pvm
From: gmj@crab.pha.jhu.edu (Geir Magnusson)
Subject: Re: PVM under Windows?
Organization: JHU Dept Physics
Date: 29 Aug 1994 18:15:34 GMT
Message-ID: <gmj-290894142102@windward.pha.jhu.edu>

In article <aorndorff.257.2E59EAE3@nhtsa.dot.gov>, aorndorff@nhtsa.dot.gov
(Andrew Orndorff) wrote:

[big snip]

> 
> Well, assuming that you have a _large_ base of Windows PC's, and that
> Windows has at least minimal multi-tasking capabilities (created child 
> processes, etc) there could be a lot of free CPU to use.
> 
> Add in something like Winsock TCP/IP support (for PVM) and Netware/VINES/
> LAN Server on each machine so that common executables and user data
> space are available, and it's pretty close to having a true workstation
> cluster.
> 
> It's one way of doing things. Same reason an OS/2 version of PVM could be
> a nice solution for some people (yes, I know one exists), or an NT version, 
> etc.

Why the insistance of using something as crummy and slow as Win3.11?  Why
not run a minimal implementation of Linux on a small partition and use
that?  It's also free...

(Or did someone suggest this already?)

geir


Geir Magnusson Jr.                       while (ms--)
Dept of Physics and Astronomy            {
Johns Hopkins University                     linux++;
gmj@crab.pha.jhu.edu                     }

