Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cwjcc!tut.cis.ohio-state.edu!rutgers!gatech!hubcap!Dennis
From: morrow@sun6.scri.fsu.edu (Dennis Morrow)
Newsgroups: comp.parallel
Subject: Matrices of computer systems
Message-ID: <5720@hubcap.clemson.edu>
Date: 9 Jun 89 11:59:46 GMT
Sender: fpst@hubcap.clemson.edu
Lines: 33
Approved: parallel@hubcap.clemson.edu

I am posting this for John Nall.  Please respond directly to him if you
have a decision-matrix sort of list of available parallel computers
which you can share with us.  THANKS!

---Dennis Morrow (morrow@sun6.scri.fsu.edu)
   Supercomputer Computations Research Institute
   Florida State University
   Tallahassee, FL 


**********************************
>From nall@scri1.scri.fsu.edu Thu Jun  8 14:49:26 1989
Return-Path: <nall@scri1.scri.fsu.edu>
Date:    Thu, 8 Jun 1989 14:50:06 EDT
From: nall@scri1.scri.fsu.edu

Subj:  Matrix of parallel systems

I think this group is getting fairly close to being able to give a
committee report to the Director, and I have been giving some thought
as to what it should look like.  As ... suggest(ed) in the last
meeting, it definitely should include a matrix of systems.  I am not
exactly sure ... but I would visualize something along
the lines of each row representing an available parallel machine, and
the columns being such things as type (SIMD, MIMD, etc), maximum number
of processors, maximum amount of memory, top theoretical performance,
top actual performance (mips and mflops), price, available software,
etc.

Is anyone aware of such a list already existing?  I hate to have to
reinvent the wheel.

John Nall (nall@scri1.scri.fsu.edu)