Path: utzoo!attcan!uunet!cs.utexas.edu!tut.cis.ohio-state.edu!purdue!gatech!hubcap!Eugene
From: eugene@eos.arc.nasa.gov (Eugene Miya)
Newsgroups: comp.parallel
Subject: Re: Myrias ships 64-pe system
Message-ID: <5476@hubcap.clemson.edu>
Date: 12 May 89 19:02:45 GMT
Sender: fpst@hubcap.clemson.edu
Lines: 26
Approved: parallel@hubcap.clemson.edu

>"The Myrias SPS-2 will not only scale to over 1000 processors;

Claims of scaling to many processors abound in this field.
We have heard from some sources that the RP3 project at IBM
has been cancelled or scaled back, for instance.  Scale is a
problem and much of it isn't linear.

An interesting piece of advice came to me from Steve Lundstrom
(I hope its okay I mention this, Steve).  He suggested that anyone
who proposed building such a scaleable architecture attempt to build
the largest possible system first.  You will learn a lot along the way.
Then, break the machine down and sell the parts (submachines).
I know this is possible with Butterflys and Hypercubes, but it is
really the learning along the way.  I consider this a good piece of
wisdom from Steve.

Longish signature follows "Type 'n' now"

Another gross generalization from

--eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov
  resident cynic at the Rock of Ages Home for Retired Hackers:
  "You trust the `reply' command with all those different mailers out there?"
  "If my mail does not reach you, please accept my apology."
  {ncar,decwrl,hplabs,uunet}!ames!eugene
  				Live free or die.