Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!uwm.edu!ux1.cso.uiuc.edu!csrd.uiuc.edu!s41.csrd.uiuc.edu!eijkhout
From: eijkhout@s41.csrd.uiuc.edu (Victor Eijkhout)
Newsgroups: comp.misc
Subject: Re: 5 billion computations per second!!!
Message-ID: <1991Mar26.015138.28898@csrd.uiuc.edu>
Date: 26 Mar 91 01:51:38 GMT
References: <2480@umriscc.isc.umr.edu>
Sender: news@csrd.uiuc.edu (news)
Organization: UIUC Center for Supercomputing Research and Development
Lines: 24

mcastle@mcs213e.cs.umr.edu (Mike Castle {Nexus}) writes:

>Was listening to National Public Radio earlier, and they had a little piece
>on computers.  They mentioned some sort of tests run at Los Alamos something
>or the other.  They said the fastest machine ran there was one made by 
>Thinking Machines, Corporation, and turned in a result of 5 BILLION computations
>per second. 

Thinking Machines in Boston make the Connection Machine,
which is (they claim) a 64K processor machine.
However, every processor is a bit-processor, and there are
16 of those on a single custom chip, so it is closer to
the truth to say it's a 4K processor machine, with a 12-dimensional
hypercube architecture. And their latest model has
a floating point accelerator per node so you have 4K
fp chips. Now, if you take an 'algorithm' that has hardly
any communication (vector plus vector, or so)
5Gflop is easy to believe. I seem to remember from 
a few years back that they can do inner products
at 20Gflop, and slightly less trivial algorithms such
as a Conjugate Gradient method at 5Gflop. (flop is
floating point op per sec.).

Victor.