Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!husc6!m2c!umvlsi!dime!yodaiken
From: yodaiken@freal.cs.umass.edu (victor yodaiken)
Newsgroups: comp.arch
Subject: Re: parallel systems
Message-ID: <5772@dime.cs.umass.edu>
Date: 19 Oct 89 03:44:59 GMT
References: <35825@lll-winken.LLNL.GOV> <20336@princeton.Princeton.EDU> <7651@bunny.GTE.COM>
Sender: news@dime.cs.umass.edu
Reply-To: yodaiken@freal.cs.umass.edu (victor yodaiken)
Organization: University of Massachusetts, Amherst
Lines: 43

In article <7651@bunny.GTE.COM> hhd0@GTE.COM (Horace Dediu) writes:
>Consider the 8k processor NCUBE 2--"The World's Fastest Computer." 
>(yes, one of those).  According to their literature:
>"8,192 64 bit processors each equivalent to one VAX 780.  It delivers
>60 billion instructions per second, 27 billion scalar FLOPS, exceeding the
>performance of any other currently available or recently announced
>supercomputer."  It's distributed memory .5MB per processor, runs UNIX, 
>and is a hypercube.
>
>I don't know the price, but I bet it's less than a Cray. 

Like to see the delivered price of a 8k processor system.

>Interesting to
>talk about GigaFLOPS.  This is fast.
>
This sounds like one of those total b.s. measures obtained by
multiplying the number of processors by the max mips/mflops rate
per processor. 
>
>Who cares about shared memory?  Distributed is the only way to scale.
>Everybody realizes this since it can be proven.

Proof citation? Sketch?  

There is a lot of mythologizing about parallelism. Parallel processing
is a standard technique for speed which is used in every carry-lookahead
adder, every bus, etc. It seems reasonable to believe that parallelism
will be an important technique in the future. It seems POSSIBLE that 
using multiple cpu's will be a useful technique. On the other hand
there is no reason why this technique must work, and it seems at least
as possible that cpu's should not be the basic unit of parallel
computation. 

>It's no secret that every big iron maker and every supercomputer shop is
>developing parallel machines.  These are still modest efforts (<100 cpu's),
>but the leading egde is now in the 10k coarse grained, 64k fine grained
>processors.  This should scale nicely to 1M processors in the next decade.
>After that we can expect some kind of new barriers to come up.  

I admire your confidence, but am unconvinced.  Evidence?

victor