Path: utzoo!mnetor!uunet!husc6!cmcl2!beta!hc!lll-winken!lll-crg.llnl.gov!brooks
From: brooks@lll-crg.llnl.gov (Eugene D. Brooks III)
Newsgroups: comp.arch
Subject: Re: Cray architecture
Message-ID: <5364@lll-winken.llnl.gov>
Date: 24 Mar 88 23:54:39 GMT
References: <7762@alice.UUCP>
Sender: usenet@lll-winken.llnl.gov
Reply-To: brooks@lll-crg.llnl.gov.UUCP (Eugene D. Brooks III)
Organization: Lawrence Livermore National Laboratory
Lines: 25

In article <7762@alice.UUCP> dmr@alice.UUCP writes:
>Speaking of Cray architectural features, has anyone noticed that the
>Y-MP comes in a dull rectangular box?  It might as well be some silly
>IBM machine (cf.  the International Style).
>
>Fortunately, thanks to upton@ole, we know that with the 3, CRI is
>returning to its roots.  I'd say the company is well rid of Chen.
>Notice who's funding his new company?  It's clear that the guy just
>wants to build conventionally-shaped computers.  I'll take a true
>Seymour design, thanks.
I would agree that the C shaped machines are more attractive, and that
Chen was probably run out of town over the square box.  IBM will certainly
be happy with Chen as all that needs to be done is paint the thing blue.

There may, however, be method to Chen's madness....

Any look in an architecture book will show a standard crossbar to be a
square object and the Y-MP is rumored to have more than one full crossbar.
I wonder what the ideal shape for an Omega network, or any of its brethren, is?

The Cray 2 seems to be running its semaphore bits and memory banks around
in a circle, the wheel is admittedly powered by a fast rodent.

							Eugene Brooks

Disclaimer: Nothing above is to be taken as anything but idle jest.