Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 25 Jul 1999 14:07:23 -0700
From:      Summoner <summoner@uswest.net>
To:        James Howard <howardjp@wam.umd.edu>
Cc:        Phil Regnauld <regnauld@ftf.net>, freebsd-chat@FreeBSD.ORG
Subject:   Re: Unisys backing NT
Message-ID:  <379B7C8B.3DCAF146@uswest.net>
References:  <Pine.OSF.3.95q.990725122806.11440A-100000@tracy.umd.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
James Howard wrote:
> On Sun, 25 Jul 1999, Summoner wrote:
>> http://www.marketplace.unisys.com/ent/cmparch.pdf
>>
>> To put it simply, a CMP system is a (2x2)x2x8 32-way cluster-in-a-box
>> with lots of inter-machine shared memory, 96 PCI slots, the ability to
>> run 8 OSes at once, and a crossbar switch that would give any hardware
>> junkie wet dreams.
>>
>> I want two.  :-)
> 
> Does FreeBSD run on it?  Would it be useful for RSA cracking? :)

Cracking keys?  It's proven that any civilian encryption system,
given enough CPU cycles, can be broken.  It's a waste of time IMO,
but then it's your electricity bill, not mine.  I think something
like GIMPS is far more productive (and lucritive).

Unisys's changes in the architecture design are supposedly invisible
to the OS running on each sub-pod.  It's based on Intel Xeon CPUs, so
it's at least theoretically possible that FreeBSD would run on it.

The hype-pages do say that Unisys has been working closly with MS on
this one to get NT working on it, so they may very well be using a
special kernel (like what SGI did with the 320 and 540).  If that's
the case, then it's likely FreeBSD compatibility is making a sprint
for the nearest window.  But I'm no kernel hacker, so I have little
idea of what would break "our" kernel.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?379B7C8B.3DCAF146>