[Beowulf] Quasi-Non-Von-Neumann hardware in a Beowulf

Joe Landman landman at scalableinformatics.com
Thu Mar 10 17:48:42 PST 2005

Robert G. Brown wrote:

> Actually IIRC there was some effort by Intel to build something into a
> CPU or some other part of a mobo chipset, but I don't know what came of
> it.  The people who need this commercially are webvolken and encryption
> algorithms, where "unpredictable" is almost as good as "random to the
> nth bit of randomness".  Nobody uses marginally secure encryption if
> they can help it.  However, these folks are totally happy with
> kilorands/sec, let alone megarands/sec.  I need hundreds of
> megarands/sec.

(thinking aloud)

Hmmm.  I bet we could generate this (if you dont mind tt800 or its MT 
ilk), in software using some really neat and very powerful (and 
inexpensive) COTS chips.  Dont know if we could hit 0.1 GRPS (gigarand 
per second), but I bet it could come pretty close.

I have been thinking about a USB dongle for the past few months that 
provided random ints at a pretty nice rate.  The chip I have in mind for 
this might be able to be powered off the USB port also (I think, I need 
to see the overall power needs).

Could do a PCI card... would cost more to fabricate and sell.  The $30 
mark might be low.  I'd guess closer to about $100 for that, unless we 
could get significant volumes  ... (economies of scale work wonders for 
component pricing).

(/thinking aloud)

>    rgb

Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 786 8423
fax  : +1 734 786 8452
cell : +1 734 612 4615

More information about the Beowulf mailing list