1996-02-01 - parallel encryption

Header Data

From: Simon Spero <ses@tipper.oit.unc.edu>
To: cypherpunks@toad.com
Message Hash: 8fb93412992754161f073dc59e502419ca55b0eca312fcaa36b78de225871c25
Message ID: <Pine.SUN.3.91.960131051351.13875A-100000@tipper.oit.unc.edu>
Reply To: N/A
UTC Datetime: 1996-02-01 06:25:07 UTC
Raw Date: Thu, 1 Feb 1996 14:25:07 +0800

Raw message

From: Simon Spero <ses@tipper.oit.unc.edu>
Date: Thu, 1 Feb 1996 14:25:07 +0800
To: cypherpunks@toad.com
Subject: parallel encryption
Message-ID: <Pine.SUN.3.91.960131051351.13875A-100000@tipper.oit.unc.edu>
MIME-Version: 1.0
Content-Type: text/plain


Ok, so I've got my BeBox and so finally have an SMP of my own again; 
anyone want to suggest any cool crypto stuff that parallelises well?
Rogaways hashs look interesting, and nDES offers an obvious process 
network for pipelining, but what about things like running multiple 
interleaved CBC streams in parallel, with each stream starting off from a 
different IV? I can't think of any practical ways of speeding up a single 
RSA operation, although twice as many processors obviously gives twice 
the thruput.

Simon
------
"GM spent $3.6 billion giving birth to the Saturn, and it doesn't even go 
 supersonic." - Ben Rich, Skunk Works head (75-91)





Thread