1993-08-19 - Cypherpunk DES distributed DES last post

Header Data

From: nobody@alumni.cco.caltech.edu
To: cypherpunks@toad.com
Message Hash: 685655e7e41b98d89f5b9345ad5d57e1da8d9a820e7bebec552b118fcd896604
Message ID: <9308192217.AA14930@alumni.cco.caltech.edu>
Reply To: N/A
UTC Datetime: 1993-08-19 22:20:48 UTC
Raw Date: Thu, 19 Aug 93 15:20:48 PDT

Raw message

From: nobody@alumni.cco.caltech.edu
Date: Thu, 19 Aug 93 15:20:48 PDT
To: cypherpunks@toad.com
Subject: Cypherpunk DES distributed DES last post
Message-ID: <9308192217.AA14930@alumni.cco.caltech.edu>
MIME-Version: 1.0
Content-Type: text/plain


Since the list keeper has requested, let us let the DES parallel
breaking die, for the time being at least.

I mentioned a possible Cypherpunk project to set up a network
of parallel DES de-cryptors linked by cryptographically secure
anonymous remailers to avoid obvious detection (although one could
argue all those flops in the night might speak for themselves).

A quick test shows the Sparcstation IIPX doing about 100 decryptions
per second in a typical unloaded system from a user account (these
are relatively small decryptions...we assume here we are decrypting
a text which only a small portion needs to be decrypted to reveal
its appropriate entropy measure).

Or over eight unused hours at night, in the neighborhood of 3 million
decryptions.  Let's even assume we can marshall 5000 similar machines
through our combined talents, that's 15*10^9 decryptions.  Keeping in
mind the total 56-bit keyspace is 72*10^15, we're still talking 
around 5 million nights to solve the problem (~13,000 years).

If we want to solve one system per month, it looks like we can only
handle about 38 bits worth of keyspace.  Not very impressive.
So, until the next technological leap, we can probably forget about
this.





Thread