Date: Sun, 2 Mar 1997 13:39:09 -0700 (MST) From: Marc Slemko <marcs@znep.com> To: Charles Henrich <henrich@crh.cl.msu.edu> Cc: freebsd-chat@FreeBSD.ORG Subject: Re: RSA 56-bit key challenge Message-ID: <Pine.BSF.3.95.970302133526.3488G-100000@alive.znep.com> In-Reply-To: <199703021942.OAA10009@crh.cl.msu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 2 Mar 1997, Charles Henrich wrote: > In lists.freebsd.chat you write: > > >On Sun, 2 Mar 1997, Jonathan M. Bresler wrote: > >> > >> hmm....5x86-133 uses 5 minutes to do 20Mkeys. a factor of 300 > >> faster....there are faster machines out there yet. remember the > >> 6000(?) cpu intel box that the gov't bought. > > > You mean the Paragon? Massively-parallel computing would be > >ideally suited for this type of job. Each CPU grabs a chunk of the > >keyspace and then works on it totally independently of all the others. > >If you have a 4096-node system, with each CPU only capable of 100,000 > >keys/sec, you still end up with 400 million keys/sec. That would be a > >match for the world-wide effort under way with the genx.net server. > > No, he's talking about the Sandia National labs 9000+ PPRO 200 system for > simulating nuclear detonations. 9000*350K/sec is 3150 MKeys/sec, they could > break this in no time :) AFAIK, it has 7624 PPRO 200s. It is a massively parallel distributed memory architecture and only cost $55 million US. Anyone want to take up a collection for one? Guess FreeBSD's SMP code isn't up to this sort of architecture though... ISTR an article I read on it suggesting that something like it could be a good solution for companies wanting web servers!?! I love the press; everything has to relate to the web somehow...
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95.970302133526.3488G-100000>