network parallel decryption amateur style
Recent postings have gotten me to thinking: If we wrote a easily portable UNIX program to decrypt DES in parallel across our many machines, how fast could we go? Perhaps we could make use of the anonymous remailers to hide our cooperation. How many computers do we have? I have account on somewhere between 10 and 20 Sparcstation IIPX's which are very lightly loaded at night...someone else said they have 1500... PC gurus might want to also make versions for high-speed PC's on the net. It wouldn't even need to be very difficult, just maybe having one complex server which assigns keyranges to every person who mails in a request, and gets mailed back the range to check. It doesn't really have to be automatic, although that would be nice. What do y'all think? -(signature removed, because, this is a dangerous idea. That's why I like it ;)
Recent postings have gotten me to thinking: If we wrote a easily portable UNIX program to decrypt DES in parallel across our many machines, how fast could we go?
Using the fastest software DES implementation I know of, optimized for fast single-bit key change, it would take about a million SparcStation-years (at 100% utilization) to do an exhaustive DES search. Even assuming a order of magnitude faster than that (from a better implementation or faster common workstation hardware), that's still an awful lot of Sparc-years you'd have to get "the net" to donate. -matt
participants (2)
-
Matt Blaze
-
nobody@alumni.cco.caltech.edu