john-users - Re: JTR multicore vs singlecore

Openwall Project		/home Owl JtR Pro crypt pam_passwdqc tcb phpass scanlogd popa3d msulogin / Linux BIND / advisories presentations / services donations / wordlists passwords / community lists wiki CVSweb mirrors signatures
bringing security into open environments

Order Openwall Wordlists CD (20+ languages) with delivery worldwide or download

[<prev] [next>] [<thread-prev] [thread-next>] [month] [year] [list]

Date: Sun, 3 Feb 2008 13:45:41 -0700
From: RB <aoz.syn@...il.com>
To: john-users@...ts.openwall.com
Subject: Re: JTR multicore vs singlecore

> (If the translation to my language is correct), why use 2 (or more)
> cores if it's not faster ? :-/
It's been a while since I ran a cluster for mpi-john, but IIRC the incremental mode was much faster. Indeed, one MPI patch I tried years ago wouldn't do anything _but_ incremental mode.

Looking back at some benchmarks I still have the numbers for, I went from ~800k c/s DES on a single PPC970 (1.6GHz) core to over 4.8m c/s with two 970+ (2.0 GHz) and two 970s - a JS21 and a JS20. I don't know how accurate the benchmarking was, as I didn't audit the code, but was able to regularly exhaust the keyspace used by pwdump6 (LM?) in under 5 hours.

Please forgive any inaccuracies - it's been nearly a year.

RB

-- 
To unsubscribe, e-mail john-users-unsubscribe@...ts.openwall.com and reply
to the automated confirmation request that will be sent to you.