john-users - Re: PVM or MPI?

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Fri, 23 Mar 2007 03:50:54 +0300
From: Solar Designer <solar@...nwall.com>
To: john-users@...ts.openwall.com
Subject: Re: PVM or MPI?

On Thu, Mar 22, 2007 at 11:41:46AM -0500, RB wrote:
> Not being in the HPC crowd, I personally can't see much of a
> difference - they're two implementations of one parallelization
> paradigm: thread concurrency on heterogeneous machines.

Correct.  (Disclaimer: I am also not in the HPC crowd.)

> There [was]
> an MPI implementation of John during the 1.6.x days, and it worked
> okay - I used it to some success on an 8-way P-III machine.  Looking
> around, it seems they've updated it:
> http://www.bindshell.net/tools/johntheripper.

There are several MPI hacks for John the Ripper.  I have 5 of them here:

	ftp://ftp.openwall.com/pub/projects/john/contrib/mpi/

PVM was also mentioned on some occasions, but somehow noone has used it
for John the Ripper (as far as I'm aware).

I don't think either MPI or PVM should be used for an implementation
that would be usable in practice - to actually audit password hashes
rather than just demonstrate the concept in a university environment.
This has been discussed on this mailing list before.

-- 
Alexander Peslyak <solar at openwall.com>
GPG key ID: 5B341F15  fp: B3FB 63F4 D7A3 BCCC 6F6E  FC55 A2FC 027C 5B34 1F15
http://www.openwall.com - bringing security into open computing environments

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