Date: Tue, 20 May 2003 14:58:34 +0200 (CEST) From: Heiko Schaefer <hschaefer@fto.de> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: current@freebsd.org Subject: Re: gbde performance question Message-ID: <20030520144927.I60060@daneel.foundation.hs> In-Reply-To: <32161.1053433748@critter.freebsd.dk> References: <32161.1053433748@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
> >but i still don't see why the processes that are named gbde-something > >would take up _that_ much cpu time ... does that look reasonable and > >explicable to you? > > The crypto work _does_ take time. I am only just starting to measure > how much for my GBDE paper now, so I don't really have anything > to compare your numbers with. well, from what i read, any modern cpu should be capable to do raw aes in the order of magnitude of hundreds of MB/s. that's why i wonder if gbde does _that_ much more than just nubercrunching on aes. or could the aes implementation be ineffecient compared to those that i read about (see link in my original mail) ? or have more rounds, keylength, or anything of that kind ... i'm looking for any reason to account for what i imagine is a factor of roughly 10x against what i think raw aes on bulk data should take. this generally puzzles me - but particularly in the light of the hardware purchase that i will almost definitely make to get rid of these flipping bits. i've had enough. and if gbde has any sane reason to take as much cpu as it does, i'd be tempted to go close to intel's top cpu models. even though i'd prefer no to :) regards, Heiko -- Free Software. Why put up with inferior code and antisocial corporations? http://www.gnu.org/philosophy/why-free.html
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030520144927.I60060>