Date: Fri, 27 Oct 1995 09:51:09 -0700 From: David Greenman <davidg@Root.COM> To: "Ron G. Minnich" <rminnich@sarnoff.com> Cc: FREEBSD-HACKERS-L <freebsd-hackers@freebsd.org>, Larry McVoy <lm@slovax.engr.sgi.com> Subject: Re: New lmbench available (fwd) Message-ID: <199510271651.JAA00187@corbin.Root.COM> In-Reply-To: Your message of "Fri, 27 Oct 95 09:28:52 EDT." <Pine.SUN.3.91.951027092536.13429D-100000@terra>
next in thread | previous in thread | raw e-mail | index | archive | help
>actually the big surprise for me was walking the results and seeing >freebsd outrunning linux in so many areas on the 100 mhz boxes. I knew >that it was marginally faster in places but the margin this time (except >for ctx) was surprising. Also the aix ctx results are interesting: kind of >shows the advantage of single-address-space operating systems, as opposed >to the unix model. > >It's useful to show freebsd performance at the limit. But it's also >useful to show it on a plain vanilla 133 mhz box without $$$ boltons. > >BTW ttcp on freebsd on 100BT interfaces (SMC) is at about 56 Mhz. These >are neptune, i understand triton would be better. Yes, Triton works *much* better. You should be able to get full 100Mbit performance using a pair of 133Mhz Triton-based machines - I can tell you that I get nearly 80Mbits when doing TCP from a 90Mhz Triton machine. The pipeline burst cache also makes a tremendous difference, BTW... -DG
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510271651.JAA00187>