Date: Fri, 10 Feb 95 8:55:57 MST From: terry@cs.weber.edu (Terry Lambert) To: kuku@gilberto.physik.rwth-aachen.de Cc: jhay@mikom.csir.co.za, freebsd-questions@freefall.cdrom.com Subject: Re: enet throughput Message-ID: <9502101555.AA11184@cs.weber.edu> In-Reply-To: <199502101027.LAA25688@gilberto.physik.rwth-aachen.de> from "Christoph Kukulies" at Feb 10, 95 11:27:26 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > John Hay -- jhay@mikom.csir.co.za > > > > > > freebsd enet performance doesn't look too good down here. > > > from freebsd -> irix i see 700 kbytes/sec. This is using 3c509s, isa bus, > > > p90 systems, the 12/22/94 snap. > > > > > > from freebsd -> freebsd i see 200 kbytes/second. Linux on similar boxes, > > > same cards, sees 980 according to a friend. > ^^^ > I believe linux figures are cheating because the linux fs cache. > I wonder how linux looks when you transfer a really large file (>> physical > memory). Of course you really meant to say "asynchronus writes in violation of the NFS specification" instead of "fs cache", since BSD has an equivalent or better VM caching mechanism. Yes, async writes are something which ought to be allowable, but not on by default because of the unreliability and server reset intolerance they cause. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9502101555.AA11184>