Date: Thu, 29 Nov 2001 12:27:37 -0800 (PST) From: Julian Elischer <julian@elischer.org> To: John Polstra <jdp@polstra.com> Cc: hackers@freebsd.org Subject: Re: Netgraph performance Message-ID: <Pine.BSF.4.21.0111291224400.5212-100000@InterJet.elischer.org> In-Reply-To: <200111292014.fATKEmI70068@vashon.polstra.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Thanks! :-) The fact that it works in high speed applications is a pleasant side effect of the fact that we tried REALLY HARD to make if low-ish overhead, but we really originally wrote it for T1 speed devices (and lower). The main design goal was to try make it easy for people to reconfigure it in ways we hadn't thought of and to make new modules to extend it. (also to avoid having a special control program for each new kind of node). On Thu, 29 Nov 2001, John Polstra wrote: > In article <Pine.BSF.4.21.0111291012500.5212-100000@InterJet.elischer.org= >, > Julian Elischer <julian@elischer.org> wrote: > > Netgraph is a prototyping tool, which has enough performance to be > > useful in non-performance-critical applications. (such as all sync > > interfaces). It is not designed for gigabit interfaces etc. >=20 > You are selling Netgraph way too short. I've been using it > intensively with gigabit interfaces, and it performs very, very well. > For my application (which involves generating and responding to a > whole bunch of network traffic) it has yielded a good 4-5 times better > performance than any other alternative I've found. >=20 > John > --=20 > John Polstra > John D. Polstra & Co., Inc. Seattle, Washington = USA > "Disappointment is a good sign of basic intelligence." -- Ch=F6gyam Tr= ungpa >=20 >=20 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0111291224400.5212-100000>