Date: Thu, 30 May 2002 17:21:17 +0200 From: Andre Oppermann <oppermann@pipeline.ch> To: Richard A Steenbergen <ras@e-gerbil.net> Cc: "Louis A. Mamakos" <louie@TransSys.COM>, Attila Nagy <bra@fsn.hu>, Luigi Iannone <Luigi.Iannone@lip6.fr>, freebsd-net@FreeBSD.ORG Subject: Re: MPLS Message-ID: <3CF6436D.ADA51D7F@pipeline.ch> References: <Pine.LNX.4.44.0205291108080.7798-100000@scribble.fsn.hu> <3CF4A64A.EE220611@pipeline.ch> <200205291413.g4TEDLRG075458@whizzo.transsys.com> <3CF4E483.2510639@pipeline.ch> <200205291522.g4TFMdRG076033@whizzo.transsys.com> <3CF4FCFC.3D760508@pipeline.ch> <20020529180204.GK33611@overlord.e-gerbil.net> <3CF51E7C.E9A47960@pipeline.ch> <20020529233411.GO33611@overlord.e-gerbil.net> <3CF5F4ED.369BCE83@pipeline.ch> <20020530150612.GP33611@overlord.e-gerbil.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Richard A Steenbergen wrote: > On Thu, May 30, 2002 at 11:46:21AM +0200, Andre Oppermann wrote: > > The theorie about the LC-Trie is that it'll fit into L2 cache for the > > entire default-free forwarding table. > > Versus the reality of doing bit operations instead of byte operations. In > my tests, the mtrie was always faster, even on a celeron with a piddly L2 > cache. We'll find out. We're designing a framework into which different kinds of tables can be implemented. We can move on to profiling then. > > This might work on a decicated router system. FreeBSD is being used > > for servers as well. > > Please explain how that would not work for servers? It would work but not optimal because the packet flow is different for locally terminated/generated packets. > > Also to not to change the userland/kernel interface wrt routing we have > > to keep all prefixes in the kernel RIB. This means we have to be able to > > do longest prefix matching. > > This is a non sequitur. All routes will be available through the kernel > RIB, but for exact matches only. When is a longest prefix match needed > there? When the routing daemon instructs us to remove the prefix 10.0.0.0/8 when we also have 10.0.0.0/9 and 10.128.0.0/9. > > Anyway, I'm confident we'll come up with something that is well > > balanced and much faster and more memory conserving than what we > > have today with the patricia trie. (A default-free kernel RIB consumes > > approx. 30MB of kernel memory in 4-STABLE/5-CURRENT at the moment). > > Not to mention the outrageous amounts of memory consumed by the caching > mechanism. Where? Do you mean rt_metrics? > Oh, and it should be able to support multiple nexthops per prefix, and > load balance across them. I think even Linux has this support now, and an > actual FIB. Yes, and policy routing... -- Andre To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3CF6436D.ADA51D7F>