> Date: Fri, 21 Sep 2007 21:46:02 +1000 > From: Norberto Meijome <[EMAIL PROTECTED]> > Sender: [EMAIL PROTECTED] > > On Thu, 20 Sep 2007 23:54:49 -0400 > Richard A Steenbergen <[EMAIL PROTECTED]> wrote: > > > Honestly, FreeBSD routing code is pretty poor as far as a modern router > > goes. If you throw enough CPU at it you can brute force your way through > > plenty of things, but in the context of modern commercial routers it > > doesn't even play in the same league (even for a software-only router). > > Interesting.... what is the golden aim of software based router we should be > trying to reach? > > IMHO, comparing routing code in software vs. hardware routing, or vs a > RTOS seems a bit useless (unless the comparison is something like 'we > could do things much faster this way, but that would force us to go > down the path of hard RT OS..) Which is different to saying 'well, > this and that part of x and Y are inefficient / too expensive for the > latest cpu models." > > I'm not bagging you , but there's always the balance to be had - > something very flexible (as *BSD / Linux ) vs something very rigid > (programmatically) but very fast / scalable... i'm interested in > seeing how / whether we efficiency can be increased without losing the > flexibility (of course, @ the cost of time,etc...something needs to > give :D )
Ever run into a non-existent 'olive'? Or even a J series Juniper? Juniper put together a very impressive software based routing system that is FreeBSD based. -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: [EMAIL PROTECTED] Phone: +1 510 486-8634 Key fingerprint:059B 2DDF 031C 9BA3 14A4 EADA 927D EBB3 987B 3751
pgpehd6ywzliK.pgp
Description: PGP signature