[46703] in North American Network Operators' Group
Re: Load balancing in routers
daemon@ATHENA.MIT.EDU (Donn Lasher)
Mon Apr 8 18:10:04 2002
Message-Id: <5.1.0.14.2.20020408150648.00b685c0@mail.clearskynet.net>
Date: Mon, 08 Apr 2002 15:09:50 -0700
To: NANOG <nanog@merit.edu>
From: Donn Lasher <dlasher@clearskynet.net>
In-Reply-To: <B8D711B2.1987%joe@via.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Errors-To: owner-nanog-outgoing@merit.edu
>Paul's statement about CEF is interesting. It's probably the first public
>statement I've ever heard where someone was praising CEF. Usually
>discussions about CEF are accompanied by liberal amounts of swearing...
Just my anecdotal 2 cents:
2@Cisco 4000M, 4 T1's, load balanced. 5 meg of IP traffic, no filters.
no ip cef - 90% CPU - router on its' knees
ip cef - 40% router alive and kicking
Other than the memery overhead CEF adds to BGP, CEF is painless, and
worthwhile.
