[36574] in North American Network Operators' Group
RE: gigabit router (was Re: Getting a "portable" /19 or /20)
daemon@ATHENA.MIT.EDU (David Schwartz)
Tue Apr 10 22:33:36 2001
From: "David Schwartz" <davids@webmaster.com>
To: <alex@yuriev.com>
Cc: <nanog@merit.edu>
Date: Tue, 10 Apr 2001 18:48:53 -0700
Message-ID: <NCBBLIEPOCNJOAEKBEAKGEJGOEAA.davids@webmaster.com>
MIME-Version: 1.0
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: 7bit
In-Reply-To: <Pine.LNX.3.96.1010410200131.6377N-100000@cathy.uuworld.com>
Errors-To: owner-nanog-outgoing@merit.edu
> CEF should be called Customer Enrangement Feature. It is a very very very
> bad idea to have linecards be anything else than forwarders. They
> should not
> make any intelligent routing decisions. There should not be a
> tons of copies
> of routing table on line cards. That is what creates problems.
>
> Alex
CEF allows linecards to be forwarders. They don't make any routing
decisions, they just forward packets according to a routing table. (Routing
= deciding where packets should go, ie building a routing table. Forwarding
= sending packets to their destination, ie using a routing table.)
The reality is that having only one copy of the routing table creates an
inevitable bottleneck. For the same reasons this won't work on a regional
network, it won't work on a single router if the router is sufficiently
complex. The same techniques that work to scale the Internet as a whole work
inside a box.
Why do you think central fowarding is superior to distributed forwarding?
DS
