[27956] in North American Network Operators' Group
Re: The Size of OSPF Network
daemon@ATHENA.MIT.EDU (Vijay Gill)
Fri Mar 31 18:42:07 2000
Date: Fri, 31 Mar 2000 18:40:04 -0500 (EST)
From: Vijay Gill <wrath@cs.umbc.edu>
To: HANSEN CHAN <hchan@newbridge.com>
Cc: nanog list <nanog@merit.edu>
In-Reply-To: <38E512CE.94BAB47F@newbridge.com>
Message-ID: <Pine.SOL.3.95.1000331183531.12408B-100000@mailserver-ng.cs.umbc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: owner-nanog-outgoing@merit.edu
On Fri, 31 Mar 2000, HANSEN CHAN wrote:
>
> Dear all,
>
> For the big ISP networks (tier 1 or 2) that happens to use OSPF, what is
> the typical number of OSPF areas in the network? According to OSPF
> books, a typical area is consisted of 200 routers. Are those guidelines
> ever followed in real ISP network deployment?
>
> Any input is highly appreciated.
In an ISP context, given that IS-IS and OSPF are roughly equivalent in
terms of resource consumption on a theoretical level (individual
implementations may vary depending on the various data structures,
algorithms used etc. for costing models), I see no problem with running
300-400 routers in a area. Of course this depends on how stable you
expect the underlying transport fabric to be and what you use OSPF for
(e.g. ibgp-hack) and how smart you are when you go about it (i.e. a well
designed IGP structure may have 500-700 routers in one area and not have
significant issues).
For an enterprise type network with OSPF as the sole protocol and
depending on the number of adjacencies, your mileage may vary
considerably.
/vijay
