[28741] in North American Network Operators' Group
Re: Optical Crossconnects and IP
daemon@ATHENA.MIT.EDU (Michael Shields)
Sun May 14 22:10:34 2000
To: Tony Li <tli@procket.com>
Cc: Bora Akyol <akyol@pluris.com>, nanog@nanog.org,
craig@aland.bbn.com
From: Michael Shields <shields@msrl.com>
Date: 15 May 2000 02:08:34 +0000
In-Reply-To: Fletcher E Kittredge's message of "Sun, 14 May 2000 13:43:39 -0400"
Message-ID: <874s80efa5.fsf@challah.msrl.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Errors-To: owner-nanog-outgoing@merit.edu
In article <200005141743.e4EHheW09551@sss1.gwi.net>,
Fletcher E Kittredge <fkittred@sss1.gwi.net> wrote:
> http://smg.ulb.ac.be/Preprints/Fortz99_29.html
>
> mentions it was in Infocon 2000...
>
> And oh, yes, most definitely worth the read.
A very interesting paper, but it does have a key assumption that may
limit its applicability:
"The above definition of the general routing problem is equivalent
to the one used e.g. in Awduche et al. Its most controversial
feature is the assumption that we have an estimate of a demand
matrix."
The approach laid out by Fortz and Thorup is very pleasant in its
ability to show that a simple weight-based model can come close to
optimal traffic engineering. But without any way to quantify how
accurate our estimate of the demand matrix is, we cannot know whether
the projected weights are actually even close to optimal. And
considering the elasticity of demand, I would argue that an accurate
demand matrix cannot be constructed for most Internet backbones given
currently available tools and understanding.
I would be happy to be wrong about this.
--
Shields.
