[138984] in North American Network Operators' Group
Re: Regional AS model
daemon@ATHENA.MIT.EDU (Graham Wooden)
Thu Mar 24 17:51:12 2011
Date: Thu, 24 Mar 2011 16:51:06 -0500
From: Graham Wooden <graham@g-rock.net>
To: nanog@nanog.org
In-Reply-To: <8B604AC6-01B7-45C7-8DCD-5559758E371B@zaidali.com>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Quoting Zaid Ali <zaid@zaidali.com>:
> I have seen age old discussions on single AS vs multiple AS for =20
> backbone and datacenter design. I am particularly interested in =20
> operational challenges for running AS per region e.g. one AS for US, =20
> one EU etc or I have heard folks do one AS per DC. I particularly =20
> don't see any advantage in doing one AS per region or datacenter =20
> since most of the reasons I hear is to reduce the iBGP mesh. I =20
> generally prefer one AS and making use of confederation.
>
> Zaid
>
Hi Zaid,
What timing - this is fresh on my mind too as I am in the middle of =20
doing this myself with three locations, all with independent edges =20
with different transit providers. I actually do have a private Layer2 =20
circuit between, with one site being in the middle. I only have one =20
public AS, but I have selected doing the confederation approach (which =20
some may consider to be overkill with only three edges).
Each site has their own set of IPs and would originate out of their =20
respective edge, and using EIGRP metric changes at each core to get =20
0.0.0.0/0 from another edge if the local fails. Each edge is then =20
announcing each others' subnets with an extra pad or two to keep the =20
asymmetrical routing down (the private L2 isn't as fast as my transits).
Good luck with your deployment!
-graham
