[c-nsp] BGP and remote POPs with individual upstreams

[Kevin Graham]

> My initial thoughts are to BGP peer between POPs with a higher local-pref

> for the local outbound traffic and to prepend between the POPs so inbound
> traffic is more likely to take the shortest path inbound.
>
> Is this too simplistic? Prone to trouble? What gotchas should I be looking
> at, or other designs should I be considering?

This is entirely reasonable. If you wanted, you could also look at doing
each POP as a federation, which would get similar results as well.

For egress traffic with different transit providers, you'll inevitably still
leak some traffic between POP's with more-specifics learned via a single
provider, but this should be small.

An AS-prepend of non-local-prefixes would take care of ingress reasonably
(again, expect some small surprises), though if you have the same provider
to multiple POPs, you could also get there w/ just MED's.

>  My Cisco bookshelf isn't helping me much with this...

Add Internet Routing Architectures to it then, it is well worth the read.