[ppml] 2005-1:Business Need for PI Assignments

[Stephen Sprunk]

Thus spake <Michael.Dillon at radianz.com>
> > > In the end, I think that strict geographic allocations which
> > > follow the topology of the network, are the only workable
> > > solution.
> >
> > Make up your mind -- do you want strict geographic allocations
> > or do you want addressing that follows topology?  Those are
> > only the same thing for very simple end user scenarios.
>
> I want strict geographic allocations that follow the geography of the
> network. This is roughly similar to real-world geography except that
> the distance between cities is not measured in straight lines because
> the fiber routes are not straight lines.

The topology of the network rarely follows geography.  Both ISPs and
enterprises tend to have many, many sites that are not connected to
exchanges or private peering/transit to every other provider in that
location.

For a concrete example, say each state is given a geographical registry
(extrapolate to national or continental levels instead if desired).  A
provider in Detroit may reasonably choose to get its connectivity solely
from Chicago and Cleveland.  Either (a) the provider uses Illinois or Ohio
addresses, meaning their customers can only multihome (or change providers)
to other providers also serving Illinois and Ohio, or (b) the provider uses
Michigan addresses and must announce more-specifics into the global (or at
least US) routing tables for each customer block.  Both scenarios are worse
than what we have today.

The only way to make this work is to force providers (by law) to peer with
or purchase transit from all other providers in a given location to be
allowed to offer service there.  I can't help but think the economic effects
of this would be far, far worse than what we suffer with today, nor have I
seen any studies of these problems by the various folks proposing geographic
addressing.

> > Real world example: picture an enterprise with 40,000 sites in
> > the US.  Should each of those sites be numbered from the local
> > state/city pool?
>
> Should? I'm talking about "may" here. I want us to allocate
> geographic addressing so that they MAY number each site from
> their local city pool or they MAY number each site according to
> some other plan. I want to create a situation in which the
> existing IPv6 address allocation plan has to compete in the
> real world with a geographic addressing plan.

Unfortunately, many others proposing the same addressing model are planning
to use it to replace, not augment, the current model.

> > The reason it's not being pursued is that (a) the topology does
> > not follow geography today and (b) providers have [no]
> > motivation to make it so.
>
> Topology does follow geography.

No, it does not.  The vast, vast majority of networks I personally have
worked on have little if any correlation to geography at the IP level.

Geographic addressing makes things simpler for sites that are connected only
to other sites in the same area, and that may be nice for consumers and some
businesses, but it makes things worse for operators of more complicated
networks.

S

Stephen Sprunk      "Those people who think they know everything
CCIE #3723         are a great annoyance to those of us who do."
K5SSS                                             --Isaac Asimov