Re: Regionally aggregatable address space for multihoming

[Iljitsch van Beijnum <iljitsch@muada.com>]

On Tue, 12 Jun 2001, RJ Atkinson wrote:

> At 07:49 12/06/01, Iljitsch van Beijnum wrote:
> >I guess this is true for the really large networks. Maybe this is 
> >because they've lacked a good incentive to interconnect at more 
> >locations. Geographic aggregation could be the necessary incentive. 
> >They  are already present in many locations and local fiber is cheap, 
> >so it could happen. On the other hand, it may not.

>         Local fibre is cheap in some places, expensive in others.

You are right, it can be expensive. I was thinking along the line of hardware
and investments. Obviously, for non-telco ISPs the market price is more
relevant.

>         Given that, I don't see how trying to force a local
> exchange system will be effective anytime soon.

Does anyone have a good overview of the current world wide situation? As far
as I can tell routing is pretty good these days, I hardly ever see packets
taking the scenic route halfway around Europe or the US these days. Of course
I see mostly traffic that has NL as the source or destination, so this
doesn't mean all that much.

>  And I don't
> necessarily think that forcing such will actually provide
> significant improvement.  I'm open minded about the latter,
> but would need to see more substantial explanation of how
> such actually improves the health of the global routing
> system

Well, let's reverse the argument: how will assigning very small blocks of
address space to multihomed networks with no particular rationale help us? At
least with a geographic assignment policy some people can filter some
announcements in some parts of their network some of the time.

As I see it, my original proposal (or the proposal by Tony Hain) aren't going
to fix all routing issues forever, because we have to plan for too many
multihomed networks (a billion) and we cannot depend on enough local
interconnectivity between ISPs. On the other hand, it should help a bit and
it won't break anything.

So we have to think of something better. I still think geography can be a
good "better than nothing" attribute when there is no better information
available, but we'll see.

Iljitsch