[13465] in North American Network Operators' Group
RE: IPv8 < IPv6
daemon@ATHENA.MIT.EDU (Rodney Joffe)
Thu Nov 6 15:26:53 1997
From: Rodney Joffe <rjoffe@genuity.net>
To: "'NANOG'" <nanog@merit.edu>
Date: Thu, 6 Nov 1997 13:12:53 -0700
> Alan Hannan said:
>
> Physical topology is likely to map to geographic topology.
> Circuits certainly do take odd L1 paths to connect L1 endpoints,
> but these are exceptions, not he rule.
>
> Accordingly, not allocating in a geographic fashion lends to
> deaggregation, which is bad.
>
> A good sentiment, but not neccessarily practical. The one constant we
> can count on is that things change.
>
So, if I look at my crystal ball, and predict that I will need n
addresses in a particular physical location, using your planning
exemplar I will allocate n addresses in an aggregate. The fact that I
only need 1/2 n now means that I will temporarily waste 1/2 n. But what
happens when I need n + 1 addresses in that city? Which is the lesser of
two evils?
So there is also a good argument to *not* rely on a congruence
of physical and geographic topology.
Some form of NAT still appears preferable.
>
>
Rodney Joffe
Chief Technology Officer
Genuity Inc., a Bechtel company
http://www.genuity.net
