1) There are a lot of ways to interwork IPv4 and IPv6, particularly for
getting to content. (The main problem with most such mechanisms is that
they can not support new applications. Requiring really new things to
use the new infrastructure (IPv6 or something else) seems quite
reasonable to me.
2) My point is that unless things fragment in very strange ways
(possible, but difficult to actually see happening), the number of
prefixes in the Internet can not grow very far. 90% of all addresses
are already in the allocation cycle. (Not all are allocated, and I am
not saying we are within 10% of the top.) Basically, new ISPs can not
get enough IPv4 space to do business. So they can not get space and
fragment it, because even fragmented it simply won't do the job.
Yes, it is quite possible for economics to cause shifts in behavior, and
it is possible for this to cause far more serious fragmentation. But,
in terms of planning, that does not seema likely path.
Put differently,
1) If we (the IETF and related communities) do not solve the v4/v6
interworking problem to a useful degree, then it does not matter what we
do. Economic and utilization pressures will force really unpleasant
implementation hacks. Discussing architecture for such an environment
is probably useless, because the constraints are too tight.
2) If we do solve the itnerworking problem, while it is worth makign
IPv4 work better, it is probably only strictly necessary to ensure that
we have an architecture for IPv6 that can actually cope with the
pressure we put on it.
Currently, the IETF is working very hard on point 1.
If we do not get a good answer to point 2, then we are going to simply
give ourselves more rope to be hung with, and it won't be long before we
are dangling and choking on our own mess.
Yours,
Joel
Scott Brim wrote:
Joel M. Halpern allegedly wrote on 02/03/2010 09:49 EST:
(Robin, I am using your headline as a hook for my comments. This topic
has been bouncing and I wanted to suggest another perspective on this.
It probably has very little to do with what you were trying to get at in
your note.)
IPv4 will clearly be around for a LONG time.
IPv4 will, I expect, continue to be used for a LOT of content for a LONG
time.
However, I am not sure that matters for the RRG work.
The IPv4 Internet works. The routers, and the routing system, cope with
the current pressures. To my way of looking at things, the question is
how will the Internet routing system cope with growth.
But, definitionally, there really is not that much growth left in IPv4.
On the other hand, although IPv6 is tiny now, unless the entire Internet
stops growing, v6 will become massive. If we do not work out an
architecture that can cope with growth, when that pass the current IPv4
size, and keeps growing faster, we will be up the proverbial creek
without any control whatsoever.
Hence, I think it is actually quite reasonable to have an architecture
and approach which only addresses IPv6 Internet Scaling. Without
denying that IPv4 will be here, and important, for a very, very, long time.
Joel, I don't quite get it. First, if you only deal with v6 then would
you create two parallel Internets? If so you need routing and
addressing interworking. Second, there will be pressure for growth in
the number of IPv4 _prefixes_ and if we don't design a way to support
it, people will find ways we don't like.
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
