[arin-ppml] Controlling the IPv6 address consumption rate

[William Herrin]

On Fri, Oct 15, 2010 at 12:52 AM, Fred Baker <fred at cisco.com> wrote:
> On Oct 14, 2010, at 6:24 PM, William Herrin wrote:
>> Hokay. Sure. Did it offend you that I described that I described the
>> local LAN as robbing from routing? Would you prefer something along
>> the lines of, "From the routing perspective, IPv6 is only 64 bits
>> large, not 128?" I humbly apologize for my choice of words.
>
> I think there is a question of requirements.
>Hopefully I can say this without people throwing bricks.
>
> The requirements IPv6 was developed against include but are not limited to
> http://www.ietf.org/rfc/rfc1726.txt

Fred,

Respectfully, IPv6's design requirements from the '90's are not at
issue here. IPv6 is built; the protocol is what it is now, not what it
was imagined to become then.

The question I pose now is: with the protocol we have, how do we make it last?

Since 1994, when RFC 1726 was published, it has become commonplace for
households to have an always-on distinct network connected to the
Internet with a router. And its commonplace to carry around a network
on one's cell phone, supplying both the phone and one's laptop
computer. In a small but growing number of cases, that phone supplies
a car or bus full of laptops.

So the better part of two decades after RFC 1726, it takes no great
imagination to foresee that a selecting a top network count an order
of magnitude smaller than world population (10^9 as 1726 suggested
that IPv6 design to) isn't going to last as long as we'd like.


> Now, you can argue that the requirements were incorrect.
>If so, please advise: how many LANs in the world do you
>need to be able to individually enumerate, and what are
>your aggregation requirements? How many hosts per LAN?

I don't know. Any guess I might make will surely prove as laughable as
1726's 10^9.

Fortunately, at an operations level there's a smarter question.

We have decades of experience with how quickly ISPs of various
customer counts consume percentages of the address space. How much
more slowly do we want them to consume percentages of the IPv6 address
space?

Determine that answer, and it's easy math from our IPv4 experience to
set the upper bounds for how much address space an IPv6 ISP may hold.
Whatever bits are left is what's available for the ISP to employ
whatever routing and assignment practices they select.

Regards,
Bill Herrin


-- 
William D. Herrin ................ herrin at dirtside.com  bill at herrin.us
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004