[ppml] And as for assignments...

[Jonathan Barker]

All who have responded,

I basically asked "What's wrong with my logic" just to make sure I 
wasn't missing anything. I can do /64s... you're right, I have so much 
address space with just a /32 under v6 that I could serve 4 billion 
customers - a figure I can say with certainty I'll never attain.

But, once upon a time, IPv4 didn't have classless routing capabilities. 
That was built in later. If there's demand, all these other features can 
surely be built in for smaller blocks as well, whether IETF standard or 
not... (if a couple of near monopolies agree, that is. ;-) By the time 
we get most of the Internet on v6 - who knows what v6 will actually look 
like.

I suppose I've held a little too much conservatism over from IPv4 
constraints, as I have been very constrained when it comes to IP 
addressing. We're the most IPv4 efficient ISP I know of - but I just 
cannot understand why a single family dwelling would ever need or want 
2^64 addresses. It's just so phenomenally wasteful. It's a little like 
building a 20 Gigawatt nuclear power plant for each and every household 
who needs power to run their PC. :-)

Jonathan
>
> Methinks you haven't played with v6 much; a subnet is /64 except in 
> very specialized cases like PTP links or loopbacks.  While it is 
> _possible_ to do what you propose, you're going to lose a lot of the 
> advantages of v6 and make life much more difficult for both you and 
> your customers by doing this.
>
> What is stopping you from giving each customer a /64?  With your /32, 
> that means you can have up to 4B customers, which should be sufficient 
> :)  Why bother trying to assign less?  v6 is not v4; there is no need 
> to go to extreme lengths to limit subnet sizes.
>
>> My question is - What is wrong with my logic, in that most
>> people who are commenting don't think in these terms?
>
> Whether you agree with their logic or not, the IETF has designed IPv6 
> to work with /64 host subnets, and a lot of things are difficult to 
> impossible if you don't follow that model.  Even if you manage to get 
> minimal services working, you're going to find a lot of bugs in 
> vendors' implementations or not be able to use the more novel services 
> if you insist on a non-standard subnet size.
>
> S
>
> Stephen Sprunk         "God does not play dice."  --Albert Einstein
> CCIE #3723         "God is an inveterate gambler, and He throws the
> K5SSS        dice at every possible opportunity." --Stephen Hawking