IPv6 PI allocation

[Iljitsch van Beijnum]

On 18-mei-2007, at 16:23, Andrew Alston wrote:

>> Let me check... Available IPv4 address space as of right now: 1233
>> million addresses. Obviously those aren't going to last THAT long,
>> but for now, we have enough. So it makes sense that people continue
>> deploying IPv4.

[...]

> 6.5 billion people on the planet, 1.1 billion people with net  
> access... 5.4
> billion people without addresses...

[...]

> Or wait, do the unaddressed portions of the world not matter?

In a sense, they don't: if you aren't on the internet, you don't need  
address space.

But that's not the point. The point is that we're going to have a  
problem in the future. We don't have a problem today. If it's going  
to rain tomorrow, are you going to stay inside the house today in  
order to avoid getting wet? Of course not. You prepare for tomorrow  
and enjoy the nice weather while it last.

People don't have to adopt IPv6 today. What they have to do is  
prepare so they can adopt it around the time we run out of IPv4  
address space. In order to do that, they have to be convinced that  
we're going to run out of IPv4 space at a given point in time. Today,  
we're not in the position to provide predictions of sufficient  
quality to convince people. In another one or two years that problem  
will solve itself (I think the magic threshold is when what we have  
left is 3 x what we use per year.)

> Then we can talk about the ipv4 depletion mathematical models... I  
> suppose
> Geoff's numbers mean nothing?  We ARENT facing depletion by end of  
> 2009?

Geoff predicts the IANA global pool will run out by that time.  
Currently, the RIRs hold some 400 million addresses they haven't  
given out yet, so even if Geoff's projection pans out we'll have  
until 2011.