Re: [sig-policy] prop-087: IPv6 address allocation fordeployment purpose
(Tomohiro -INSTALLER- Fujisaki/藤崎 智宏) said the following on 17/08/10
11:54 :
>
> I'm sorry if I misunderstand your comment, but the problem is not
> number of addresses but prefixes of the addresses.
>
> If ISPs use a same address prefix for all users, it is possible to
> shorten the encoded prefix. I'm not sure how many ISPs request and get
> additional address blocks under same IPv4 address prefixes, but I
> suppose there are not so many ISPs that use a single prefix for all
> of their customers.
Okay, so the concern is about the situation where an ISP (I'll use
private address space for my example) might use 172.16.0.0/16 for one
group of 6rd customers. And then are allocated 172.17.0.0/16 for the
next group. So they can't use the last 16-bits of the IPv4 address for
6rd because the IPv6 addresses would end up being the same in each pool.
Can't they just use the last 17-bits then - or some other way of
identifying the difference?
If each 6rd customer is receiving an IPv6 /48 this way, each IPv4 /16
allows a total of 65536 addresses. And presumably a total of 65536
customers.
If the ISP has two IPv4 /16s in use for their existing customers, that'd
be 131072 IPv4 addresses in total - and 131072 /48s in IPv6 is the same
as a /31. And an IPv6 /31 would be able to encode the last 17-bits of
the IPv4 address for 6rd as I mentioned above.
And if customers are getting less than an IPv6 /48 by way of 6rd, then
there are more bits available to encode the IPv4 address in.
As you can see, I'm still having trouble working out what the problem
is. There must be some other scenario...?
Thanks!
philip
--