> Sean said that traditional multihoming would be "very difficult".
Actually, Sean's statement was that "IPv6's current addressing
architecture" makes multihoming very difficult, and that is the point
that is untrue.
> You replied that "This is not true" (which I take to mean
> that multihoming is not very difficult), and then go on to describe
> something that sounds very difficult to me (maintain longer prefixes,
> make multihomed customers pay for it, get the ISPs to agree to
> handle such longer prefixes).
This is exactly what we have in IPv4 today, except that multihomed
customers aren't really paying. And maybe they won't pay in IPv6
either. The market can sort that out. The point of the IPv6 addressing
architecture is to make that sort of multihoming a _possibility_ and
an _optimization_ rather than a _requirement_. In contrast, today's
IPv4 has lots of long prefixes in the DFZ with no clear way of placing
an upper bound on the number of prefixes that must be maintained in
the DFZ to provide reachability to all sites. In IPv6, the small
number (8K's worth) of TLAs should do the trick. That still leaves
room for many more "value-add" routes when one considers that today's
IPv4 already can/must handle 75K worth of prefixes in the DFZ. The
difference is in IPv4 it is a requirement to maintain all 75K+ routes
in order to just get reachability.
> You say that "multihoming is still quite posible", but nobody said
> that it was impossible, just difficult. For me, your statement
> certainly reinforces the idea that multihoming in IPv6 is indeed
> very difficult.
As others have pointed out, IPv6 is also developing a multiple
addresses per end-node approach to multihoming. This approach appears
to be more scalable that the current IPv4 solution. Indeed, people
seem to point to multihoming as one of the biggest threats to
continued scaling of the IPv4 routing infrastructure. So one of IPv6's
multihoming approaches is no worse than IPv4, while another appears to
be significantly better. So once again (in contrast to assertions made
by some), IPv6 *does* have some features that should make IPv6 routing
scale better than it does in IPv4.
> I read your statement as follows:
> IPv6 does not solve the multihoming problem. Instead, it tries
> to minimize the damage by:
>
> 1. discouraging the use of multihoming, primarily may making
> multihomed customers pay more for it.
> 2. forcing paths to multihomed sites to be less efficient (at
> least for all but one of the ISP connection points) and or,
> 3. limiting the regions of the internet for which multihoming
> is effective for a given customer.
> Is this an accurate representation?
Absolutely not, as I hope has now been made clear.
Thomas