[ppml] v6 Multihoming (was Re: IPv4 "Up For Grabs" proposal)

[Scott Leibrand]

Seth Mattinen wrote:
> William Herrin wrote:
>   
>> Because of the change in how IP addresses are justified and assigned,
>> the IPv6 DFZ has only a couple thousand routes and is expected to have
>> fewer than 100,000 routes at full deployment. This will make it
>> possible for folks on the DFZ to both spend less money -and- do a
>> better job of keeping the Internet stable.
>>
>> The hitch is: until IPv4 goes away, you're not talking about 100,000
>> routes. You're talking about 100,000 IPv6 routes PLUS 220,000 IPv4
>> routes. So it gets worse before it gets better and until IPv4 goes
>> away, it doesn't get better.
>>     
>
> So... explain multihoming in an IPv6 world to me.
>   

Basically, you can multihome in IPv6 the same way you do in IPv4.  If 
you qualify for PI addresses in IPv4, you also qualify for IPv6 PI (in 
the ARIN region).  You can announce your PI /48, or you PA block for 
that matter, in BGP to your upstreams.

The main reason you can do the same thing in IPv6 as in IPv4 and get 
half the routes is that you're no longer allocating multiple discrete 
netblocks to a single ASN, so the ratio of routes to ASNs in the table 
is closer to 2:1 instead of 5:1.

Is that what you were asking?

In addition, IPv6 supports host multihoming, where a host has multiple 
IP addresses and uses whichever is appropriate.  Extensions like shim6 
are being standardized to allow session failover, which will open up a 
new realm of possibilities for small-site multihoming.  I suspect most 
large sites will continue to multihome with BGP and PI space, though.

-Scott