[ppml] IPv4 "Up For Grabs" proposal

[James Jun]

[ snip ]

> 
> 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.

Not quite, the IPv6 DFZ is currently at around ~800 routes.  


> 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.

It's not that of a big problem than some people are speculating actually.
Most modern TCAM sizes can perform 1M entries, even at a bad case
implementation view point, you still have plenty of space to shove IPv4 DFZ
and IPv6 DFZ altogether, put a several thousand of your interior routes, and
on top of that play with MPLS and multicast entries.


For most carriers right now, there's more urgency to upgrade hardware in
order to support higher bandwidth load than due to routing table.  Routing
table issue is mostly secondary issue for many carriers -- many of them
upgrade their gear mostly out of need to support higher port capacities and
scaling for increased bandwidth use by their clients.

Additionally, a number of networks have designed their core in such a way
that it doesn't even have to carry much of DFZ anymore (see MPLS).  This
reduced opex translates to availability of funds to do something more useful
(i.e. purchase edge routers that are more beefy with increased FIB capacity
or otherwise).

james