[ipv6-ops] Re: So why is "IPv4 with longer addresses" a problem anyway?

[Mark Smith]

Hi Fred,

On Wed, 26 May 2010 17:01:08 -0700
Fred Baker <fred at cisco.com> wrote:

> 
> On May 26, 2010, at 2:32 PM, Mark Smith wrote:
> 
> > IMHO, if the Internet was running IPX (32 bit network number, 48 bit node address), we'd still be running it, and wouldn't now have the problem of running out of addresses. 
> 
> well, yes, but we might have some issues with route scaling. IPX enumerates LANs, not edge networks or the ISPs that serve them. 
> 

My understanding, based on what Radia Perlman says about IPX in
"Interconnections", is that IPX network look up was actually longest
match, rather than exact match. In the time I worked with IPX, I never
saw or heard of anybody aggregating IPX networks, however it would seem
the spec might have allowed it.

A quick google search seems to show IOS supported specifying a mask on
an IPX static route, so that also suggests that IPX routing is longest
match:

http://www.cisco.com/en/US/docs/ios/12_2/atipx/command/reference/2rfipx3.html#wp1024096

> We could make the same comment about CLNS and NSAPs; CLNS enumerates networks, and within them, subnets and hosts. The networks would be the edge networks, not the ISPs that serve them.
> 
> IPv6 isn't perfect. The nice thing about it is that we can reasonably enumerate hosts, subnets (/64), edge networks (/48 or longer), and ISPs (/32 or thereabouts). 

Regards,
Mark.