North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: When IPv6 ... if ever?

  • From: Richard A. Steenbergen
  • Date: Sat Sep 02 16:49:24 2000

On Sat, 2 Sep 2000, Roeland M.J. Meyer wrote:

> 3) Conservatively, every recent linux node has IPv6 kernel support and 
> every Windoze box can do IPv6 (not to mention what we have just heard
> from the BSD camp <g>). This is over 50% of the leaves out there
> (conservatively). At what point is there sufficient market penetration  
> of the technology to consider rolling  IPv4/IPv6
> interoperability/capability on the core routers and switches (something 
> short of 100%, I would hope)?
>
> 4) Is it maybe that Sun, HP, Intel, Cisco, IBM, and the telco's, aren't
> all quite ready yet?

There are two key problems which are preventing the widespread use of
IPv6, IP Allocation, and network vendor support. Support for all hosts is
actually one of the least of the problems.

Think of it this way, major networks want to deploy IPv6... How do they do
it? They certainly can't do it on their primary backbone links and
routers, the support from vendors is simply not there. Even if there was
working code, they wouldn't dare deploy it on their production network,
the code is too unstable (especially IPv6 routing protocols), and they
risk looking unreliable in comparison to those who don't even make the
attempt to support IPv6. So what do they do? Without the network there is
not the demand for high traffic IPv6, and without the demand there is
little desire to build the network. Should they buy seperate routers, try
desperately to make IPv6 work well on a spare 7200, and hope not to get a
black eye from customers who expect the same level of routine-ness we are
experienced with in IPv4? Should they provision more circuits because of
this? Build a parallel network supporting IPv6, without a current customer
demand? Or do they say, "we'll wait until the vendors get it right"? Even
if you get past all that, there is still the very obvious fact that for
certain vendors, the levels of performance we expect today are because of
extensive tuning for IPv4 speed, and we know we won't be seeing this level
of performance right off the bat from IPv6. And while they wait, there is
no usable IPv6 infastructure, less interest and development from vendors
who don't really see a "need", and less demand from users who know there
is no point when they can't find a network infastructure to support them.

The other problem is IP allocation. ARIN and others are used to their
comfortable role as IP Nazi ("No IP for you, come back, one year!"), they
have happy extortion-scheme pricing on it, and they're not about to give
it up voluntarily any more then NetSol would have without being forced...
If you doubt this, go look at the current policies for obtaining IPv6
space.

-- 
Richard A Steenbergen <[email protected]>   http://www.e-gerbil.net/humble
PGP Key ID: 0x138EA177  (67 29 D7 BC E8 18 3E DA  B2 46 B3 D8 14 36 FE B6)