[ale] Can a class-b space be sold? How much?

[Michael H. Warfield]

On Thu, 2010-03-18 at 13:04 -0400, Pat Regan wrote: 
> Hash: SHA1

> On 03/18/2010 12:06 PM, Michael B. Trausch wrote:
> > I know that at least one environment that I worked in exhausted their 
> > 10.0.0.0/8 space, and that is why they have an additional /8 on their 
> > own network for their own machines.  They just firewall the /8 normally.

> Tell them they should have segmented a little more carefully.  I'd be
> surprised to find any single organization approaching anywhere near 16
> million hosts :).

Comcast.

At a NANOG meeting several years ago, Comcast did a presentation on how
and why they were migrating all their network management over to IPv6.
The gist of it was that, between cable modems and cable converters and
DTA's, etc, etc, etc, they had exceeded their ability to manage all the
devices, just the internal private IP addresses to the devices, from a
common data aggregation point.  Because it was a management and data
aggregation issue, multiple network segments with duplicate private
addresses behind NAT's was not an option.  Their only viable choice was
to begin migrating all the device management over to IPv6.  Now, they're
finally offering a beta test of IPv6 to the end customer as well.  They
saw the handwriting on that wall years ago.

I wouldn't be surprised if AT&T and similar ilk weren't far behind.

Now, even with all the conservation and recovery that's been performed,
we're still looking at v4 address exhaustion by the end of 2012.  But
recovery of v4 space isn't the answer either.  It's not the whole
problem, it's just the one with the highest profile.

DON'T buy into the myth that recovering all these /8's (or even the
multitude of /16's) out there would help much.  As each of those have
been recovered and retasked, the core routers have all taken major load
hits as the routing tables have grown explosively.  That's the flip side
of the coin that IPv6 was designed to address (ooopppsss - no joke
intended there), route aggregation.  The more we recover and retask v4
address space, the more fragmented it gets and the worse the router
problems becomes.  We're already past that break even point with over a
1/4 of a million routes in the core routers, but everyone (in the IPv4
world at least) seems to want to ignore that harsh reality that we're
already overtaxing the core routers.  IPv6 has more routed capacity, in
terms of routed /48 full networks, right now in the core routes than all
of the IPv4 routed HOST addresses combined and doing it with less that
1/100 of the routes in the routers.

IPv6 is not just IPv4 with fat addresses.  IPv4 has far more problems
than merely running out of addresses.

> Pat

Regards,
Mike
-- 
Michael H. Warfield (AI4NB) | (770) 985-6132 |  mhw at WittsEnd.com
   /\/\|=mhw=|\/\/          | (678) 463-0932 |  http://www.wittsend.com/mhw/
   NIC whois: MHW9          | An optimist believes we live in the best of all
 PGP Key: 0x674627FF        | possible worlds.  A pessimist is sure of it!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 482 bytes
Desc: This is a digitally signed message part
Url : http://mail.ale.org/pipermail/ale/attachments/20100318/d566246c/attachment-0001.bin