[AusNOG] Less than 10% of IPv4 Addresses Remain Unallocated

[Andreux Fort]

On Thu, Jan 21, 2010 at 10:57 AM, Paul Grehan <pgrehan at onthenet.com.au> wrote:
> Geoff,
>
> This always seems to be a moot point ending in a finger pointing exercise
> between SP's and vendors (particularly those that mass produce low cost CPE,
> and those that provide their chipsets)...
>
> Is APNIC and other RIR's getting much traction from these vendors as I see
> SP's as only half the equation....

I don't really follow this reasoning: How much equipment do RIRs
purchase from vendors?  As far as I know, it's not a lot.  Let's face
it - that dictates how much traction a customer gets.

Economics dictates that at some point, it'll become less expensive to
buy IPv6-capable equipment than to purchase v4 address-space.  Until
then, customers need to demand equipment (or build their own) that
supports this.  And where it supports this, demand support from the
vendor in placing a priority on fixing IPv6 bugs that still remain.  I
know that Lorenzo Colitti has done a lot of work here (not the reason
Erik and him received the Ito award, though) but his employer doesn't
use all of the major vendor's equipment, and I know nothing of other
large networks.  However, bigger customers (read: bigger than we have
in oz) get priority.

Then there's the internal questions of management networks and the
likes - when do you transition that, or do you run two? (I reckon
that'll be staying v4 for some time to come, other than for greenfield
players in the wake of 2012).

<ob-nog> Is it just me or did the Mayans predict the IPv4 address
exhaustion, too? ;-)


>
> -Paul
>
> Geoff Huston wrote:
>
> On 21/01/2010, at 10:31 AM, Skeeve Stevens wrote:
>
>
>
> Geoff,
>
> That is an awesome collection of data and is very useful to use in providing
> information to customers and entities.
>
> Do you have that information summarised together on your website somewhere
> that we can point them to?
>
> One of the biggest issues we have is explaining this simply to people who
> have no idea.
>
>
>
> Thanks Skeeve.
>
> This information is at  http://ipv4.potaroo.net  (updated on a daily basis)
>
> I know its long-winded and perhaps this page heads too far into the
> mathematical techniques that underlie the projection, so I'd appreciate any
> feedback folk may have about improving the presentation of the data on that
> page.
>
>
> [For those who have asked me in the past about the variability of this
> prediction, perhaps I should also point out that a plot of the various
> predictions dates over time is at
> http://www.potaroo.net/tools/ipv4/predict.png, and the change over time in
> the amount of time left to the time of predicted runout is at
> http://www.potaroo.net/tools/ipv4/timeleft-predict.png.]
>
>
> thanks,
>
>   Geoff
>
>
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
>
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
>
>



-- 
Andreux Fort (afort at choqolat.org)