[AusNOG] SMH: "No room at the internet"

[Mark Newton]

On 19/05/2010, at 1:15 PM, Daniel Hooper wrote:

It’s hard for smaller operators to justify rolling out ipv6 at the moment unless the big boys get on board, maybe it’s time for us smaller operators to start putting our foot up someone’s backside at our upstream providers to get ipv6 moving forward in a realistic time frame.

I'm not sure why "the big boys" actions make any difference whatsoever to
what you're doing.

What support do you expect from them?  What could they do that'd make any
difference?

 My scenario is this at the moment, I have a small base of customers utilising ipv4, I could go ahead and re-deck our core out and ipv6 enable it, however my up streams aren’t doing ipv6 yet (and no replies back from them as to when they will be), so I’ve basically got no global ipv6 connectivity. Management isn’t going to open the cheque book until the customers are screaming for it and my providers don’t seem to give a hoot about it at the moment, possibly more screaming from us will push this.

Everyone'll make their own business decisions, sure.  But the path you've
described looks pretty risky to me, and isn't likely to yield benefits proportional
to its costs.

I can see where it comes from, though:  The perception from people who
don't know what they're talking about  is that IPv6 is new, and adopting it
is a risk.  So if management wants a risk-averse path, delaying IPv6 until
everyone else has jumped on board might look like the right answer, because
it'd reduce the chance that you waste a bunch of effort on something that
might not turn into reality.

Unfortunately it's based on a false premise, and is consequently bass-
ackwards.

IPv6 is happening, on timescales you can't control.  So any risk mitigation
strategy that's based on a supposition that it won't happen, or that you have
influence over its rate of adoption, is flawed.

The business risk isn't that you'll waste expensive effort on something
that doesn't come to pass -- It's that this IPv6 thing will happen, and you'll
be caught flat-footed, unable to respond because you don't know how.

You'll be able to recognize the signs of that happening when your first
customer comes to you asking for IPv6 and you have to say, "No."

IPv6 isn't impossible;  it isn't even hard.  But it does require planning and
preparation, mostly because you can't just assume that whatever you're
doing with IPv4 maps seamlessly to IPv6.  Your geeks need to retrain
themselves, and you'll need to find some way to help your salescritters
as well.


Also worth mentioning at the moment is the small amount of resources on the internet that are ipv6 capable, I cant see a huge amount of traffic on the network being ipv6 (so no point buying another pipe to another upstream just to gain global ipv6 connectivity)

It isn't about traffic.  It's about the Internet continuing to function in the way
everyone expects it to function after Sep/Oct next year.

If you're waiting for traffic, that means you're going to be a late adopter,
and the v6 traffic you see which eventually convinces you to move is
probably originating from the customers who used to be connected to
your network. :-)

 Unfortunately I don’t believe for a second the small community of tech savvy internet consumers is going to make enough of a song & dance about ipv6 for it to have any real impact.

Almost no customers are going to make a big song and dance about IPv6.

But we're all going to do it anyway.

All of us in this industry know that there's IPv6 demand coming down the
pike that we need to be ready to service.  Thanks to Geoff's exhaustive efforts
we've even known the demand spike's date for about the last five years,
so we're all going to look like a pack of dills if we aren't ready for it, aren't
we?

Some of us will be ready to service it, and will have happy bill-paying
customers as a result.  And others won't start preparing until the demand
spike hits, and will have customers dawning to the slow realization that
their service provider can't satisfy their needs.

For businesses whose bread and butter comes from satisfying customers
connectivity needs, I know which alternative looks riskier.  Does your boss?

  - mark


--
Mark Newton                               Email:  newton at internode.com.au<mailto:newton at internode.com.au> (W)
Network Engineer                          Email:  newton at atdot.dotat.org<mailto:newton at atdot.dotat.org>  (H)
Internode Pty Ltd                         Desk:   +61-8-82282999
"Network Man" - Anagram of "Mark Newton"  Mobile: +61-416-202-223





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20100520/3ead99ae/attachment.html>