[AusNOG] Wrote an article for APNIC some here maybe interested in - "UDP over IPv4 – a stepping stone to IPv6?"

Mark Newton newton at atdot.dotat.org
Mon Mar 27 17:23:34 EST 2017


With all due respect:

This year, 20% of Google users (a rough proxy for internet users — most users do a Google search or watch a youtube video every now and then) will be hitting google.com <http://google.com/> with native, non-tunneled IPv6.
https://www.google.com/intl/en/ipv6/statistics.html <https://www.google.com/intl/en/ipv6/statistics.html>

That is, for all intents and purposes, a successful mass market adoption. 

That’s up from 1% four years ago, and increasing exponentially. I project that we’ll pass 50% of internet end users using native IPv6 at some point during 2019.

If you are not part of that, you are objectively failing.

“Where I think we need to be seeing more discussion” is about winding-up endless bullshit discussions and just making it happen.  If you think we need more discussion around business cases and architectures when 20% of the world has actually done it already, then you are part of the problem. Maybe your business model or architecture is deficient, and that makes it uniquely difficult for you. Maybe Australians just don’t know how to run ISPs, and that’s why this country hasn’t quite nudged 8% adoption whilst North America, the ones who just five years ago were spraying threads all over NANOG about how hard and impossible it was, are over 30% adoption.

Get on with it.

<glares at TPG>

   - mark


> On Mar 27, 2017, at 2:27 PM, Paul Wilkins <paulwilkins369 at gmail.com> wrote:
> 
> Mark,
> Thanks for posting this great summary of approaches to delivering ipV6 over UDP. It's important that we have active discussion, so the community both knows what approaches are being considered, and where the industry as a whole is moving. Because migration to ipv6 isn't purely a technical problem. There's the interplay of technology and business models that creates the ecosystem which can't exist without consensus solutions. And obviously starting a conversation is the first step to any consensus solution.
> 
> Where I think we need to be seeing more discussion, is around the business cases for the various architectures, scale, and opportunity costs. For instance, where is the road map once you have heavy use of ipv6 over UDP, and you discover lack of flow control is no longer specific to points of aggregation, but has become a feature of flows across the network?
> 
> Someone I think once thought, we'll just tack on 96(32 actually) extra bits and the address problem would be solved. The problem is the economies of scale for ipv4 are better than ipv6, and no one considered the operational costs of running both stacks, to allow the bridging needed as a fundamental part of a migration and successful evolution.
> 
> Kind regards
> 
> Paul Wilkins
> 
> On 25 March 2017 at 15:10, Mark Smith <markzzzsmith at gmail.com <mailto:markzzzsmith at gmail.com>> wrote:
> https://blog.apnic.net/2017/03/24/udp-ipv4-stepping-stone-ipv6/ <https://blog.apnic.net/2017/03/24/udp-ipv4-stepping-stone-ipv6/>
> 
> Regards,
> Mark.
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net <mailto:AusNOG at lists.ausnog.net>
> http://lists.ausnog.net/mailman/listinfo/ausnog <http://lists.ausnog.net/mailman/listinfo/ausnog>
> 
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20170327/e7c1d394/attachment.html>


More information about the AusNOG mailing list