John W. O'BrienI'm starting to wonder, if Amazon continues to buy up as much of the IPv4 address space as they can, will AWS eventually become the only network---not the only IaaS platform, the only *network*---that can practically absorb IPv4 Internet growth of any kind? #IPv6
Carl@neirbowj They can't be the *only* one, there need to be telcos as well for customers to connect. But I agree, their IPv4 hoover is stong and should be stopped.
@edolnx I'm not trying to judge (yet) whether this is good or not, and, if so, for whom. I'm observing that Amazon just dropped ~$200M on the equivalent of a /8.87 from T-Mobile, which means that T-Mobile thinks $200M cash is worth more *right now* than over 9M IPv4 addresses, and Amazon thinks the opposite. In the "Weekly Global IPv4 Routing Table Report" on the NANOG mailing list, the volume of v4 space Amazon is announcing dwarfs most other operators.
@edolnx T-Mobile thinks IPv4 address space will be worth less to them in the future, and Amazon thinks it will be worth more to them. They might both be right. If so, what does that mean about the rest of the networking world? T-Mobile benefits from IPv6-friendly mobile platforms, Android and IOS. Amazon stands to benefit from all the IPv6-laggards: Verizon FiOS, GitHub, every enterprise and enterprise SaaS platform...
Named Bird@neirbowj @edolnx The adoption of IPv6 will go on, more and more things will start requiring it. Then ISP's can make do with less and less IPv4 addresses for their CG-NAT. Websites will sooner or later start becoming IPv6-only, then most ISP's will be forced into IPv6 regardless, resulting in a feedback loop of adoption.
When everything that is consumer-facing has IPv6, support for IPv4 will start to decline, eventually becoming useless.