Last week NTT America, the US arm of the Japanese giant NTT Communications (NYSE:NTT, news, filings), announced it had expanded its IPv6 backbone to five sdxc datacenters. Of course IPv6 is the designated successor to IPv4, whose address space will run out completely in a few years. What does that mean? Just that IPv4 was designed long ago, when the internet we know today seemed unfathomably large.
We've been hearing for years IPv6 was needed to avoid catastrophe, yet nothing ever seems to happen. Well, in the US anyway - as seems all too frequent these days on the internet, IPv6 leadership has come from international sources. Global Crossing has been rolling out IPv6 steadily for a while now, but if you're not reading the PRs you don't hear about it much in the US. Even Level 3 started dabbling in IPv6 in Amsterdam before bringing it to the US. In general, US carriers know IPv6 and have tested it and all say they are ready for it, but aren't deploying it commercially because they just don't need to yet. Of all carriers internationally though, NTT has been perhaps the most aggressive in using and promoting IPv6, so it is perhaps natural that they should make this move befoe others. Earlier this month at CES they even discussed their commercial IPTV over IPv6 implementation in Japan that handles 76 channels, 10K on demand titles, and 13,000 karaoke titles. Their expansion of IPv6 to the US may give them a leg up when adoption starts to rise.
So when does IPv6 adoption take off in the US? I think it won't start until people feel pain first, that's just how it is. I think David Siegel described the situation well in his discussion of the phases of IPv4 exhaustion. Because the benefits of IPv6 are mainly the large address space, it won't take off until people start really squabbling over what's left in IPv4. But when that starts to happen, I think the wave will build rapidly.