The end of one calendar year always inspires prognostication about the next. I’m not going to go so far as to make any specific predictions about networking for 2016, but there is one networking topic that should at least be on your radar for 2016: IPv6. I’m also not going to proclaim 2016 as “The Year of IPv6” or anything like that (arguably, that has already happened), but I will say that this should be the year that you start to take IPv6 seriously if you haven’t already.
As of today, four of the five Regional Internet Registries (RIRs) are simply out of IPv4 space, and only AFRNIC, the African RIR, has any IPv4 left to assign. The big milestone this year was when ARIN, the RIR for North America and part of the Caribbean, issued the last IPv4 address space in its free pool. There’s been a robust secondary market for IPv4 for years, where one can buy “used” address space from third parties as opposed to obtaining “new” address space assigned by an RIR. Prices for IPv4 are rising as the impact of RIR exhaustion and resulting scarcity sets in. We’re nearing the point where the cost of obtaining sufficient IPv4 space outweighs the accommodations necessary to just use IPv6. Some organizations reached that point a long time ago: carriers realized that large mobile deployments required IPv6 to have any hope of sufficient address space.
Right now, a lot of content and services are only available on IPv4. Content providers haven’t seen demand and they’ve actually been worried about making user experience worse by making content available over IPv6. As IPv6 support in operating systems and applications has increased, and as networks devices start to support it, devices obtain IPv6 addresses automatically. So-called “dual stack” devices with both IPv4 and IPv6 addresses are more common, but just because a device has an IPv6 address doesn’t mean that it can reach everywhere on the IPv6 Internet. In these early deployment days, sometimes IPv6 works only on the local network because the ISP or larger enterprise network doesn’t support IPv6. In such scenarios, an application can receive an IPv6 address in a DNS response and try to connect, only to frustrate users with timeouts or other failures.
The IETF (Internet Engineering Task Force), the standards organization responsible for the development of IPv6, realized this state of affairs was a big problem and hurting IPv6 deployment. The result was an algorithm governing application behavior published in 2012 called “Happy Eyeballs.” The idea is pretty simple: an application attempts to detect the state of IPv4 and IPv6 connectivity and use whichever transport performs best. This straightforward change simultaneously improved user experience and helped clear a path toward more IPv6 adoption by making content publishers less afraid to use IPv6. If your ISP supports IPv6 (e.g., Comcast for all broadband customers) and you’re using a modern web browser (e.g., the latest version of Chrome, Safari, Firefox, and Opera), you’re using Happy Eyeballs already.
Other factors are spurring IPv6 deployment. The Internet of Things (IoT) is a hot topic; every day it seems some new kind of device is connected to the Internet. But considering the IPv4 address space shortage, these devices—and there are going to be a lot of them—won’t use IPv4. They’ll leapfrog straight to IPv6. Using IPv6 for these devices has the potential to avoid the connectivity difficulties introduced by network address translation (NAT), which is ubiquitous in IPv4. In an IPv6 environment, true end-to-end connectivity is possible, which was the state of affairs in the early days of IPv4. While exactly how the Internet of Things will evolve is unclear, we can safely predict there will be an explosion of connected devices all wanting to communicate with one another, and IPv6 will make that not only possible but easier.
People like to refer to an IPv6 “transition,” but it’s really just a matter of adoption. The more we deploy IPv6-capable devices on IPv6-capable networks, the more content and services will become available via IPv6. More and more traffic will flow over IPv6 and IPv4 will diminish. But, just as a COBOL programmer can still make a living today, we’ll have IPv4 with us for a long time.
Considering the recent IPv4 address exhaustion milestones, the deployment of IPv6 for networks with large numbers of devices (like mobile and IoT), and the increased prevalence of Happy Eyeballs, I think we’ll look back on 2016 as a significant year in the deployment of IPv6. Maybe that’s a prediction after all.
This article is published as part of the IDG Contributor Network. Want to Join?