How I Learned to Stop Worrying About IPv4’s Limitation and Love IPv6

If you remember back before the Internet was a “thing,” first of all don’t tell anyone – you’ll really date yourself. Seriously, for those of us who witnessed the birth of the “World Wide Web” and the phenomenal growth of the Internet, it’s hard to believe that we’ve exceeded the capacity of the original IPv4 addressing structure. You’ll recall that IPv4 employs a 32-bit system that results in more than 4 billion unique 16-digit addresses. Where have they all gone?

It’s estimated that nearly 3.9 billion people of the almost 7.6 billion human inhabitants in the world have Internet access (as of June 2017). But remember, in the developed world, it’s likely that each person has more than one device that connects to the Internet, each requiring a unique IP address. Cisco reports that 429 million mobile devices and connections were added in 2016. Mobile-connected tablets also increased 26% to 184 million, and the number of mobile-connected PCs increased 8% to 136 million.

Organizations around the world use up a fair share of IPv4 addresses—there are 27 entities, including the U.S. Department of Defense, that hold about 671 million addresses. And now with the Internet of Things (IoT) coming online to the tune of 8.4 billion connected things this year —projected to grow to 20.4 billion devices by 2020—you can see how we’ve outgrown IPv4 quite easily.

Lucky for us, the smart people at the Internet Engineering Task Force (IETF) foresaw decades ago that we were destined to need a new Internet addressing system with exponentially more room for growth. In 1998, IETF published the basic IPv6 protocol (RFC 2460), and has been working to refine and update it ever since. Now IPv6 is a full Internet standard documented with IETF RFC 8200.

  • Basic protocol (RFC 2460) published in 1998
  • Basic socket API (RFC 2553) and DHCPv6 (RFC 3315) published in 2003.
  • Mobile IPv6 (RFC 3775) published in 2004
  • Flow label specifications (RFC 3697) added 2004
  • Address architecture (RFC 4291) stable, minor revision in 2006
  • Node requirements (RFC 4294) published 2006
  • IPv6 promoted to a full Internet standard (RFC 8200) published 2017


IPv6 addresses are 128 bits long, which allows an astounding 2128 addresses or 340,282,366,920,938,463,463,374,607,431,768,211,456 addresses (3.4X1038 or approximately 340 undecillion). It’s safe to say that we should be in good shape for some time!

Although many organizations have been working to adopt IPv6, others are still behind the curve. There are numerous reasons why this is. In some cases, it’s simply due to a lack of knowledge in the organization about how IPv6 works. There may even be some unconscious fear – after all, we’ve all come to understand the format of IPv4 addresses and how they work. In comparison, IPv6 addresses seem like an unwieldy beast, and more like a MAC address than a networking protocol – case in point… 2001:0DB8:0012:0034:0000:0000:0000:1111.

The good news is that there is no reason to fear or avoid IPv6 addressing. It’s actually pretty simple to understand, and is even easier to implement than IPv4 because you don’t have to deal with subnets.

In my latest article for Network World, I’ve explained IPv6 addressing for enterprises. When you’re got that under your belt, I recommend this IPv6 deployment guide that will give you the high-level steps to start your migration toward IPv6 and remove your dependency of the limits of IPv4.

If you work at an enterprise IT shop and you have not yet start to plan for and implement IPv6, then now is the time to start this process. You would not be doing your organization any favors to further delay IPv6 deployment. If your organization ends up being an IPv6 technology laggard, there are undesirable side-effects.

For even more educational posts about IPv6, explore the GTRI blog. GTRI also offers IPv6 training.

Scott Hogg is the Chief Technology Officer (CTO) for GTRI.