Will IPv6 ever kill IPv4? If so, when?

A friend of mine almost had a heart attack recently when he stumbled across this site on the Internet. My friend lives in Canada’s largest city and he manages the datacenter of a fairly large manufacturing enterprise that has many mission-critical legacy systems hosted on their network. The site referred to is called “IPv4 Flag Day” and it says that several software vendors, Internet service providers, and content providers are planning on removing IPv4 backward compatibility from their products and services starting Feb. 1, 2030.

Yikes, that’s only 11 years away! From a large business perspective that’s like the blink of an eye!

A little research on Wikipedia indicates that this is actually a parody site, though from the discussion on Reddit you probably wouldn’t catch on to that unless you read very carefully. Still, it’s maybe not such a bad idea to contemplate having us IT pros someday shut down IPv4 addressing, both on the Internet and also within corporate networks around the world. After all, IPv6 is so much bigger and better than IPv4, isn’t it? So why not just give everyone a push to migrate?

IPv4 vs. IPv6: The mobile factor

IPv4 addresses

Some big companies have heard the siren call of IPv6 and have jumped headfirst into the waters. Examples of companies like this that come to mind are Google, Facebook, and Netflix. What has motivated these companies to migrate their infrastructures and services so they focus almost exclusively on IPv6? The habits of those who consume the services these companies offer.

The last half-decade or so has seen an unmistakable decline in the PC market as mobile phones have become the preferred platform for people to communicate, shop, bank and watch entertainment. And while PC sales declines seem to have finally leveled off in the last couple of years, this is largely due to businesses refreshing obsolete legacy hardware as opposed to actual growth in many of the businesses themselves.

Google and Facebook were also founded as companies in the late 1990s and early 2000s, respectively, at a time when migration from IPv4 to IPv6 was seen as being inevitable for businesses because of the anticipated exhaustion of IPv4 address space. The tech media, which often tries to hype things up to win eyeballs, often painted the impending exhaustion of the IPv4 address space as of comparable importance to the Y2K disaster. But the Internet didn’t blow up when we ran out of IPv4 addresses, something that didn’t actually happen until a few years later than it was earlier predicted to have occurred — and which in actuality even today hasn’t yet happened since you can still buy and sell IPv4 addresses on the IPv4 resale market. Being forward-thinking, however, companies like Google and Facebook planned early to move everything to IPv6 as quickly as possible to ensure their continued growth as businesses as the mobile device marked exploded toward the point where IPv4 addressing could no longer support the number of phone-carrying users of their services.

Large telecommunications service providers followed closely behind the footsteps of pioneering tech companies like Google. Starting early on in this decade, providers began implementing plans for enabling delivery of their telecommunication services over IPv6 instead of just IPv4. For example, Telstra the largest telecommunications service provider in Australia began, offering IPv6 Internet connectivity in 2012 and finished implementing delivery of IPv6 mobile services in 2017. T-Mobile, which provides telecommunications services in the USA and parts of Europe also caught on early to the need to offer IPv6 connectivity for mobile users so they began seriously rolling out IPv6 on their network in late 2013 and early 2014. British Telecom (now BT Group) finished their IPv6 rollout for broadband customers late in 2016 (see PDF). And in South Korea service provider SK Telecom which has more than 30 million mobile customers now provides full IPv6 support for their LTE subscribers — in fact more than half of all the network traffic SK Telecom carries is now IPv6. Similar stories can be told for other providers like Comcast, AT&A, and Verizon.

IPv4 vs. IPv6: The cost factor

IPv4 resale
Shutterstock

Still, there are holdouts, especially in the corporate world. Many large and midsized companies still haven’t finalized or adopted plans for migrating their internal networks from IPv4 to IPv6. The reason for such resistance is simply market forces at work. Businesses have to justify large expenditures before they open their wallets to pay for them. And migrating a complex heterogeneous network with multiple sites and hard-to-replace legacy systems to IPv6 can be neither cheap nor easy. If you run a business then you need a good reason to spend money. One good reason, therefore, for offering IPv6 is because your customers demand it. Google, Facebook, and Netflix foresaw that IPv6 would eventually be needed to keep their customer base growing and ensure their customers were happy. Large telecoms also saw the writing on the wall in this regard, as do most large cloud computing companies.

Paradoxically, however, the very fact that companies offering cloud services now mostly support IPv6 has reduced the urgency for many customers to migrate their network infrastructures to IPv6. After all, why should I need to put my company though the hardship of moving our network, systems, and applications to IPv6 if I can simply offload that need by moving our workloads and apps into Amazon’s cloud? Even though our business is growing rapidly, I really don’t need any new IPv4 addresses, so why should I care whether IPv4 addresses have run out?

A slow walk

And finally, there are the bean counters, the ones who hold control over how much gets spent on what at your company. It may be the case that completely replacing IPv4 with IPv6 internally and only keeping IPv4 as a service at the edge of your corporate network may result in significant cost savings over time. But while arguments for IPv6 improving your ROI can, with some difficulty, be presented in a convincing fashion to your finance team, the unfortunate reality of the situation is that some of that ROI can only be recouped when the other parties your business deals with — your partners and your customers — also go full-on IPv6 themselves.

And that may take a long, long time to happen.

If ever.

Featured image: Shutterstock

About The Author

1 thought on “Will IPv6 ever kill IPv4? If so, when?”

Leave a Comment

Your email address will not be published. Required fields are marked *

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Scroll to Top