Despite IPv6 adoption, IPv4 Still Hangs in There – Maybe Forever

IPv4 Exhaustion Continues but will Co-exist with IPv6 for Years to Come

It’s been 22 years since IPv6 was introduced as a draft standard by the IETF. Why are we still talking about IPv4? Hasn’t it died already?

Well….no, IPV4 is still very much alive and despite IPv4 exhaustion, many, many organizations, and users continue to use IPv4 now and likely will many years into the future. IPv4 and IPv6 will coexist for years.

Mark Twain once said, “the report of my death was an exaggeration.” The text was from a cable sent by Mark Twain from London to the press in the United States after his obituary had been mistakenly published. May of 1897 the English correspondent for the New York Journal, Frank Marshall White, contacted Twain in London to inquire about his health, and then later, to ask for comment about reports that Twain was on his deathbed.

The same is true for IPv4. Globally, IPv4 exhaustion has been reached with the free pools of IPv4 addresses from the Internet Regional Registries (IRR) fully allocated. However, despite considerable effort to encourage IPv6 adoption, IPv4 is still very much alive. IPv6 adoption has definitely accelerated in the last few years but is not the dominant protocol used.

So, IPv4 is fighting back…. it’s not dead, but it has become very expensive. IPv4 exhaustion can be resolved by organizations purchasing more addresses on the open market, now reaching prices of $24 each. Many organizations including higher education, enterprise and ISPs and rural cable operators have existing IPv4 address pools and infrastructure that they must continue to leverage. So, solving IPv4 exhaustion with CGNAT solutions makes sound business sense.

Full IPv6 adoption is actually the summation of three independent technology drivers – devices, networks and content (web sites). If any of these are not fully IPv6, the two addressing schemes will co-exist. Consider the following:

• Organizations can solve IPv4 exhaustion with CGNAT. Many universities, for example were provided a large set of “free” IPv4 address pools years ago and subsequently built their network infrastructure to be IPv4 compatible. This allocation has been strained as students and faculty now have five or more devices requiring connectivity to university networks and resources. CGNAT, with NAT44 or NAT444 can expand IP address pools by 40-60X, helping budget-constrained organizations extend investment without purchasing new IPv4 numbers on the open market.
• Connected devices and applications are not all IPv6 compatible. IPv6 support has been mandatory in both Android and iOS devices for years, but there are many older devices still connected. Old technologies do not replace new ones, they just overlap. For example, while newer mobile devices and laptops are IPv6 compatible, most 3G and all 2G devices were not. Consider that of the 8.9B mobile subscriptions forecast by 2025, 1.7B will still be 2G/3G devices, according to Ericsson. In addition, older applications and devices may not be IPv6 compatible. For example, some rural cable operators have postponed IPv6 due to incompatibility of home cable routers, where the cost to change out those devices is too high and the process too disruptive to subscribers.
• Subscribers, employees, students, users still want to access IPv4 only sites. Today, while high-traffic web destinations such as Google, Yahoo, Wikipedia, Facebook, Netflix and YouTube all support IPv6, only 17 percent of websites use IPv6, and globally, only about 30 percent of Google searches use IPv6. That means that almost two-thirds of Google queries are accessing IPv4-only websites.
• Many organizations cannot justify the near-term cost and disruption that a data center and network change-out to IPv6 will entail. Switching to IPv6 is costly and time consuming. All connected devices must be inventoried and changed out or reconfigured. There is a risk that a needed device or application will not work and cause service disruption that is time consuming to troubleshoot and fix. An IPv6 transition takes a great deal of detail-oriented effort by network administrators for testing and production, and in some cases, they must re-architect entire networks. Balanced against the daily demands plus the need to move forward on strategic initiatives like 5G, cloud, virtualization, edge, and others, IPv6 can be a high-effort, low-return project. Since 2017, the DoD, for example, has attempted twice to transition to IPv6 and halted both times due to security concerns and lack of trained personnel.
• Even for service providers that have moved to IPv6 within their networks, the transition is slow. Service providers, especially mobile network operators expect significant cost benefits from simplifying their networks with IPv6. The top tier-one mobile network operators have all aggressively changed to IPv6, with T-Mobile, Verizon, AT&T in the U.S. at 70 percent + of traffic and Reliance, Chunghwa, BT and others in the top-20 exceeding 60 percent of traffic. However, of the 351 service provider networks measured, only 100 of them had greater than 50 percent of their traffic on IPv6.

All organizations must balance rapidly increasing IPv6 devices and traffic volume against other network technology initiatives such as software-defined networking, cloud and edge. Connected devices, including cellular IoT devices are expected to exceed 24 billion by 2025 – doubling quantities in 2019. The additional 10B + new devices will likely be IPv6. So, most organizations have to manage a growing base of newer IPv6-enabled devices with older IPv4 devices connecting to both IPv4 and IPv6 content. The two environments will have to co-exist for some time and to accommodate the other technology transitions that are needed.

The real question is not so much the speed of IPv6 adoption, but how long IPv4 will hang in there. As described in a recent Heavy Reading survey and webinar, 80 percent of operators see CGNAT as a required function, even in their cloud-native, 5G standalone (SA) networks that they will be deploying over the next three years. Although the relative percentage of traffic that remains IPv4 certainly gets smaller, traffic is growing so rapidly with so many new devices that operators are having to augment older technology even as they plan and build out the new – just to keep up with demand.

A10 Networks provides CGNAT technology to service providers, higher education, government, and enterprise. The A10 Thunder® CFW solution includes CGNAT, IPv4 to IPv6 migration and core network and data center firewall functions and solves IPv4 exhaustion while providing a path to IPv6. The Thunder CFW solution is still very much in demand for its high performance, completeness of features and all-in-one licensing model. The solution enables organizations to extend depleted IPv4 address pools and infrastructure, build a seamless transition path for IPv6 migration, and secure growing volumes of vulnerable IPv6 addresses. Our customers continue to include Thunder CGN in their existing data centers and as they move to edge, mobile edge computing (MEC), cloud-native infrastructures and 5G.

How long will IPv4 hang in there? It is going to be awhile. In the meantime, A10 Networks is continuing to provide both CGNAT and IPv4 to IPv6 migration technology that meets our customer needs now and into future network evolutions.

For more information on Thunder CGN, please visit  https://www.a10networks.com/products/thunder-cgn/

Share this post:

Share on X (Twitter)Share on FacebookShare on LinkedInShare on Email

Categories:

5G Cloud & Networking Rural Broadband

---

Previous Post

Kostas Katsaros on Edge Computing, IIoT and Mobile Networks

Next Post

Fast Detection Protects Critical MEC Services from DDoS Attacks

---

Terry Young

|

November 17, 2020