Back in late 2015, I had drilled down on the dilemma of the different versions of Internet Protocol addresses, known as IP version X (IPvX). I penned a missive on how close we were to running out of the fourth version “IPv4” addresses (that will be coming up shortly as a Tech Talk). At that time there were about 17 million IPv4 addresses left. Today, that number has decreased to roughly 1.2 million, according to the RIPE network coordination center. Considering there are nearly 4.3 billion IP addresses available, that 1.2 million is fly spit in the Pacific Ocean. Literally, we are out of IPv4 addresses for all intents and purposes.
It is not “the IPv4 sky is falling” time, although some seem to think it is. Various organizations that serve the IP market, such as the Asia Pacific Network Information Centre (APNIC) ran out, and the Internet Assigned Numbers Authority (IANA) free pool was exhausted way back in 2011. The same is true for some other, similar organizations.
However, there have been various stave-off technologies developed which have offered some breathing room. For example, there are special reserve pools. And v4 addresses are also returned to the pool. There are reserved addresses as well. There is also some movement to reclaim/reuse unused and hoarded addresses, particularly the old school /8 numbers that were claimed early on and held for future usage. However, this is like plugging holes in a dike that is about to blow.
As well, technology is coming to the rescue. Network Address Translation (NAT) and Port Address Translation (PAT) have made it possible to survive this far. As the names state, this technique translates a particular v4 address into another v4 address, where there is one instance of NAT44 used by the broadband Internet access subscriber CPE device. More on that in the Tech Talk.
There is also the ability to translate v4 into v6, but that is somewhat problematic because they are not entirely compatible. For example, a v6 network is not able to browse a site running on a v4 web server without translation.
Those in the thick of the IP address game have turned to NAT as a way to avoid transitioning to v6. As I mentioned earlier, if one is heavily invested in v4, adding v6 has its challenges. And, NAT has given them some breathing room. But make no mistake, at some point everyone is going to play in the v6 arena.
Why this is of more concern now than a few years ago, when this hit the first critical knee, is that we are on the verge of deploying a whole slew of new devices. Platforms such as 5G, the Internet of Anything/Everything (IoX), autonomous vehicles, smart “X”, motes, the list goes on and on and most, if not all will require IP addresses.
Much of this is still in various stages of development – meaning the drain on IP addresses is small at this time. However, once things like the IoX start to become prolific, with billions of new devices deploying, we really are going to run out of the v4 addresses. The only question is when.
So, with all of this on the table, why has IPv6 been so slow in uptake? Not that it is all that new. It has been around since 1988. The industry saw this coming years ago, and thusly developed v6.
With all that is coming down the pike, migration is starting to pick up momentum. Once the aforementioned platforms start to proliferate, so will the adoption of v6.
IPv6 adds a number of new protocols that are necessary, especially for the IoX, including:
· Internet Control Message Protocol version 6 (ICMPv6). Responsible for providing diagnostics and error reporting. Replaces ICMPv.
· Multicast Listener Discovery (MLD). This is a series of three ICMPv6 messages that replace version 2 of the Internet Group Management Protocol (IGMP) for IPv4 to manage subnet multicast membership.
· Neighbor Discovery. Responsible for the interaction of neighboring nodes and includes message exchanges for address resolution, duplicate address detection, router discovery and router redirects. Replaces Address Resolution Protocol (ARP), ICMPv4 Router Discovery, and the ICMPv4 Redirect message.
As well, and, as to be expected, v6 is more efficient at packet handling, and offers better overall performance, including security enhancements.
There is, obviously, some deployment of v6 going on. ISPs and carriers top the list, especially in the mobile network area. For example, 90 percent of T-Mobile traffic is going over v6. Verizon has just over 80 percent on v6. Comcast and AT&T have their networks at 63 percent and 65 percent, respectively, according to World Ipv6 Launch.
Next, come the big website holders. Alexa, for example, says that nearly 30 percent of their top 1,000 websites can be reached over a v6 network. Other segments, such as the enterprise have been much slower to adopt, however.
But at some point, economics will come into play. The usual rule of the less of something is available, the more it will cost will apply. That will begin happening with v4 addresses soon. It may even turn out that the cost of v4 will be that much more expensive than v6, incentivizing v6 adoption.
Will we ever truly run out of v4 addresses? Eventually, of course. However, for the time being, the technical sleight of hand of technologies such as NAT, coupled with reuse, reserve and the like will keep the pool available at least until next-generation platforms get serious.
One good move would be for a collective effort of the major uses to switch to v6 at some particular point in time, collectively. That way the disruption would be more of a spike than a period. I have heard some rumors regarding that.
Personally, I am all in on using v6. Performance is better, security is better, and I do not have to worry that my v6 addresses will become high-maintenance or hijacked because all of a sudden they have a high value.