I remember debating which protocol would dominate networking. Would IBM embrace IP over their SNA and Token Ring? Was Netware, then the dominate network system, going to support IP? Was Banyan Vines going to become more popular outside of certain government networks? Would ATM make the WAN and LAN distinction irrelevant? When IBM selected IP, the die was cast. And when the Internet became public, 25 years ago on April 30, 1993, the protocol debate was settled.
In 1993 no one thought the 4 billion possible IP addresses would be exhausted in their lifetime. The early pioneers snatched up millions of addresses just for their own use because they could, and no one really seemed to worry, there were billions of unused addresses. IP addresses themselves did not cause a goldrush.
With the evolving use of subnets and lower cost routers, one public IP address could be used to support a network of 10’s of thousands of users. So there was not a lot of worry about running out of addresses, even though work was proceeding on the next generation, IPV5 and IPV6.
But as the Internet exploded, so did the IP address use. By 2010, only about 10% of the addresses remained. And by 2016, we were pretty much out. No new addresses could be assigned. That is, an organization such as your ISP could not get any more to assign to you. They had to make due with the block of IP addresses they had already been assigned. The initial Internet vision of every host having it’s own IP address became impossible with IPV4.
With the Interent of Things, mobile devices, OTT streaming video players, tablets, and even vacum cleaners, toys, and toasters, IP addresses are in demand and there simply are not enough to go around, In fact, there are no more. Well, there are some more but they are second-hand. Harvesting unsued addresses from large organizations (MIT, IBM,and other early Internet pioneers), and carefully reusing existing inventory has allowed the IPV4 addresses to continue. But we are coming to the end and we’ve know it for a long time. That’s why there is the IPV6 address space.
At a recent Connecticut Educational Technology commission meeting, we talked about IPV6 migration.
Basically, we’re out of IPV4 addresses for the Connecticut Education Network (CEN), where I’ve played a role since its inception. And we are teaching network users how to migrate to IPV6.
IPV6 (IPV5 existed but never materialized) provides more address space than there are grains of sand in the desert. In fact there are 2 to the 128th power (340,282,366,920,938,463,463,374,607,431,768,211,456) IPV6 addresses. That’s 340 undecillion addresses. If you could scan a million addresses per second, it would take about three times the age of the universe to hit them all.
There are many reasons to switch to IPV6, including easy subnetting, easy filtering, no need for NATs, security is built-in, and much more. There are only two disadvantages that I can think of:
- I don’t think I can remember one (I can easily say “192.168.1.100” but not “2601:199:c200:aeb0:203c:de8e:d9bc:438c”)
- I have to learn something new
And of course, all of my network gear has to know how to work with IPV6.
As I write this, my ISP (Comcast) has provisioned my router with an IPV6 address. It never needs to change, so I will never need “DyDNS” or similar tools. My DNS provider supports IPV6 so any domain I own can be reached with an IPV6 address. Windows and Linux both support IPV6. And we’ll be adding IPV6 support to the Discover Video appliances soon. Maybe you are already using IPv6 at home but didn’t realize it.
If you are not already using it, brush up on IPV6. It’s not difficult.