We have spent years, decades actually, listening for IPv4 addresses running out. I seem to remember that the first time I read about it was in the very, very late 1990s or, at the latest, between 2000 and 2001. And it is true that, more than 20 years later, in some regions of the world such shortages are still not It has been noticeable in a marked way, but although it sounds remote to us, in others it has become a real problem, forcing the adoption of IPv6.
As a quick reminder, IP addresses are the unique identifier of a device on a TCP/IP-based network. Currently, the majority we use IPv4, in which IP addresses are made up of 32 bits divided into four blocks of one byte each. That is, each of the possible permutations between 00000000.00000000.00000000.00000000 and 11111111.11111111.11111111.11111111. To make them more readable, they are usually expressed in base 10, rather than base two. Thus, for example, the more than usual 192.168.0.1 is, in base 2 (binary) 11000000.10101000.00000000.00000001.
If we count we will see that there are a total of 4,294,967,296 IP addresses, a number that when IP was defined seemed more than enough. However, there are two factors that we must take into account to understand the current situation. The first is that the volume of devices connected to the Internet today was simply unimaginable in the late 1970s and early 1980s. And the other important point is that, when it started to be used, several ranges were reserved for specific uses, so they cannot be assigned to connections.
In this Wikipedia entry you can consult all the IPv4 reserved ranges, which determine the IP addresses that cannot be used. And although today maintaining some of these ranges makes all the sense in the world, since they are used on a regular basis, such as those reserved for private networks. Others, however, have ceased to make senseand freeing their IP addresses would be an important relief.
And that’s what Seth Schoen, co-founder of Let’s Encrypt, is working on. intends to release the ranges 240/4, 0/8, 127/8 and 225/8-232/8, reserved since the initial implementation of the protocol, and that if added to the available IPv4 addresses, they would mean around 419 million additional addresses. This is not a solution to the problem, but at least for now it would mean a relief and, moreover, a more efficient use of IPv4.
In the middle, and above all in the long term, the only real solution is the transition from IPv4 to IPv6, and in this case the IP addresses are made up of 16 bytes, that is, 128 bits, 64 to identify the network number and the other 64 for the device identified with it. This gives a total of 10 28that is, 79,228,162,514,264,337,593,543,950,336 IPv6 addresses, more than enough even in the most optimistic scenarios regarding IoT deployment.
