The Internet Protocol: Past, Present, and Future – Part 3
In the final part in our series on the Internet Protocol, we discuss measures being taken to mitigate the exhaustion of IPv4 addresses, as well as IPv6.
Previously, we talked about the birth and history of the Internet Protocol, the communication protocol that allows the internet as we know it to exist, and the creation of and problems with the current version of the protocol, IPv4. In this final article, I’m going to talk about some attempts to mitigate the exhaustion of IPv4 addresses, as well as IPv6, recently activated and trying to gain traction as the solution to the problem.
Slowing the Death of IPv4
Rather than roll out new infrastructure to support the new protocol version, many institutions and ISPs are instead looking to ways to mitigate the exhaustion of IPv4 addresses. Whether through subnetting or the reassignment of unused addresses, they’re looking to swell the remaining pool of available addresses in order to slow the death of IPv4. While this doesn’t directly deal with the root of the problem, let’s discuss these solutions.
First is subnetting. Many people have asked me why their IP address looks different between their home networks and the address the rest of the internet sees. The reason for this is what’s known as “subnetting.” In essence, a DHCP server, or “Dynamic Host Configuration Protocol”, assigns an IP address from a very limited range on a private network so that communications between computers on the network can occur, while all outbound traffic is sent through the router with a single IP address. This allows people to have dozens of devices connected to their home network without the ISP needing to assign dozens of IP addresses to the home; instead, they can assign a single IP address, greatly reducing the number of addresses they need to hand out.
Subnets are defined as networks within a specific range of IP addresses, usually denoted with CIDR notation. The most common subnet is the class-C subnet, with a subnet mask of 255.255.255.0. This means that the first three bytes of the address are unchanging, with the last byte the differentiating part of the address; it can also be written 192.168.1.0/24, meaning that 24 bits are used only for the network prefix, with the host being identified in the final 8 bits. This subnet is the most common, because it is used in nearly every private home network.
Subnetting, however, only slows the need for addresses, allowing a single address to be defined for an entire network. This opens up far more addresses, though certain ranges are off-limits and servers still need a unique public IP address, bringing us to the next stop-gap, reclaiming addresses. Many addresses that were assigned to organizations in the early days of IPv4 have still been unused, but cannot be assigned to anything else because they’re held in reserve for that organization. Others were assigned and used, but the server assigned to it has gone offline, leaving the address to sit unused but assigned. ISPs and regional IP assignment organizations have been culling these unused and abandoned addresses and returning them to the regional pools of addresses, ready to be reassigned as needed. However, even this solution only kicks the proverbial can down the road.
IPv6 – The Solution
The ultimate solution to the address exhaustion problem is the new version of the Internet Protocol, IPv6. IPv6 is an expanded address space that allows for far more addresses than IPv4, effectively eliminating the exhaustion problem, barring some massive uptick in networked devices (entirely possible in the next few decades).
Unlike IPv4, which uses a 32-bit address space, IPv6 uses a 128-bit address space, allowing for 2128 addresses, or roughly 34 followed by 37 zeroes. To put this into perspective, this is enough to give every person on the planet, roughly 7 billion people, their own personal set of addresses the size of the current IPv4 address space.
Let that sink in for a minute.
In essence, IPv6 eliminates the need for more address space. There is no way with modern technology that the amount of addresses allowed by IPv6 will be exhausted. It is quite possible that, given advances in technology and miniaturization, one day IPv6 will be used up, but until then it more than meets our requirements.