Universities and Underused IPv4 Address Space
By IPv4 Staff
May 11, 2023
Many universities have large pools of unused or under-utilized IPv4 address space. How much is it worth? Why do many universities have unused address space? How can they free it up?
How much is it worth?
IPv4 addresses ranged in value between $40 and $60 in 2022. A university that transfers about 65,000 IPv4 addresses (a common /16 block of addresses) can raise $3.5 million from a dormant asset.
Why is there unused address space?
The internet is an experiment that escaped the lab.
Early support for large-scale networking by the US government focused on Open Systems Interconnect (OSI), which was being developed by telecommunications companies. This system was complex and expensive. Standards were slow to develop even though there was a need for a workable system, especially among the academic and research communities.
In response, academia created the Internet Protocol to serve a simple but pressing need: identifying devices on a network and exchanging data among them accurately. The earliest device-identification and location system dates from 1973, with the creation the very first versions of addressing identifiers. However, it wasn’t until the development of IPv4 by ARPA in 1981 that a system (protocol) gained widespread use. Although it is called IPv4 it is actually the first “Internet Protocol” version assigned. Versions 0 and 1 are reserved and versions 2 and 3 were never assigned. This system – in use worldwide today – identified devices on the new network. IPv4 offers approximately 4.3 billion possible unique identifying number configurations, which was considered more than sufficient at the time.
In the early use of IPv4 – and so the distribution of addresses – was limited because use was among a relatively small group of researchers and academics. These users relied on a single record-keeper, Jon Postel, to keep track of who had which addresses. (He kept a notebook.) Organizations needed to be insiders to connect to the early networks, ARPANET and then NSFNET, so there was no reason to deny requests for addresses. There were, after all, billions. Far more than it was imagined might ever be needed. In 1991, the US government removed restrictions on who could connect to the early internet. Even then, there was no strong demand for IP addresses.
At that time, computing was expensive and slow. One consequence of this was a need to simplify routing protocols. Protocol developers did this by having just three sizes of network, which they called classes.
- Class A networks had about 16 million addresses
- Class B networks had just over 65,000 addresses
- Class C networks had 256 addresses
When organizations needed multiple Class C networks, they would get a Class B, even if they’d only use a small portion of it. An organization that needed just 2,000 IPv4 addresses would have needed eight Class C networks.
Many universities found themselves in this position. They needed a few thousand IPv4 addresses and so ended up with a Class B: about 65,000 addresses.
Because they did not need all those addresses, they could afford to put administrative convenience ahead of conservation. In some cases, this resulted in lots of unused or underused addresses in a patchwork of internal assignments.
Example of a Class B network with large amounts of free space distributed across it.
Universities and Colleges
Due to the early purpose of the Internet, research institutions were given large numbers of addresses. Which meant colleges and universities were distributed overly-large classes of addresses. Many have them today, under-utilized and very valuable.
CIDR
A more finely sub-divided system of address “blocks” was needed. The system that was created to respond to this problem, CIDR (Classless Inter-Domain Routing), is a routing system in which network engineers can distribute IP addresses based on the size of their specific network.
CIDR (pronounced ‘cider’) was introduced in the early 1990s and is a classless domain routing system. This new system empowers organizations to only get the amount of IP addresses they need in their networks, without wasting unused space. This is more efficient than the previous system because it doesn’t distribute excess addresses.
CIDR blocks permit owners of IPv4 addresses to subdivide and transfer (sell) there excess holdings if they possess a large “class.” The trick is that the numbered addresses one intends to sell must be consecutive. Transferred blocks have to have sequential integrity to have practical value.
For more information about CIDR, see the following:
How can it be freed up?
When looking to access the value in unused IPv4 address space it is worth looking at the alternatives to. They are:
- Do nothing and transfer the unused space in separate transactions based on whatever sequences are unused and available
- Renumber so that all used space is in one part of the block (sequentially) and transfer the remainder
- Renumber and look for ways to reduce your own use, so more space can be transferred
Do Nothing
It is possible to transfer blocks as small as 256 addresses. Universities could decide to leave things as they are and just transfer unused space. This could look attractive but it comes with costs. It requires more transfer transactions and adjusting routing and security policies. This means it could have higher risks and costs.
Renumber
Another approach is to place everything that needs IPv4 addresses in one part of the existing address block. This leaves a large and contiguous range of addresses available for transfer.
IPv4.Global can connect sellers with skilled consulting engineers who can help them renumber their network. The university would then have all its infrastructure in a single, continuous block and a large contiguous block available for transfer.
Larger blocks are attractive and likely to attract a premium, making the renumbering doubly valuable.
For more information on renumbering, see Renumbering IP Addresses.
Renumber +
The previous two options don’t change how much IPv4 address space the university uses. But it is now possible to reduce IPv4 address requirements. Integrating address sharing technologies – generally called NAT – with IPv6 can dramatically reduce the need for IPv4 addresses.
Client devices, like laptops or phones, have all been IPv6 ready for over a decade. Most client devices don’t need a permanent unique address. And security policies often don’t allow client devices to have a unique address, anyway.
Introducing address saving technologies alongside renumbering can more than pay for itself. Client LANs are generally far easier to renumber than routing and server infrastructure, so this can be a quick win.