Internet Protocol version 4 (IPv4) is the address system that we use today - you may be familiar with addresses that look like this:
The numbering system is hierarchical and provides a way to uniquely identify every computed on the network - about 4 billion devices can be identified in this way. At one time this might have seemed like enough but our modern world now needs far more than 4.3 billion network addresses. All the organisations responsible for allocating IP addresses have pretty much run-out: RIPE NCC, the regional Internet registry for Europe, was the second RIR to deplete its address pool on 14 September 2012. So getting any new IPv4 addresses is difficult and there is a market in the re-use of existing addresses. For example: Microsoft paid $7.5 million to acquire 666,624 IPv4 addresses from Nortel's during their liquidation sale.
The solution to the problem is IPv6 which increases the address space very significantly - enough address space for many decades to come. But the trouble is that IPv6 and IPv4 are not inter-operable. The widespread adoption of IPv6 is going to be a slow and an expensive business. To give you an idea of how slow IPv6 adoption is consider this: less than 1% of traffic reaching Google is currently using IPv6.
Note: If you are wondering why we have jumped from IPv4 to IPv6 the reason is that the denomination IPv5 was used for an experimental protocol for streaming so to avoid any possible confusion we now have IPv6. (IPv1, 2 and 3 where also experimental and never widely adopted).