IPv6 : prospects and problems : a technical and management investigation into the deployment of IPv6

Abstract

IPv4 has been used for over twenty years, and will most likely be used in many years ahead. However, we are now experiencing that the IPv4 address space is running out, resulting in restrictions on who will be able to get these types of addresses assigned to them. Methods such as Network Address Translator (NAT) have been developed and implemented in order to save the IPv4 address space. It is said that this is not a good enough solution, as such techniques introduce new problems at the same time solving some. A new version of the Internet Protocol, IPv6, has been developed and is likely to replace IPv4. IPv6 has been developed to solve the address problem, but also new features are designed to supposedly enhance network traffic. In our thesis we give an overview of the problems with IPv4. This includes the limited address space and the limited quality of service. Further we present the features of IPv6 that are meant to solve these problems and add new possibilities. These are: New address format, the IPv6 header and Extension headers to mention some. Further we have investigated and here present how the transition from IPv4 to IPv6 is expected to take place, followed by a thorough description of the transition mechanisms. One of the original intentions on the development of IPv6 was that IPv4 and IPv6 have to be able to coexist for a long period of time. Transition mechanisms have therefore been designed to make this possible. There are three main types of mechanisms: - Tunnelling - Translation - Dual-stack. Each of these mechanisms requires different configuration and implementations in hosts and network. Technical research on transition mechanisms states that these are not good enough for all IPv6/IPv4 scenarios and need improvements in order to make IPv4 and IPv6 coexist smoothly. There are a lot of transition mechanisms that are agreed upon as being good for general use and then there are transition mechanisms that are good for certain scenarios and not for others. Some scenarios still lack a good translation mechanism. As a result of this, IPv6 networks are being built separately from IPv4 networks. In Asia commercial IPv6 networks are offered, while the process is slower in other parts of the world. The reasons for not building IPv6 networks are many, and not agreed upon. Some believe it is because of economical restrictions, while others claim it is technical reasons and that it exists far too few applications supporting IPv6. The number of IPv6 enabled applications is growing. Large companies like; Microsoft Corporation, Cisco Systems Inc, Apple Computers Inc., Sun Microsystems Inc and various versions of Linux include support for IPv6. The deployment of IPv6 is expected to happen at different times in different parts of the world. We have investigated the status of IPv6 globally and in Norway. The main results are that the roll-out has reached the furthest in Asia where commercial IPv6 networks already are offered. The activity in Norway is still small, but growing. It was desired to run an experiment in order to prove or disprove some of the information we gathered on how IPv6 interoperates with IPv4, but because of limitations in the network at Heriot-Watt University we were not able to do this. Instead we have focused on a project by Telenor R&D; “IPv6 migration of unmanaged networks-The Tromsø IPv6 Pilot”. We also gathered some information from people working at Norwegian ISPs in order to address some of the aspects of the upgrading.