The architecture of the Internet was developed in the late 70’s by the Internet Engineering Task Force (IETF), over the defined stack of TCP/IP protocols. The infrastructure of the Internet, since the TCP/IP model was applied in 1983, has grown at an accelerated rate. The different equipment manufacturers, the administrators and developers had to work a lot to be able to adapt to this growth. Approximately towards the end of the 80’s, and the beginning of the 90´s, due to the growth and popularity that the Internet started having, certain flaws in the design of the IPv4 protocol started to show.
Among the main faults, one can mention: the exhaustion of public addresses, the non existence of a standard security method, a poor implementation of quality of service scheme and headers of variable length with too detailed fields (some of them are no longer used).
These points, among others, encouraged the development of a new protocol, called IPv6. In it, the IPv4 problems detected are solved, and new functionalities are added to the protocol, taking into account previous experience.
Among the main advantages of the IPv6 protocol, we can find:
- Extended address space (128 bits).
- A simplification of the Header format (fixed routing Header, options in extension Headers).
- Auto-configuration of the IPv6 address in all the devices.
- Hierarchical addressing architecture, allowing agregation techniques (Global addresses).
- Native security scheme implemented in the protocol.
- Categorization of the traffic flow (new field to identify a certain traffic stream).
- Anycast reach addresses.
In the protocol, the link operations have been optimized and new mechanisms were developed, such as neighbors’ detection, multiple addresses per interface, lifespan of the addresses and interface ID assignment.
Although the IPv6 protocol is a well-known standard and it has been deployed in several environments successfully, the IPv6 deployment tends to be postponed in some operators since the migration to IPv6 is not an easy task. The IPv4 and IPv6 protocols are not compatible. The greatest difficulty is that this compatibility with IPv4 must be maintained while one migrates to IPv6 because migration is a process which is extended in time, and many services today are only offered for IPv4 (they have not migrated to IPv6 yet). The key to success in the migration process resides in such compatibility: in being able to use both protocols until all the services migrate to IPv6. Most Internet’s nodes are going to need to use both protocols for a long time. Because of this, it is very important to analyze the migration methods to IPv6 and to chose the most suitable one, depending on the operator’s network. It is known that any attempt to have a quick transaction is destined to fail and that gradual transaction schemes are needed.
The migration techniques can be gathered in three big categories:
- Dual-Stack:IPv4 and IPv6 coexist in the same devices and networks. These methods imply the use of both protocols in parallel in the devices.
- Translators: Translation from IPv4 to IPv6 and from IPv6 to IPv4 to allow the communication between them.
- Tunneling: Transport of IPv6 packets in IPv4 tunnels and IPv4 packets in IPv6 tunnels are ideal when the core doesn’t support IPv6.
Some recommendations have been defined for the IPv4/IPv6 coexistence and for the transition to IPv6. An example of migration techniques are Dual-Stack, NAT44(4), Stateful NAT64 + DNS64, 6rd, Dual Stack – Lite (DS-Lite), Gateway Initiated Dual Stack – Lite (GI-DS-Lite), 6PE and 6VPE.
The mechanisms defined are tools which can be combined and adapted to each situation. Nowadays, new migration schemes are being created to facilitate and accelerate the transition to full IPv6.
More information at: