How to Transition to IPv6
Many events take place during the summer Olympic Games, which happened in London during the late summer of 2012. Even the most casual observer of the competition would notice the intense effort involved, particularly in the team-oriented events such as track and field. Critical to the success of the relay race is the passing of the baton, in which the racer for a specific leg of the race finishes their segment and transfers ownership of the race to the next participant of their team. The world of information technology is experiencing this exact type of shift as it moves from the familiar comfort of IP Version 4 into the “brave, new world” of IP Version 6.
Effective transitions seldom take place overnight, and this includes network technologies as well. Nearly always, when installing new components into an infrastructure, there is a period of migration and/or coexistence before committing fully to the new environment. IPv6 is no different, which has led to the development of various mechanisms designed to phase the protocol into existing networks. This two-part series will introduce and discuss the major options and, in some cases, the subcategories of these transitional mechanisms.
One of the very first approaches used with regard to IPv6 is the dual-stack solution, which simply means that both IPv4 and IPv6 protocols exist simultaneously on devices on the network. IP addressing for each protocol is configured on each LAN workstation as well as each device in the data path (routers and switches do not show the configured addresses). Each protocol exists separately from the other; no interaction between the two takes place in this model.
Advantages of Dual-Stack
Deployment of a dual IPv4 and IPv6 solution has distinct advantages over other potential solutions. To begin with, use of the respective IP stacks means that no tunneling mechanisms in the internal network are required, nor are the headaches than can occur when using them. Second, using this solution allows for migrating devices on a more gradual basis, rather than all at once. Third, most network equipment manufacturers, including Cisco, recommend dual-stack as the most effective transition mechanism.
Disadvantages of Dual-Stack
Deployment of a side-by-side solution with dual protocols instantly creates a greater amount of complexity for the individuals and teams supporting the network. This automatically adds additional work to the operational environment, both in terms of personnel and equipment. Machines using the dual-stacks will experience greater loads because of the additional processing involved.
Network Address Translation (NAT) Protocol
When the address depletion with IPv4 became self-evident, the introduction of RFC 1918 Private Addressing along with NAT helped to solve the immediate problem. One of the many advantages of IPv6 is that with a vast address space at its disposal, NAT would no longer be necessary for daily operations. That being said, NAT is a solution for transition from IPv4 to IPv6. The same dynamics of traditional NAT exist for NAT-PT, except in this case translation is taking place between each respective protocol.
Advantages of NAT-PT
As was the case with traditional NAT, deployment of the address translation process can be isolated to the router or firewall performing this function, avoiding the need to involve every device on the network. Another advantage is that IPv6-only hosts have the capability to interact with IPv4-only hosts and vice-versa.
Disadvantages of NAT-PT
The original conception of the Internet Protocol envisioned end-to-end communication between hosts, with no intervening devices involved, a concept clearly broken through the use of NAT, regardless of the specific version involved. Another disadvantage of this transition mechanism is that it introduces a single point of failure (the NAT device), which goes against best design principles. Another problem with NAT-PT is that it provides basic connectivity and neglects the more advanced features of IPv6. While this transition mechanism was among the first to be proposed and/or used, it is generally discouraged for use. It should be noted that another version of NAT, referred to as NAT64, is available, but has some of the same disadvantages previously discussed.
Next week we’ll finish with Tunneling.
Reproduced from IP Version 6 Transitions Mechanisms
IPv6 Transitions Mechanisms Series
- How to Transition to IPv6
- Using Tunneling to Transition to IPv6