IPv4 to IPv6 Transition Strategies and Deployments CCIE Security Coaching Institute in New Delhi

Network Bulls
www.networkbulls.com
Best Institute for CCNA CCNP CCSP CCIP CCIE Training in India
M-44, Old Dlf, Sector-14 Gurgaon, Haryana, India
Call: +91-9654672192

Several deployment models exist to migrate from an IPv4 network to IPv6. During a transition time, both protocols can coexist in the network. The deployment models are

• IPv6 over dedicated WAN links
• IPv6 over IPv4 tunnels
• IPv6 using dual-stack backbones
• Protocol translation

Each model provides several advantages and disadvantages with which you should become familiar. The following sections describe each model.

IPv6 over Dedicated WAN Links

In this deployment model, all nodes and links use IPv6 hierarchy, addressing, and protocols. It is not a transition model, but a new, separate deployment of IPv6. The WAN in this model uses IPv6. The disadvantage of this model is that additional costs are incurred when separate links are used for IPv6 WAN circuits during the transition to using IPv6 exclusively. As shown in Figure 8-6, a company needs both IPv6 and IPv4 networks in sites A and B during the IPv6 deployment and transition. The networks are connected using separate WANs.

IPv6 over IPv4 Tunnels

In this deployment model, pockets of IPv6-only networks are connected using IPv4 tunnels. With tunneling, IPv6 traffic is encapsulated within IPv4 packets so that they are sent over the IPv4 WAN. The advantage of this method is that you do not need separate circuits to connect the IPv6 networks. A disadvantage of this method is the increased protocol overhead of the encapsulated IPv6 headers. Tunnels are created manually, semiautomatedly, or automatically using 6to4.

RFC 3056 specifies the 6to4 method for transition by assigning an interim unique IPv6 prefix. 2002::/16 is the assigned range for 6to4. Each 6to4 site uses a /48 prefix that is concatenated with 2002.

Figure 8-7 shows a network using IPv4 tunnels. Site A and Site B both have IPv4 and IPv6 networks. The IPv6 networks are connected using an IPv4 tunnel in the WAN.

Dual-Stack Backbones

In this model, all routers in the backbone are dual-stack, capable of routing both IPv4 and IPv6 packets. The IPv4 protocol stack is used between IPv4 hosts, and the IPv6 protocol stack is used between IPv6 hosts. This deployment model works for organizations with a mixture of IPv4 and IPv6 applications. Figure 8-8 shows a network with a dual-stack backbone. All the WAN routers run both IPv4 and IPv6 routing protocols. The disadvantages are that the WAN routers require dual addressing, run two routing protocols, and might require additional CPU and memory resources. Another disadvantage is that IPv4-only and IPv6-only hosts cannot communicate with each other directly; dual-stack hosts or network translation is required (covered next) for IPv4 and IPv6 hosts to communicate.

Dual-Stack Hosts

Hosts require dual stacks (IPv4 and IPv6) to communicate with both IPv4 and IPv6 hosts. In this environment, host applications can communicate with both IPv4 and IPv6 stacks. When using dual stacks, a host uses DNS to determine which stack to use to reach a destination. If DNS returns an IPv6 (A6 record) address to the host, the host uses the IPv6 stack. If DNS returns an IPv4 (A record) address to the host, the host uses the IPv4 stack. Using dual stacks is the method recommended for campus and access networks during a transition to IPv6.

Protocol Translation Mechanisms

One of the mechanisms for an IPv6-only host to communicate with an IPv4-only host without using dual stacks is protocol translation. RFC 2766 describes NAT-PT, which provides translation between IPv6 and IPv4 hosts. NAT-PT operates similarly to the NAT mechanisms to translate IPv4 private addresses to public address space. NAT-PT binds addresses in the IPv6 network to addresses in the IPv4 network and vice versa. Figure 8-9 shows a network using NAT-PT.