NAT (Network Address Translation) was proposed in 1994. When some hosts in the private network have been assigned local IP addresses, but now want to communicate with hosts on the Internet, NAT can be used. NAT is defined in RFC 1631.
The original purpose of NAT is similar to CIDR, and it is also to slow down the depletion of available IP address space. The implementation method is to use a small number of public IP addresses to represent a large number of private IP addresses. Over time, people have found NAT very useful for network migration, network convergence, server load sharing and other applications.
IPv4 was first created in the 1970s, earlier than the current Internet, earlier than the World Wide Web, earlier than online ubiquitous broadband services, earlier than smartphones. At the beginning of its creation, IPv4 has 4.3 billion addresses, which are extremely rich for the negligible experimental TCP/IP network to be supported. However, the number of people connecting to the Internet has exceeded 3.2 billion, and there are a large number of other devices connected to the Internet.
No matter what scale the IoT will develop to in the future, the current 4.3 billion addresses are far from meeting the demand. From a capacity perspective, we ran out of IPv4 addresses in the mid-1990s. We only use the expanded IPv4 available address for the Internet of Things that is far beyond the IPv4 address capacity through many means. So IPv6 is not necessary, but there are still many difficulties before the transition to IPv6 network.
1. The Internet lacks centralized management and is an alliance of a large number of autonomous systems managed independently. Therefore, there is no way to force or coordinate everyone to switch from IPv4 to IPv6.
2. The network needs a lot of financial resources, manpower and technology to fully support IPv6.
3. IPv6 is not backward compatible with IPv4. IPv6 was first born in the 1990s. At that time, the designers believed that operators would actively deploy IPv6, and almost no one thought that IPv6 deployment would face many obstacles.
NAT64 is a stateful network address and protocol conversion technology, which generally only supports access to IPv4 network resources through IPv6 network side user initiated connection. However, NAT64 also supports manual configuration of static mapping relationships to enable IPv4 networks to actively initiate connections to access IPv6 networks. Although most of the current devices support IPv6, there are still many old devices that only support IPv4. These devices need to be connected through IPv6 network in some way. NAT64 can realize IPv6 and IPv4 network address and protocol conversion under TCP, UDP and ICMP protocols. And because IPv6 is not compatible with IPv4, there must be necessary migration mechanisms, such as dual stack, tunnel and conversion.
1. Dual stack interface: The simplest way to maintain the coexistence of IPv4 and IPv6 is to configure two protocols for the interface. The version of the IP protocol used depends on the version of the packets received from the device or the address type returned by DNS when querying the device address. Although dual stack is the expected migration method from IPv4 to IPv6, the premise is that the migration process must be completed before IPv4 addresses are exhausted.
2. Tunnels: Tunnels also solve coexistence problems. The tunnel allows a device or site of one protocol version to pass through the network segment (including the Internet) of another protocol version, so two IPv4 devices or sites can exchange IPv4 packets through the IPv6 network, and two IPv6 devices or sites can also exchange IPv6 packets through the IPv4 network.
- Conversion: The conversion technology is to change the packet header of one protocol version to the packet header of another protocol version, thus solving the interoperability problem between IPv4 devices and IPv6 devices.
