While static NAT provides a permanent mapping between an inside local address and an inside global address, dynamic NAT allows the automatic mapping of inside local addresses to inside global addresses. These inside global addresses are typically public IPv4 addresses. Dynamic NAT uses a group, or pool of public IPv4 addresses for translation.
Dynamic NAT, like static NAT, requires the configuration of the inside and outside interfaces participating in NAT. However, where static NAT creates a permanent mapping to a single address, dynamic NAT uses a pool of addresses.
Note: Translating between public and private IPv4 addresses is by far the most common use of NAT. However, NAT translations can occur between any pair of addresses.
The example topology shown in the figure has an inside network using addresses from the RFC 1918 private address space. Attached to router R1 are two LANs, 192.168.10.0/24 and 192.168.11.0/24. Router R2, the border router, is configured for dynamic NAT using a pool of public IPv4 addresses 220.127.116.11 through 18.104.22.168.
The pool of public IPv4 addresses (inside global address pool) is available to any device on the inside network on a first-come first-served basis. With dynamic NAT, a single inside address is translated to a single outside address. With this type of translation there must be enough addresses in the pool to accommodate all the inside devices needing access to the outside network at the same time. If all of the addresses in the pool have been used, a device must wait for an available address before it can access the outside network.