NAT Overview

NAT is a powerful tool in connecting networks. One of the biggest challenges in networking is addressing, and the challenge only becomes bigger as networks scale and grow connected. NAT gives us great flexibility in our use of addresses. With NAT, we can use a variety of different or similar address schemes. Simply put, NAT gives us the ability to convert one address to another. However, it is an oversimplification to say that NAT only translates addresses; NAT is a solution for a number of problems:

• With the depletion of public addresses on the Internet, NAT has come to the rescue. For companies that do not have enough valid unique addresses to connect all their hosts to the Internet, NAT can be configured on a router bordering the Internet and convert an organization's addresses that are on the inside of the network to legitimate addresses on the public network. The translation or mapping can be one to one, many to one, or ranges, as we discuss later.

• NAT can also be used in the migration from one Internet service provider (ISP) to another, where you receive a new IP addressing scheme. Typically, when using NAT, you only need to change a few public addresses as opposed to renumbering your entire private IP network.

• NAT can also be useful for merging networks. Suppose two companies that are using the same address range merge. We can use NAT to convert between the two networks, presenting each side with an address that is unique within the range they are using.

• NAT can even be used for destination-based load balancing. NAT can take incoming server requests destined for a single address and disperse them in a round-robin fashion to multiple servers, each with its own unique addresses.

NAT is defined in RFC 1631, and Cisco has supported NAT on its routers since IOS release 11.2.