A. Three main types of NAT exist. In order of complexity (from simple to complex), they are
- Static NAT--With this type of NAT, a NAT router maintains a table that associates each internal IP address with a corresponding external allocated (i.e., registered) Internet IP address. With static NAT, you must register an IP address for every machine that connects to the Internet. This approach isn't used very often because it doesn't save on registering IP addresses. However, static NAT can be useful for making devices accessible from the Internet--the external IP address will always point to the internal address stored on the NAT router.
- Dynamic NAT--With dynamic NAT, a NAT router maintains a list of registered Internet IP addresses. Every time an internal client tries to access the Internet, the router maps it to one of the registered IP addresses that isn't currently in use. As a result, you need registered IP addresses only for the number of concurrent Internet users.
- Single-Address NAT/Overloading/Masquerading/Network Address Port Translation (NAPT)--With this type of NAT, a NAT router has only one registered IP address. The NAT router maps each internal client that needs to communicate with the Internet to a different port from the registered IP address. The router writes the address request in the form x.x.x.x:y--for example, 10.0.0.1:100 would be IP address 10.0.0.1, port 100. Responses from the Internet include the originating port so that the router knows which internal IP address to map the response to.
The figure below illustrates the use of single-address NAT. The NAT router in the figure maintains a translation table that specifies the port that each internal IP address uses for external communication, as follows:
|Internal Address||External Address|
This type of NAT is the most popular form used.