The difference between static and dynamic NAT is that once the five global addresses are manually assigned when using static NAT, they will never change. The only way to change them is by using the console program to manually reassign them. When using dynamic NAT, the router will map a local IP address to a global IP address whenever a request is made. Since there are only 5 global IP addresses in the example above, there can only be 5 mappings at any one time. In other words, much like static NAT, only 5 local machines can access the Internet at any one time. However, contrary to static NAT, the router will discard the mapping between the global and local IP addresses after a certain length of time (which is quite long so rarely happens), or after the session is finished (an example of a session is when requesting a web page, the entire page has completed downloading). The most common implementation of NAT is to define a range of dynamic addresses to be used by hosts, but assign static addresses to your servers if you wish for them to be accessible from outside your network.
NAPT
NAPT is an advanced version of NAT that uses IP port numbers in the network address translation process. It is much more widely implemented on networks today due to the fact that it uses only a single global IP address (as opposed to NAT which uses a range of global addresses), thus providing greater cost savings. For Internet access for everyone on the network through a single IP address (a single user account), NAPT is the right choice.
When a packet on the local network arrives at the router and needs to be sent to the Internet, it already has a (local) source IP address and a (local) source port number that was generated when the packet was
82 | Configuration and |
Management
