Friday, December 8, 2006

Subnetting an Intranet

When intranets are over a certain size, or are spread over several geographical locations, it becomes difficult to manage them as a single network. To solve the problem, the single intranet can be subdivided into several subnets, subsections of an intranet that make them easier to manage. To the outside world, the intranet still looks as if its a single network.

1. If youre building an intranet and want it to be connected to the Internet, youll need a unique IP address for your intranet network, which the InterNIC Registration Services will handle. There are three classes of intranet you can have: Class A, Class B, or Class C. Generally, a Class A rating is best for the largest networks, while a Class C is best for the smallest. A Class A network can be composed of 127 networks, and a total of 16,777,214 nodes on the network. A Class B network can be composed of 16,383 networks, and a total of 65,534 nodes. A Class C network can be composed of 2,097,151 networks, and 254 nodes.

2. When an intranet is assigned an address, it is assigned the first two IP numbers of the Internet numeric address called the netid field and the remaining two numbers called the hostid field are left blank, so that the intranet itself can assign them, such as 147.106.0.0. The hostid field consists of a number for a subnet and a host number.

3. When an intranet is connected to the Internet, a router handles the job of sending packets into the intranet from the Internet. In our example, all incoming mail and data comes to a router for a network with the netid of 147.106.0.0.

4. When intranets grow-for example, if there is a department located in another building, city, or country-there needs to be some way to manage network traffic. It may be impractical and physically impossible to route all the data necessary among many different computers spread across a building or the world. A second network-called a subnetwork or subnet-needs to be created.

5. In order to have a router handle all incoming traffic for a subnetted intranet, the first byte of the hostid field is used. The bits that are used to distinguish among subnets are called subnet numbers. In our example, there are two subnets on the intranet. To the outside world, there appears to be only one network.

6. Each computer on each subnet gets its own IP address, as in a normal intranet. The combination of the netid field, the subnet number, and then finally a host number, forms the IP address.

7. The router must be informed that the hostid field in subnets must be treated differently than non-subnetted hostid fields, otherwise it wont be able to properly route data. In order to do this, a subnet mask is used. A subnet mask is a 32-bit number such as 255.255.0.0 that is used in concert with the numbers in the hostid field. When a calculation is performed using the subnet mask and the IP address, the router knows where to route the mail. The subnet mask is put in peoples network configuration files.