Subdividing a Network into Subnets

Class A networks contain more than 16,000,000 IP addresses, and Class B networks contain more than 65,000 addresses. Out of context, those might seem like large numbers. When you consider the fact that all hosts in the same IP network cannot be separated from each other by a router, the numbers end up looking huge. These days, LANs with more than a few hundred devices on them are unusual, and single LANs with more than 1000 devices are extremely rare. If you use a Class A or even a Class B network for all the devices on a LAN, many IP addresses will go unused.

Subnetting provides one solution by reducing a lot of the waste. Subnetting refers to a process by which the network engineer in charge of a particular TCP/IP network number, in effect, changes the rules. Next, the chapter shows you an example in which lots of IP addresses are wasted, followed by the same general example, with subnetting being used to reduce waste.

The Problem: Wasting IP Host Addresses

As an example of the possibility of excessive IP address waste, examine Figure 10-8, which uses three Class B networks.

Figure 10-8. Wasting IP Addresses: Room for a Ton, Only Need a Few

The design in Figure 10-8 requires three networksin this case, three different Class B networks. Each Class B network has 2¹⁶2 host addresses (65,534) in itfar more than you will ever need for each LAN. Only a few IP addresses have been used so farone for each computer, plus one each for the router interfaces. (Remember: Each network interface has an IP address. So, in network 150.2.0.0, two addresses are used by the two computers, another is used by R1's right-side Ethernet interface, and a fourth is used by R2's left-side Ethernet interface.)

IP routing processes still work well, based on the fact that all hosts in the same network are in the same place. For instance, all hosts with addresses that start with 150.1 need to be on the Ethernet on the left side of Figure 10-8. However, more than 65,000 IP addresses exist for each of the three networks that are sitting around, unused.

The Solution: Subnetting Saves IP Host Addresses

IP subnetting relaxes the rules a little bit. Without subnetting, the following rules apply:

1. Devices in the same Class A, B, or C network cannot be separated from each other by a router.

2. Devices in different Class A, B, or C networks must be separated from each other by a router.

These two rules might have been somewhat intuitive to you, based on the earlier examples. However, to make sure routing works well, in Figure 10-8, all addresses that start with 150.1 need to be on the left-side Ethernetthat's Rule 1. Hannah and Dean, in networks 150.1.0.0 and 150.2.0.0, must be separated from each other by a routerin this case, R1. That's essentially Rule 2.

Subnetting allows the network engineer to subdivide a Class A, B, or C network into smaller piecescalled subnetsand treat each subnet by the old rules used for networks. So, to subnet a network and reduce IP address waste, the rules are changed a little:

1. Devices in the same subnet cannot be separated from each other by a router.

2. Devices in different subnets must be separated from each other by a router.

This concept is better explained with an example. In Class B network 150.1.0.0, all hosts whose addresses begin with 150.1 are in the same network. An IP network is just a group of hosts with some part of their IP addresses holding the same value. Subnetting allows the network engineer to configure the network devices such that they think that the first 3 octets of the addresses identify the network. In this one example of subnetting, the network engineer can create the following subnets:

• All addresses that begin with 150.1.1

• All addresses that begin with 150.1.2

• All addresses that begin with 150.1.3

• And so on…

Although all the addresses are still in Class B network 150.1.0.0, the networking devices will not think of the network as one big group, but instead, as a lot of smaller groups, called subnets. A subnet is just a subdivision of a larger Class A, B, or C network. The term subnetting refers to the process whereby the engineer decides to create subnets. Figure 10-9 shows this same network diagram, now with subnetting implemented.

Figure 10-9. Using One Network with Multiple Subnets

As in Figure 10-8, the design in Figure 10-9 requires three groups of IP addresses. Unlike Figure 10-8, this figure uses three subnets, each of which is a subnet of a single Class B network (network 150.1.0.0). Notice from Figure 10-9 that each subnet is much smaller than the original Class B network, but there are lots of subnets.

IP network numbers represent networks, and similarly, IP subnet numbers represent each subnet. The subnet number has the same value in the first part of the number as all the host addresses, and 0s in the last part. For instance, 150.1.1.0 is one of the subnet numbers in Figure 10-9. All addresses in the subnet begin with 150.1.1, so the subnet number includes those numbers as well. Because the fourth octet can be any valid number, the subnet number is simply 0 in the last octet.

Using subnets in the Figure 10-9 network saves IP addresses. This same internetwork of three Ethernet LANs only uses a part of Class B network 150.1.0.0, as opposed to the internetwork in Figure 10-8, which fully uses three Class B networks (150.1.0.0, 150.2.0.0, and 150.3.0.0). Also, if you add another LAN, instead of needing a brand new Class A, B, or C network, you have lots of unused subnets such as 150.1.4.0, 150.1.5.0, and so on.

This example shows just one way to use subnetting. You can subnet in many ways, including subnetting Class A, Class B, and even Class C networks. For now, you understand the basic concepts; just be aware that IP subnetting can get a fair amount more complicated than what's covered here.