Illustrated TCP/IP by Matthew G. Naugle Wiley Computer Publishing, John Wiley & Sons, Inc. ISBN: 0471196568   Pub Date: 11/01/98

Chapter 51
Subnet Mask Template

Not sure about the previous example? Let’s break it out. To identify the subnets is a little tricky. The previous slide is shown again. As you can see, the vertical line separating the host and subnet portions of the address is the dividing line. The first bit in the subnet portion of the address is set to 1. The subnet would not be 1. In calculating the value of the subnet, the whole third field is taken into consideration. Therefore, since that bit is set, it is actually a binary 8 (the fourth bit). Therefore, the first subnet number will be a 0. Each subsequent subnet will be a multiple of 8.

In the previous example with each of those subnetwork numbers, we could possibly have 2046 hosts per subnetwork number. This is a little more realistic than not subnetting. Not subnetting gives us 65,534 hosts. We were assigned one IP address and, with subnetting, we were able to make better use of the address without having to reserve more addresses (network numbers). Also, with subnetting, only one IP address is in the Internet routing tables, even though we have 32 subnets on our network. The Internet routing tables do not care about subnets. We used one Class B network number and have 32 subnets available to us from the one Class B network. Without subnetting, we would have one network number and up to 65,534 hosts assigned to it.

How did we get 32 possibilities? Using 5 bits for the subnet mask gives us 32 possible combinations (0 to 31), or 2ⁿ5. Remember, we can move the mask anywhere in the 14 available bits. The subnet mask could have used all 8 bits in the third octet, which would give us 256 subnet numbers (all 0s and all 1s being allowed).

Subnet Mask Template

How do we write a subnet mask? It is always written in decimal and shows the number that will be used to mask the bits. For example, let’s use the IP address 130.40.132.3. Using the first 5 bits of the first host field (the third octet) yields 248 (convert the first 5 bits to binary 11111000). The byte is read as a whole 8 bits even though part of it is used for the subnet and part for host assignment. This means the subnet mask for that IP address will be 255.255.248.0 in decimal. This is the mask that we have assigned to the network address of 130.40.132.3. We will always use 255 in the network potion of the subnet mask. The 248 is used to tell the network station to use the first 5 bits (5 bits binary is 248 decimal) of the network address, not for a host ID, but for a subnet. It tells a network station which bits to use for a subnet mask. The remaining 11 bits (the remaining 3 bits of the third octet and 8 bits of the fourth octet) should be used for the host ID. This allows for 32 subnets with 2046 hosts on each subnet.

Therefore, the IP address of 130.40.132.3, with a subnet mask of 255.255.248.0, yields the network number 130.40, subnet number 128, and host ID 1027.64