The network ID assigned to our network applies to the entire network. This includes all subnets and all hosts in all subnets. Each subnet, however, needs to be identified with a unique subnet identifier or subnet ID, so it can be differentiated from the other subnets in the network. This is of course the purpose of the subnet ID bits that we took from the host ID bits in subnetting. After we have identified each subnet we need to determine the address of each subnet, so we can use this in assigning hosts specific IP addresses.

This is another step in subnetting that is not really hard to understand or do. The key to understanding how to determine subnet IDs and subnet addresses is to always work in binary form, and then convert to decimal later. We will also look at a “shortcut” for determining addresses in decimal directly, which is faster but less conceptually simple.

Let's go directly to our examples to see how subnet IDs and addresses are determined. We number the subnets starting with 0, and then going to 1, 2, 3 and so on, up to the highest subnet ID that we need.

We determine the subnet IDs and addresses as follows

1. Subnet ID: This is just the subnet number, and can be expressed in either binary or decimal form.
   
   
2. Subnet Address: This is the address formed by taking the address of the network as a whole, and substituting the (binary) subnet ID in for the subnet ID bits. We need to do this in binary, but only for the octets where there are subnet ID bits; the ones where there are only network ID bits or only host ID bits are left alone.

Seem complicated? Let's go back to our examples and we’ll see that it’s really not.