Subnetting is the process of subdividing a given network address to smaller sub networks. In order to create sub networks, we borrow bits from the host portion of a given network address. The figure below lists the classful IPv4 address and their default subnet masks. A classful address uses its default mask as defined by the value of the first octet.
Reynolds & Postel,‘ASSIGNED NUMBERS’, RFC 870, pp.2
For example, if we look at the network 188.8.131.52 /8, as per the table below this is a classful network address belonging to class A. It has a default mask of /8 (255.0.0.0). That means first 8 bits are reserved for the network portion and the remaining 24 bits are reserved for the host portion.
However, if we borrow bits from the host portion of default class A network address 184.108.40.206 /8 and create sub networks, then those resulting subnetworks become classless network addresses. Therefore, classless means away with the default classful subnet mask and on with a custom subnet mask.
Let’s consider the classful network 192.168.10.0 /24 to demonstrate the process of subnetting. Also, let’s assume we have to assign these subnets to the below mentioned departments. Each department must be in a network of their own.
- Human Resource
Figure Local area network
Step 1 – Start with what you are given
Firstly, let’s examine what we have got.
- 192.168.10.0 /24 is a class C network
- It’s a classful network address.
- The default subnet mask is /24 (255.255.255.0)
- The number of host IDs we can assign is 254 (2^8-2)
|Binary Representation of 192.168.10.0||1100000.10101000.00001010.00000000|
|Binary Representation of Subnet Mask 255.255.255.0||11111111.11111111.11111111.0000000|
|Network & Host Bits||nnnnnnnn.nnnnnnnn.nnnnnnn.hhhhhhhh|
Note: we cannot change 192.168.10.0. This is our address space. Only thing we can do is borrow host bits (h) which is going to result in reducing the number of hosts and extending the default subnet mask.
Step 2 – Determine the number of subnets that are needed
- We need four subnets to assign to the four departments.
Step 3 – Determine the number of network bits needed
Now we need to decide in order to create 4 sub networks how many host bits we must borrow. To calculate the number of subnets that can be created from the bits borrowed, use the formula below.
Number of networks = 2^n
n= bits borrowed
Note: The last two bits cannot be borrowed from the last octet because there would be no host addresses available. Therefore, the longest prefix length possible when subnetting is /30 or 255.255.255.252.
The figure below depicts how many subnets we can create based on the number of host bits borrowed.
Figure Subnetting a /24 network
From the above figure, we can see all we need is to borrow 2 bits from the host portion and that will produce 4 subnets out of 192.168.10.0 /24.
Step 4 – Calculate the new mask
Figure below shows the original mask.
Now, let’s examine the mask that results from borrowing 2 bits from the host portion.
We can see the result is prefix length /26 or subnet mask 255.255.255.192. This will be the subnet mask for all of the resulting 4 subnets.
Figure Subnet mask 255.255.255.192
What are the resulting subnetworks?
The figure below shows the possible bit combinations for the last octet which determines the resulting four subnetworks based on the new subnet mask.
|Network #||Binary for Last Octet||Network Address||Subnet Mask||Prefix Length|
|Network 1||192.168.10.00 00 0000||192.168.10.0||255.255.255.192||/26|
|Network 2||192.168.10.01 00 0000||192.168.10.64||255.255.255.192||/26|
|Network 3||192.168.10.10 00 0000||192.168.10.128||255.255.255.192||/26|
|Network 4||192.168.10.11 00 0000||192.168.10.192||255.255.255.192||/26|
The resulting subnetworks
Step 5 – Confirm the number of host bits and calculate the number of hosts
To calculate the number of hosts that can be supported, use the following formula
Number of theoretical hosts per subnetwork = 2^n
Number of valid hosts = 2^n – 2
n= number of bits remaining in the host portion
Note: For any host ID,
- All bits within the host portion CANNOT be set to 1 because this host ID is reserved as a broadcast address.
- Also, all bits in the host portion CANNOT be set to 0 because this signifies the IP network ID.
When calculating the number of usable hosts, we apply the formula 2^n -2. The n is the number of available bits in the host portion and “-2” is to deduct two addresses we cannot use due to above two reasons.
Therefore, number of hosts = 2 ^ 6 = 64 and the number of valid hosts = 2^6 – 2 = 62.
The figure below shows the host addresses in network 192.168.10.0 /26.
- IPv4 Network address is 192.168.10.0 and contains all 0 bits in the host portion of the address.
- First IPv4 host address is 192.168.10.1 and contains all 0 bits plus a rightmost 1 bit in the host portion of the address.
- Last IPv4 host address is 192.168.10.62 and contains all 1 bits plus a rightmost 0 bit in the host portion of the address.
- IPv4 Broadcast address is 192.168.10.63 and contains all 1 bits in the host portion of the address.
Figure Address ranges for all four subnets
Now, we can assign the subnets to each department.
Figure Local area network