Wow … this might seem a daunting term for you non network techies out there, but let me explain what subnetting is and you will see that it’s not so complicated after all.
What is the goal of subnetting?
The goal of subnetting is to make efficient use of an IP address space. At a certain moment in time the company work for opened its first branch in … let’s say the US.
The network technicians back then had a choice out of three private IP address ranges:
- 10.0.0.0/8
- 172.16.0.0/20
- 192.168.0.0/16
(Let me tell you that you still have numerous companies that choose to use public IP addressing internally for which there is no excuse. It is simply NOT done.)
But so … back to the network technicians. Let’s that they chose the 10.0.0.0/8 range. Could they use this network configure all the computers, servers, printers and whatever devices they would have on that one subnet? YES, they could. Would it be smart? NOOO!!!
There are a number of reason why this is not a good approach ranging from bad planning for the future, security, performance.
So instead, what they started doing is cutting this range in several segments, some big, some smaller and some really small.
This art of segmenting the network in different smaller networks, which we can sub-nets, it called subnetting.
How does it work?
When you are subnetting, the first thing to do is plan for the future and build a logical structure into your subnetting scheme.
When you create subnets you have two part of a subnet:
- the network bits
- the host bits
The network bits in a subnet are fixed and are reflected by 1s in the binary representation. The host bits are the 0s in the subnet in the binary representation.
In the 10.0.0.0/8 subnet the binary representation of /8 would be 11111111.00000000.00000000.00000000 (255.0.0.0) which means that every address in this network will start with 10. and the rest are bits that are available by the hosts (with the exception of the first and the last address in the range).
That would still leave 255 x 255 x 255 – 2 addresses available for assignment which is huge.
Now let’s say that we plan for the future and we want to separate ranges by country, we could designate the second byte to the country
Let’s say we have 4 countries that we are active in by now:
- USA
- France
- Belgium
- Brasil
So we could say that we do our subnetting as follows:
- US 10.0.0.0/16: you will have 255 x 255 hosts available
- France 10.1.0/16
- Belgium 10.2.0.0/16
- Brasil 10.3.0.0/16
This would work although I doubt we would ever have 255 countries in our company.
Let’s make them a little bigger:
- US: 10.0.0.0/14
- France: 10.4.0.0/14
- Belgium: 10.8.0.0/14
- Brasil: 10.12.0.0/14
So each country could have 2 to the power of 18 devices. You subtract 2 addresses for the network and the broadcast address and you and up with 262144 devices per country.
