Networking is the Future of Audio Pt. 2 - Subnets
Now that we know what an IP address is, we need to start looking at how IP addresses fit within the network as a whole. This where subnets come in. For an audio engineer, it may be easiest to begin by thinking about a subnet the same way you think about a stage box connecting inputs to a console.
A subnet is effectively the network “stage box” within which your IP address exists.
This “stage box” is defined by an IP address range. In our stage box analogy, you could consider this range to be the number of channels available on the stage box; such as a sixteen channel stage box.
One thing to note: It is important to consider a subnet to be the range that both the device address falls within and what other addresses the device can communicate with.
The way you would define a subnet is by using what is called a subnet mask. A subnet mask is represented similarly to an IP address, as it is also made up of four octets. However, is it critical to know that there are only 33 different options of subnet masks rather than the four billion different IP addresses.
The most common subnet mask is 255.255.255.0. This is also known as a /24 network (for more information on the slash notation of subnet masks, definitely check out CIDR notation). Unfortunately, this is where we have to go back to binary to understand what is happening.
255.255.255.0 in binary is 11111111.11111111.11111111.00000000
It is a bunch of 1s and then a bunch of 0s. The way a subnet mask is determined is based on how many 1s there are, starting from the left and working your way to the right. Here we see twenty-four 1s, which is how we get the /24 designation.
In a subnet mask, the 1s represents the network itself (or the stage box to the console in our analogy).
The 0s represent the host addresses available (or the inputs actually utilized on the stage box).
If we take our original IP address of 192.168.37.64 with a subnet mask of 255.255.255.0 we are left with:
Network defined as - 192.168.37.X
Host addresses defined as - 192.168.37.0 - 192.168.37.255
Because we have eight bits following the defined network, we are free to use all eight for any IP address within that range, with a total 256 possible IP addresses. In the same way, you would have up to sixteen inputs on your stage box. We only allow addresses with the first three octets matching to access each other. If you were in a festival situation with more inputs on stage and another band set up behind yours, their inputs would still exist, but they do not exist in your stage box and you cannot access them.
Now, let’s say we need a much bigger stage box for a bigger band. We can create a subnet mask that allows for more inputs. With the same thought process, we need less 1s and more 0s. Let's take away four 1s and make them four 0s, which leaves us with the subnet mask:
11111111.11111111.11110000.00000000
If we convert this to base ten, we are left with 255.255.240.0.
(For the keenly aware, 11110000 is equivalent to 240 in base ten)
With twelve 0s instead of 8, we can now have up to 4096 IP addresses within our subnet instead of the original 256. This should be plenty of space for our “expanded stage box” with many more than “sixteen inputs”.
That is a lot of IP addresses and a lot of device management to deal with. This leads us into routers and the use of a protocol called DHCP to help us manage these devices.
