Networking can be pretty complex. Subnetting is a base concept of networking that really tends to scare people away. The truth of it is that there are alot of documents that try to explain how it works, but I fear that too many of them try to throw the math at you as well as the logic and that tends to make it a bit difficult to understand. I'd like to introduce you to Dlostboy's easy subnetting. All that is required is an understanding of how binary (base2) numbers work. Once you get to the point that you know what number 01110010 translates to in our base10 system and what an IP address is, then you have enough knowledge to tackle this.

IPv4 addresses are 32 bit numbers. That means that it takes 32 1's & 0's to represent the number. They are grouped into four 8 bit numbers separated by decimal points. Subnet masks are numbers that let you know how many other similar numbers are part of the same group, or network (subnet). You should have already known this if you're wanting to subnet.

Now here's the simplicity: Networks are usually portrayed as either things like 10.5.2.0/24 (read "10 dot 5 dot 2 dot zero slash 24") and 10.5.2.104/29 commonly known as Ascend notation or as 10.5.2.0 255.255.255.0 and 10.5.2.104 255.255.255.248 which is a fully written out IP/subnet. Both are respectively equivalent and it's good for you to know how both methods work.

We'll start with the first one. 10.5.2.0/24. The number after the slash represents how many bits are in the subnet mask. We'll skip the terminology for a minute and just say that it tells you how many IP's are in the subnet. Now all subnets have to have an address at the top that is "unused" and another at the bottom. These addresses are used for things like broadcast messages (arp for instance) If the subnet is a /24, then you take 24 from 32 (how many bits are in an IP), which leaves you 8. The largest # that an 8 bit binary number can be is 255. (11111111). That tells us that there are 256 addresses (remember, we start numbering from zero) in this subnet. Take out the top and bottom one and that leaves us with a possible 254 addresses in this subnet. The IP's 10.5.2.1 through 10.5.2.254 are all usable. Simple, huh?

Let's work with the second example: 10.5.2.104/29. 32 minus 29 equals 3. The highest number that 3 bits can represent is 7 (111). That leaves us 8 IP's (starting from zero again). Throw away the top and bottom and we're left with 6 usable IP's: 10.5.2.105 through 10.5.2.110.

Let's look at the other method. 10.5.2.0(IP) 255.255.255.0(SUBNET). In this method, we take the highest number represented by 32 bits in 4 groups, 255.255.255.255 and subtract a number from it. In this case, we've subtracted 256 IP's from the last number (starting with zero). Which is how many IP's we're using. Take out the top and bottom and that tells you that you've got 254 usable addresses.

Second example, 10.5.2.104 255.255.255.248. 255 minus 248 = 7. Add one cause we're on a zero scale and that gives us 8. Eight minus 2 for the top and bottom and that leaves us 6 usable IP's.

I hope this takes some of the mystery out of these subnets. They're really not as hard as they initially appear.

  ©2000, ©2001 LostInformation