Andrew Levicki: Subnetting (and certification) tips

A couple of weeks ago Andrew Levicki, who is writing for us regularly on IT topics and certification, introduced himself in his first post. Today Andrew shares some advice on Exam 70-291 (Implementing, Managing, and Maintaining a Microsoft Windows Server 2003 Network Infrastructure) and Exam 70-293 (Planning and Maintaining a Microsoft Windows Server 2003 Network Infrastructure) and a technique for how to do subnetting during an exam.

> > >

Hi there! How’ve you been? Since we last spoke, I’ve taken and passed two more Microsoft exams, the 70-291, and the 70-293. The 293 was my 9th Microsoft exam pass in just over 9 weeks. It’s been very hectic, all things considered, but I said I’d keep you updated every couple of weeks, so here I am.

Firstly, about the exams, if you’ve done these ones, then maybe you’ll agree with me and if you’re studying for them, they’re kind of similar to each other, inasmuch as the topics overlap, it’s just that the 293 focuses on planning whereas the 291 is more oriented towards implementation. I passed them both without problems, but I wish now that I had studied for the 293 first, then the 291. The Microsoft material for the 293 introduces concepts at a higher level that helps bring you into the nitty-gritty of the 291 curriculum. So that’s my advice, anyway: do the 293 first, then the 291. Your mileage may vary!

I was going to tell you all about how we’ve been looking at implementing a Wide Area Network solution at my company this week, but it’s actually turned into a bit of a political nightmare and also I believe there maybe too many references to Cisco Systems for my hosts’ liking, so I will shelve that. But what I also was hoping to share with you was a brilliant way to subnet that I learned on my course a couple of months back.

Now I know you all know how to subnet given a calculator, especially one of those neat online subnet calculators, and I know you all know what subnetting is and why we need it, but if you don’t, see here. But what I wanted to share with you was a technique that comes in handy when you’re in an exam situation without recourse to these handy utilities. Now if you can already subnet in your head, congratulations, you won’t need this method. But for those of you who aren’t too concerned about having a close personal relationship with binary, read on…

All you need to do this is paper and a pen / pencil, which are usually provided in Microsoft exams. Very simply, all you have to do is write the following down:

128  64    32    16    8      4      2      1

You’ll see that it is just 1, doubled again and again, but backwards in sequence up to 128.

Then add a second line, which adds them up sequentially:

128  64    32    16    8      4      2      1
128  192  224  240  248  252  254  255

This will represent subnet mask numbers.

Then just write down /1 to /32, creating a new line after every 8, so you end up with:

128  64    32    16    8      4      2      1
128  192  224  240  248  252  254  255
/1     /2    /3     /4    /5    /6     /7     /8
/9     /10  /11   /12  /13  /14   /15   /16
/17   /18  /19   /20  /21  /22   /23   /24
/25   /26  /27   /28  /29  /30   /31   /32

This is all the information you need to subnet in your exam. The /numbers show you what your “slash” notation will look like if you use the associated number from the 2nd row in the subnet mask, depending on which octet you place that number in, which corresponds to the four rows of /numbers, one for each octet.

So, for a nice and simple example, 192.168.1.0, subnet mask 255.255.255.0 is the same as 192.168.1.0/24, because “255” is used in the 3rd octet, see?

If we subnetted that to 192.168.1.0/25, we can see that the subnet mask will be 255.255.255.128, because /25 corresponds to 128 in the 4th octet. See? And there’s more. The number of columns over you go can indicate how many networks and hosts available, for networks it is simply 2 to the power of columns. In this instance, it is 2 to the power of 1, which is 2. And for hosts it is 2 to the power of the remaining columns, which in this case is 2 to the power of 7, which is 128. But the first address is the network address and the last address is the broadcast address, so there are 126 usable addresses (i.e., addresses that can be assigned to devices).

So what are our subnets in this example? The top row tells us by how many the subnets increment; in this instance it is 128. So we have:

Network Range Broadcast
192.168.1.0       192.168.1.1 – 192.168.1.126       192.168.1.127
192.168.1.128   192.168.1.129 – 192.168.1.254   192.168.1.255

Let’s try that with a slightly harder one:

For host 182.34.31.58 with Subnet Mask 255.255.252.0, identify the correct network ID and broadcast address.

Well, straight away we can see that it is a /22 because the 252 is in the 3rd octet.
There are 2^6 networks, which is 64. There are 2^10-2 hosts (10 because there are 2 columns left in the 3rd octet and the whole 4th octet), which is 1022.

The top row tells us that our subnets increment by 4, this time in the 3rd octet.
So we have 182.34.0.0, 182.34.4.0, 182.34.8.0, 182.34.12.0, 182.34.16.0, 182.34.20.0, 182.34.24.0, 182.34.28.0 – aha! That’s the subnet we want! Because the next one is 182.34.32.0, which is more than our IP address.
So our subnet is 182.34.28.0 and the broadcast address is one less than the next network, which is 182.34.31.255. Although we weren’t asked, the valid host range is 182.34.28.0 – 182.34.31.254.

Let’s try another:

Given the IP 128.167.0.0, which subnet mask will give you 155 usable subnets with at least 192 hosts per subnet?

A. 255.255.254.0
B. 255.255.0.0
C. 255.254.0.0
D. 255.255.255.0

These ones are easier with this method. Look at the subnet masks. A gives us a /23, offering us 2^7 subnets, (128). B gives us a /16, offering just the one network, this is the standard Class B mask. C is not valid, because it is a class B address to begin with and 255.254.0.0 is not valid for a Class B address. D gives us a /24, with 2^8 subnets (256), each with 2^8-2 usable hosts (254). So the answer is D.

Does that method work for you? Why not try it on a few examples that you verify on an online subnet calculator? When I go into an exam where there’s a chance that I will be asked subnetting questions, I draw that table out at the start of the exam and it’s never let me down yet! Have you got an even better method? Do get in touch and let us know.

So I’m now studying for the 70-294 exam, which is all about planning, implementing and maintaining Windows Server 2003 Active Directory. I’m hoping that this will be my 10th Microsoft exam pass. Then I will have only one more exam to do to gain the coveted MCSE. This is the stuff dreams are made of. I really am starting to believe my dad, who told me I was a genius. But he did also tell me that I had a face only a mother could love, so it’s swings and roundabouts.