Hello, Cyber Arians. I'm Kelly Hander Han and welcome the session Wednesday. This week we're gonna talk about subnet netting and try to make it quick and easy, or at least this quick and easy, its sudden and B s. So what we're gonna look at first of all, they're pre requisite knowledge to this is that you understand just a little bit of basics
about by Neary two decimal conversion.
And the reason that's important is when you look at a typical I p address one of the things
that we used to define network numbers with an I p. Who use a subject mask.
Right? And if you're familiar with the basics of I P, we have three
the fault sub net masks to 55 0
and then to 55 to 55 to 55.0,
and these air the class ful sub net masks
for Class A, Class B and Class C.
Now, if you were taught from an introductory standpoint about I p dressing in sub net masks in a really simple way to introduce this topic is to say to 55 means network and zero means host.
So when you match this up against an I P address and you look at 2 55 in the first hot cat, that tells me the first top cat of the I P address is network portion.
So you know, for a long time it's really easy to say to 55 equals Network
But that's only because we like to look at things in destiny, you and I. That's the numbering system that we're taught
The problem with looking at this in decibel is it's either all or nothing.
E Chop Head is either network or its host. There's no variation.
So in a class A, the entire first top cat represents networks. And then we have the 2nd 3rd and fourth octet for hosts
go down to class seat. I only have the final octet for hose.
So basically what we look at us, we've got a very inherently wasteful I p addressing scape.
Now, remember, we're the technology people that didn't know the year 2000 was coming. I mean, if you remember back in 1998 a bunch of us went independent going Oh, my God. It looks like 2000 is actually gonna get here. So sometimes we're not as forward thinking as we should be. That's okay. We've got some work around here,
with this I p addressing skate,
what you can tell is we don't have nearly as much support for hosts here as we do up here.
Post, Post, Post Post
on the Class C network,
it can support 254 hosts.
The Class B can support 65,534 hosts
and then a class A network can support 16.7 million. Some change. Hey, now
that seems okay, but we've got support for just 17 million I p addresses. Think about all the users on an Internet at any given time and think about for every user they have four or five different IP addresses, and that's just gonna continue to grow and grow and grow. So you've probably heard at some point in time we're running out of I p addresses.
We're really not because we
found some ways around that one of the ways this sub netting, but also, if you're familiar with network address translation, that also helps so we're not really is concerned with running out of I p addresses. And even if we were, I p Version six is coming down the line.
Supposedly, it's been coming down the line since I've been in I t. Which has been 15 plus years so
But any break that will absolve us of any sort of issues or problems with running out of I P addresses. But what we want to focus on now it's subject. Okay, so let's say that I do have a Class A network and I am gonna use the 10 Network. And there's several different beneficial reasons to using the 10 network,
that it's a private internal i p address. So ultimately, it's set aside,
could not be routed across the public Internet. This is a good choice for an internal address. The other reason I'm gonna do it is it has support for lots of hosts, right? The 10 network by default hauled into the sub net
Now how do I know that? Well, let me tell you, I don't know it for sure.
So unless I'm told differently, I'll go with defaults, and with the fault I p addressing If the first dot pet falls in the value of 1 to 1 26
then you can default to a class a address.
Now, if you're given an i P address in a subnet mask that contradict this, go with what you're given.
But if you don't have a subject mask, you can make that assumption.
2 to 20 threes. Class C. You'll notice that 1 27 is missing. That's reserved for testing purposes. It's the loop back.
All right, so I've got a Class A
I P address if we're going back to the standard class ful addressing system.
the thing with this is when I have the 10 Network, I only have a single network.
What if I want a segment traffic in my network? What if I wanna isolate traffic from one section to another? Maybe the sales group has an application that broadcasts a lot. I want to keep those broadcast off my production network. Maybe I have avoided network. It's very bandwith intensive, and I want to keep that
you know from interfering.
So what I need to do is I need this segment, my network. I need to get more networks out of this I p address.
Well, the idea is, I would happily trade some host addresses for some network addresses. I can guarantee you I don't have 16.7 million hosts.
I want more than one network. I'm happy to make a trade.
So the way we do this is we have to look at this in binary.
That's the reason to 55. Significant, anyway, is that it's all bit set the binary ones. So we can go back to this old idea of 2 55 equals networks,
and we can say binary ones
equal networks. Okay, So the first bit being to 55 all ones. Now the rest.
That's what this subject mask really looks like when we look at in binary.
So if you'll remember zero for hosts ones or for networks, I want to trade someone's. I'm sorry when it trades and see rows, and for once All right. So the way I do that, I just start stealing bits,
all right? And I take the first bid from the second test.
But what has that done for? Well, it's done a couple of things. It's changed my standard default subnet mask into something different.
What would that be now? Well, that's where your knowledge of binary comes in.
So if we look over here
now, honestly, it's not essential, you know, binary. But if you do, these numbers probably make a lot more sense. If not, take my word for this is what the numbers are in binary.
So when I steal one bid and I changed it from a 0 to 1, I no longer have a subject mask of 2550 I have a subnet mask of 255
but who cares? Well, here's why we care
when we start stealing. Bids were actually adding networks
so or support for networks. So when I steal one bit, what I'm actually creating and here's the Formula Two to the power of X
equals number of networks created.
All right. So if I steal one bit
what we call the significant
which is the octet from which were stealing
and I create for myself to to the first power.
So I've created two networks
just by changing my subject mask to 1 28 Aiken ounce pour to network. That's not enough, though. I'm 1 15 networks. So what do I do? I just keep stealing. Steal another bit.
Now my significant octet changes to 1 92
So when I steal to bits, my significant octet, it's 1 92 How many networks have I created to Second Power equals four?
Not enough. So you know what I do? Keep stealing
another bit. Steel. Three bits
to 24 is my significant on pet. I create eight. You're probably seeing a trend here, right? So if I create four
used the 2 47 that mask I'm gonna give myself 16 networks 5 to 48.
And we can keep going on and on and on and on. So the bottom line is, what I do is I take a look at this in binary,
and then I see how many bits I need the steel
for how many networks I create. So this a really handy chart between that side of the border this side of the board. You can see when I steal bits. What does my significant not cut become? And then how many networks do I create for myself
now? If we were to do this with a Class B subject mask, the same idea is the same.
Okay, so let's say we take a class B.
I'll write this up here again.
So if I take a Class B address, let's just use 172.16. That's another private I p address. So that's why we're using this. The only thing that's different
is instead of starting with the class, a sudden it mass. We start with a classmate to 55255
are already taken care of. It's the third octet. That becomes significant now.
So if I want to create a sub net, that'll support 30
networks. Well, I look at my chart. I would have to use to 48 as my significant other pet. So to support 32 networks, all in all
on a Class B network, it would be to 55255.248
That would be the subnet mask I used.
So what that means is I would have to steal five bits.
Okay. And this would be my subnet mask and decimal. This would be in binary
one. So this idea is called cider Classless Inter domain routing. And the reason it's called Classless is. Yet even though we start by looking at the default classes of I P and the default summit masks
we move beyond, there were no longer limited to just class A just class B, just classy.
So we get a lot more flexibility. Another thing with cider notation. This is a very appropriate, totally appropriate is the sub net mask. But what were generally doing today indicate submit masks
eight bit set one right.
Here's another eight says total. 16 17 18 1920 21.
Often we would represent that is a slash 21 subnet mask slashed. 21 means the same thing is 255252480
So what you're saying is, they can't you see, For instance, if you were to see slash eight,
the 1st 8 bits would be set toe one that's a 255 Ooo
Class C by default would be 24 bits set toward
so hopefully this has been helpful to you to help you understand what some of these custom sub net masks are. You can build on that you can go to some sites you can always ask, can't Google for some practice questions on subheadings. There many tutorials out there. But I'm hoping this will give you kind of a start
on what you want to be doing and just understanding. Why don't we use sometimes
Ah, standard subnet mask. Why do you sometimes see things that are a little bit different? Well, what's happening is we're getting beyond the traditional classical addressing system and we are embracing cider. Okay, good luck.