Original subtitles

>> Well it's time to culminate the subnetting topic.

As we look at the final method, variable length subnet masking.

And I know some of you might be looking, going, "Well, there was four examples

with the other one, what's up with the one example?"

Well trust me, when you see it, you're going to be like, "Okay, I only want to do one

of those," 'cause it takes some time.

And really, all VLSM is, is using the skills that you've learned and all of a sudden,

they think we've done up till now and just kind squeezing them altogether.

There's not-- I won't say there's not really any new subnetting topic that we talk about,

just kind of applications of what we've already seen.

And here's the concept.

Let me explain the concept and then we'll get into the example.

The concept is simple.

It's changing your subnet mask whenever and wherever you want.

So for example, we've been doing so far in all of our subnetting what I would consider kind

of like classful subnetting to where we find one subnet mask to rule them all.

We say, "Okay, I want 50 networks" and we break it up

and there's our one subnet mask that rules them all.

Or "I want 20 host per network," there's our one subnet mask, rules them all.

We just find one subnet mask.

But I'm looking at the example, as you've probably seen situations

where will this office has 50 users and this office has 10 users and our WAN link has two,

I mean there's two devices right there.

So finding one subnet mask to rule them all might not be the best way to go

because we'd have to figure it out base on the largest subnet, right?

But then-- and for finding it base on that, then using that subnet mask here use--

waste a whole bunch of IP addresses and same thing here.

So VLSM is really just saying well, I want to find one subnet mask for that network,

one subnet mask for that network and then pushing it all together.

So now let's move in to our example of VLSM.

Take this network environment.

We've got three networks, oh, we'll say three user networks

of 20 users, 20 users and 60 users.

And then we've got three networks of WAN links that we can't forget

about that would have like two devices each, right?

So here's our scenario and this is how you'll be able to tell if we're getting a VLSM question.

It says subnet 192.168.1.0/24, so class C, to address this work, here's the key,

use the most efficient addressing possible.

That immediately tells you, "Okay, we want those subnet masks to be efficient.

We don't want to waste IP addresses if we can."

So the key behind subnetting or sub-- VLSM is to do the same things that we've been doing.

But if we want to be efficient, start with the largest subnet first.

Okay, largest subnet, what is it on the screen?

Identify this guy right here, 60 users.

So it's using the same methods that we've seen.

So step 1 and I'll go a little faster 'cause we've seen them in the previous nuggets.

Step 1 is convert that number to binary or more specifically, find out how many bits

that number needs to be represented in binary.

So I'll throw our binary chart down here, 128, 64, 32, 16, 8, 4, 2, and 1,

so 60 in binary is going to be, let's see, 0, 0, 1.

So meaning I can stop there because I know that 60 is going to take 6 bits.

Remember, I can't get the number 60 with any less than 6 bits.

So step 2, we say, "Okay, reserve bits in the mask and find our increment."

So I'm going to say our mask originally was a /24, right, which is 1, 1, 1, 1, 1 or dot, dot,

dot, dot, dot, a whole bunch of ones, 24 ones dot and then 1, 2, 3, 4, 5, 6, 7, 8, the zeros,

I want to write out 'cause that's my playground.

This represents the first three octets of 24 bits that are there.

Okay. So reserve bits in the mask and find their increment.

Now-- so I'll ask the question, are we creating more networks or is there focus

on the networks or are we saving the host?

Saving the host, absolutely, because we're focused on 60 users, not 60 networks.

I need 60 users, 60 IP addresses on this network so I'm going to go right to left, 1, 2, 3, 4, 5,

6, and the rest can go, whoa, 7, 8.

Did I-- 1, 2, 3, 4, 5, 6, 7, no, I must have just wrote skinnier or something.

So we've got-- this is our new subnet mask for the network.

So if we were writing bit notation be /26

or decimal 255.255.255.192 would be what this is converted back to a decimal number.

So that's what our subnet mask is for that network.

That's my emphasis, our subnet mask for that network.

So now we find our network ranges or wait, our increment.

Lowest network bit converted back to a decimal number, 64, right?

And that makes me feel good because I was after 60 users

and I got the number 64 as my increment.

So step 3, we start off with 192.168.1.0 and then we just start adding 64, 192.168.1.64.

Now stop the train right there.

Let me ask a question.

How many networks of 60 do we need?

Look at it.

Just one, right?

That's it.

That's the only network of 60 that I have, so I can stop right there

because I got my one network of 60.

It's this one, 192.168.1.0 through 63 'cause the next network starts at 64, so I'm done.

Now, what I've done is created a unique subnet mask for that network.

Let's right it up there.

So I'm going to say, let's go blue, ah, let's go red.

I didn't get the right pen.

192.168.1.0 through 63/26.

Now, I write it this way because I like to see it, but know that that's not a standard.

You don't see that, you know, that they'd be like, "Hey,

that's not a standard way of writing it."

I like doing that because I can see the range and the subnet mask all at the same time.

So that tells me what that network is.

Okay. Great.

So take this, put into a little box, and do it again.

What's the next biggest network that we have?

20 users, right?

So let's just jump up top.

I like keeping them on there 'cause I want to show you how this works.

So 20 users, if I do that, how many bits to get the number 20?

No, no, no, yes.

Okay, so line moves over.

It takes 5 bits to get the number 20, okay?

Step 2, reserve bits in the mask and find my increments.

Same subnet mask.

Nothing different here, it's as if we had never done subnetting yet.

So /24 is a whole bunch of 24 ones dot 1, 2, 3, 4, 5, 6, 7, 8 zeros.

So now, I'm going to say, "Okay, am I creating more networks or saving the host?"

And the answer is we all scream, saving the host.

We need to save 5 bits of host bits, otherwise, I can't get the number 20.

1, 2, 3, 4, 5, 1, 2, 3 leftover, that's our new subnet mask, /27 or 255.255.255.224.

So those are two different ways of writing the subnet mask.

Our increments, let's grab green here, our increment is going to be 32.

Once again, makes me feel good because I'm like okay, 32 is bigger than 20.

So step 3, okay, here is where it gets a little weird, find our network ranges.

Okay, so 192.168.1.0, .1.32, .1.

Okay, wait a sec.

We're getting our increments and I look, I need two networks.

One and two of 20 users but do you see any problem here?

The first two overlap with this.

If I tail on the end ranges, this goes through 63.

The first two overlap with this.

This is the only thing different with VLSM.

I look at those first two and I get--

I go, "I'm sorry, I can't use those because they're overlapping.

They're already uses so I guess a better rule to say, a better way of thinking about it is rather

than just, you know, writing these and doing that,

you just pick up right where the last one left off."

So I didn't have to write this.

Now I did because I want you to see it

but I could have just started from 64 and felt safe doing it.

That's the beauty of the powers of two.

If I count by 128s, 64 is going to hit every value, right?

If I say, 0128, 250-- I guess 256 would be the number, then if I count by 64,

I'm going to have 064, 128, I'm going to get there, and then 192 and then 256.

I'm always going to hit the same value.

And then if I count by 32s, I'm going to get 032, 64, 96, 128,

you're always going to align these things up.

You're always going to-- they're all powers of each other.

They're all powers of two, so you're always going to--

I mean, so that-- I'm showing you that.

You're like, "What's the point?"

I'm showing you that so that when you do this and you find your one range and you say, "Okay,

that's where the next one start."

As long as you begin with the biggest subnet, you can always feel safe starting there,

you know, just saying you know what, I don't even have to right that.

I'm just going to go 192.168.1.64 and then I start adding 32 to that,

so 1.-- well that'll be 96, 1.128.

And I would say, "Okay, well I can stop there because I got my two ranges,

64 through 97 is my first one, 96 through 127 is my second one."

I've seen how this works so now, I've done my second subnetting problem.

Let me just clear off a little space in here so I can look nice and clean.

So I've done my second subnetting problem and I can come here and say, "Okay,

well this one, let's go back to red."

This one is 192.168.1.64 through 97/27.

That's this network.

And this network over is 192.168.1.9-- what do we have?

90-- wait, what am I doing?

What trickery, witchery is this?

And-- I can't count, 97 is-- that's not good.

[laughs] I heard some of you are like, "What's Jeremy doing now?"

No, this goes through 95, right?

95. My goodness, sorry.

Hang on. Hang on.

Little dissection here of some bad numbers.

So 97, so this is actually 95, right?

So this one goes from 96 through 1-- I'm making sure, 127/27.

Does that feel better?

I know-- okay, I'm sure I disturbed a number of people by doing that.

Okay, so that's this network.

Now, before we do the last piece 'cause there's one more, it's the WAN links, right?

First thing I want to say-- I want to ask you, are you getting there?

Are you kind of seeing how this works?

And you'd see why it only takes one example.

You're kind of like, "Okay, I get it.

It's multiple subnetting problems all in one."

That's what VLSM.

But I want to ask the second question that I get ask quite a bit.

A lot of people that are certification minded are like,

"How do even ask a question like this on the exam?"

Like I mean, obviously, the exam is a serious of multiple choice, drag and drop, you know,

simulation, all that kind of stuff, how do they ask a question like this

if they were going to ask me to do VLSM?

What will typically be used is drag and drop

to where what you'll see is a diagram just like I'm showing you right here.

And then you'll see a whole bunch of boxes.

And they'll just have subnets, subnet, subnet, subnet, subnets, all of these subnets.

So one box will be like 192.168.1.64/27.

One will be, you know, .96/27.

And you actually have to drag and drop those boxes to the specific areas of the network

where they belong and you just kind of plant them there

and that's how it knows whether you got the right answer or not.

So, now, and also keep in mind, the test is geared.

It's smart enough to know, you know, for instance, these two are interchangeable.

I could have put 96 through 127 over here and I could have put 64 through 95

over here, and that's totally fine.

That's not a problem, so good.

Are you ready to do the last piece?

Let's see where we can squeeze it in 'cause I want--

I like keeping them all on there so you can see what we've done.

Let's go small.

So I'm going to go down here one.

So the last thing is we need a few networks for our WAN links

of two host each, so let's start there.

Two host, what's the binary equivalent of 2?

Right there, 2 bits.

To get the number 2, I need 2 bits.

So step 2, original subnet mask /24 equals a bunch of ones,

and 24 ones, 1, 2, 3, 4, 5, 6, 7, 8 zeros.

So I say, "Okay, I need to reserve my bits in the mask and I'm going to save the host."

I need two host per network, 1, 2 are saved.

1, 2, 3, 4, 5, 6 flip over to the network side, the dark side,

I now have a /30 as my subnet mask or-- I'm not going to write.

Or let's write over here, 255.255.255.252 is the subnet mask so that is.

So again, two ways of writing the same thing.

So I'm going to come over here and I'm going to say my lowest network bit convert back

to a decimal, that'd be 1, 2, 4, that's a 4.

So now, my network range is we'll go up here, how is that?

So I'm going to start off with 192.168.1.0, right?

And I just start counting by 4s.

8, 12, now stop the train, that's going to take a while, right?

'Cause I look at all of these, 0 through 3, 4 through 7, 8 through 11, and I'll--

as I go down, that's going to take some time because these guys are all used up.

They're overlapping with this range or these ranges.

I mean, I would have to count by 4s down to 128

or what if we just went dot, dot, dot, 192.168.1.128?

Like I said, if you count by 4, you're going to hit every increment of 16, 32,

64, you're going to hit all those.

So it's very safe to say we left off at 128, let's start there.

128, 132, 136, 140, now, how many networks do I need?

Three. Good, I got them.

128 though 131, through 135, through 139.

So now, I can go line these guys up.

So top link, top WAN link is going to be 192.168.1.128 through 131/30.

Over here, 192.168-- hey, let me just write them all up.

Okay, there.

I've written them all up.

So I've now created networks for each one of those WAN links.

Let me just put there-- all there.

Now, then you might see, you know, on the exams something that has you drag the networks just

like I've done and kind of, you know, this would be your empty boxes

and you would drag the networks in there and label them.

Other ones might actually have you assign them.

So I mean, they might say, okay, this side would .129, this side would be .130, you know,

so those are the two useable IP addresses from that network or over here, you know,

our default gate way might be 192.168.1.97.

You know, the first valid IP address from here and that's what everybody points

to as their default gate way, maybe the computer.

I mean, a computer can be anything but I'll do that 1.-- I'm out of room, 98, you know,

/27, you know as the subnet mask.

So it depends on how that works, you know, what the question is looking for.

Many different ways of asking it but do you see how we've done that.

Isn't that amazing though if you look at it?

We've address this whole network and barely gone over half

of the total IP addresses that we have.

I mean that's the beauty of VLSM.

[laughs] Now, let me ask this question.

What if, you know, I said this is 60 users but look,

our subnet mask only allowed for, what is that total, 62?

What happens if this office grows and they hire five more people?

Your host because you explode your whole scheme, right?

You just outgrew the subnets and now you're like, well, you know--

yeah, that's where it just gets ugly.

So while VLSM looks awesome on paper and I don't--

hang on, don't throw the baby out with the bath water.

VLSM is awesome and it's used all over the place but it's not used in such a way

to put your subnets so tight that there's no room for growth.

You want to allow room for growth but you just don't want to be at a place

to where you're forced to waste IP addresses because of these giant subnets you had

to create for one area of your network.

So do you see why I said we'd only need one example of it?

VSLM really isn't anything new.

It's just doing what we've already done again and again and again and again.

I think, you know what, the newest thing is just to say, "Well, start with the biggest subnet,

you know, and kind of work your way down so that you get efficient use of your IP addresses."

But that is it, that's the culmination of everything subnetting.

The cool thing is at this point, you-- there's nothing that somebody could throw at you

from IPV4, you know, from normal TCPIP, where they're like, "Hey, check out this subnetting."

I mean, you've seen it all.

You've seen the binary behind it.

You've seen how to do it on networks, on host, reverse engineering and then finally,

getting super efficient using variable length subnet masking.

The only thing that I would add to this is this is truly classless,

not like you're a lesser person.

But this is truly classless IP networking.

There are, as we dive into routing and get into routing protocols,

there are some routing protocols, old routing protocols, like RIP version 1 and things

like that, that don't do this kind of thing.

They don't have the ability to do variable length subnet masking.

They can do subnetting but once you find that one mask, they have to use that for everything.

But that being said, nobody uses those protocols anymore.

So this is now a normal way of life.

I hope this has been informative for you and I'd like to thank you for viewing.