IP Addressing

IP Addressing.

Welcome to this section of the course

where we're going to start discussing IP addressing.

Now, you may be wondering what exactly is an IP address?

Well, an IP address,

also knows as an Internet Protocol Address,

is an assigned numerical label

that's used to identify internet communicating devices

on a computer network.

IP addresses are used at Layer 3 of the OSI model,

and they're also used by routers to be able to send data

from one network to another network.

Remember, when we're dealing with two devices

that are internal to our own network or LAN,

we're usually going to be dealing with Layer 2 addressing,

which is based on MAC addresses.

And that data is going to be forward around the network

via our network switches.

But, once we start going to two different networks

or even two different subnets,

we have to start using Layer 3 addressing

with our IP addresses.

Now, let me give you an analogy

that will help you understand

how IP addresses actually work.

Let's pretend that an IP address

is just like a mailing address for your home,

but we're going to use it for a computer.

Now, if I was going to send you a letter at home,

I have to have a way to uniquely identify your house

from all the other houses

that are located on the same street or within the same city.

Now, each house can stand on its own

and it can be built wherever you want it to be.

Regardless of whether it has an assigned address to it

on the front of the house or not.

This would be equivalent of me powering on my laptop

while not being connected to any networks.

Now, as long as nobody needs to know where my laptop is,

it doesn't really matter.

I can be anywhere I want,

because nobody's going to be sending me any data.

And in this case, I won't need an IP address.

Just like your house, doesn't need an address

if nobody has to send you any packages.

But if somebody wants to be able to go over to your house

or to send you a package,

they need to know how to get from one house to another.

And that's where you're going to need to know

the house number, the street, and the city name

so you can find it.

This would be the equivalent of having an IP address

assigned to my laptop.

Because once I have that IP address assigned

people can now send me data

and I can communicate with other people on the network.

Now, going back to my house example.

If you want to be able to communicate with people,

you need to know their house number, their street,

their city, and their state

that makes up their complete mailing address.

This way you can send them a package or a letter

and it will get to them, right?

Well, here in the United States,

we have a standard format for any letter or package

that we want to send and have it delivered

by the postal service.

The first line always has your house number

and a street name, something like 123 Main Street.

The second line is always going to have your city,

your state, and your ZIP code.

So it might be something

like San Juan, Puerto Rico, 00918.

This happens to be the information that we use

for Dion Training's office.

Now, when you're using the standard format,

the postal system is able to figure out

how to send the letter from your house to my office,

Because everybody knows the same common format

that we're all going to be using and how we can all read it.

So when a letter is going to be sent from your home

to my office, the post office first starts

at the second line, which is more generic.

This had my city, my state, and my ZIP code.

So the post office is going to look at the address

on your letter and determines if it can be delivered

by the same post office.

For example, if you happen to be in the same city as me,

our local post office will take it from my house to them,

and then from them to you.

You see in our postal system, each post office

is going to provide service to only one ZIP code.

So if you mailed the letter

from Beverly Hills, California,

that post office services the 90210 ZIP code.

And they're going to look at this letter

that you're trying to send to my office in Puerto Rico,

and they're going to see the ZIP code of 00918.

Those two ZIP codes don't match.

And since we're not in the same ZIP code,

that means they're going to have to go

and check the state or territory

that's listed on that second line.

In this example, the state or territory that we're using

is the island of Puerto Rico,

which is a territory of the United States.

So they're going to take the letter

from the Beverly Hills Post Office and forward it

to the distribution center,

which happens to be in Los Angeles, California.

This is then going to put on a truck

and go to that mail distribution center.

Now, once it gets to Los Angeles,

it's going to go on another truck and they're going to send it

to the South Eastern part of the United States,

to the mail distribution center

located in Jacksonville, Florida.

Once it gets to Jacksonville, Florida,

they're going to look at that city and state,

and see is it within Florida.

If it is, they load it on a truck

and send it to the general area,

either North Florida, Central Florida, or South Florida.

And then, it will keep getting forwarded,

more and more specific, until it reaches

the final destination city.

But in our case, this letter actually

needs to go to Puerto Rico.

So we can't put it on a truck.

Instead, we're going to put it on a boat

from Jacksonville Harbor.

And it's going to go from their main distribution center

to the regional distribution center,

located down in San Juan, Puerto Rico.

Now, once it gets down to San Juan, Puerto Rico,

this actually services Puerto Rico and the Virgin Islands.

Once it gets there,

they're going to look at that state territory line.

And they're going to see it's going to Puerto Rico,

which happens to be the same island this letter is now on.

So their regional facility,

which also happens to be in San Juan,

is going to look at that exact city that I wanted it to go to

by looking at the city and ZIP code.

In this case,

it needs to go from the main distribution center in San Juan

to the local post office in San Juan.

And because San Juan is a big city,

with over a million people in it,

there are a lot of different ZIP codes

and different post offices within the City of San Juan.

So they're going to check the ZIP code

and find out that this letter

needs to go to the San Juan Post Office,

that services the 00918 ZIP code.

So once again, our letter is put on a truck and it's moved.

This time, it's going from the regional center

to the local post office of 00918 in San Juan,

which will then allow it to be delivered to the end-user.

So, this letter now makes it to the local post office,

but it's still not done being moved.

At this point, we've done all the big bulk movements.

We've gone from state to state, and city to city.

And now, we are down at the local post office

that services my office,

but I don't have that letter in my hands yet.

So now, this post office at 00918

is going to check all the street addresses

for all the letters that they've received that day.

And they're going to put it onto the right truck

with the right mail person that delivers to that street.

Because each mail person usually delivers

to 10 or 20 streets per day.

Now, once this mail person has my letter,

they're going to deliver it by driving around the neighborhoods

until they reach my street.

And then, they're going to locate my office

by using the number located on our building.

In our example, 123, which was located on Main Street.

Now this is a ton of movement and routing,

and forwarding of this piece of mail

to get it from Beverly Hills, California, 90210,

to San Juan, Puerto Rico, 00918.

And this all happens in about a week,

and all for the cost of a first-class stamp

at around 60 cents.

Now that sounds like a pretty good deal to me, right?

Unless you're the mail carriers,

you have to make this all happen.

Now let's jump back to our networks and our computers.

The same type of thing happens with our data,

but we don't have to put a first-class stamp on it.

Instead, we're going to make sure it gets routed

until it finds the right location.

And that location is determined

by the IP address of the source and destination.

So if you want to communicate with other computers

outside your network or sub-net,

you need to know their IP address.

That becomes the destination.

An IP address can come in one of two formats.

IP addresses can either be IPv4 or IPv6.

This stands for Internet Protocol version 4

or Internet Protocol version 6.

In this section of the course,

we're going to discuss both of these formats

and why each one is used,

and how information is going to flow

at Layer 3 of the OSI model,

using these two different types of IP addresses.

Now we're going to be working

in domain one, Networking Fundamentals,

throughout this entire section of the course.

We're going to be covering two objectives in this section,

specifically Objective 1.4 and 1.6.

Now, Objective 1.4 is where we're going to spend

most of our time.

It states that given a scenario,

you must be able to configure a sub-net

and use appropriate IP addressing schemes.

Objective 1.6 will also be briefly covered

because it states you must be able to explain the use

and purpose of network services.

So let's jump into IP addressing

in this section of the course.

Internet Protocol version four addressing.

In this lesson, we're going to talk about IPv4

or Internet Protocol version four.

IPv4 is extremely popular

and is the most common type of IP addressing

that's used in our networks.

In fact, if you're like most people,

you've already seen an IPv4 address before.

When you look at them,

they're written as a series of four decimal numbers

separated by.s.

Some examples of this are IPv4 addresses like 10.1.2.3

or 172.21.243.67.

As you can see, each IPv4 address is made up of four parts

to form that address.

This is known as a.ted decimal notation.

When you're referring

to each of those four individual parts,

we call these an octet

because they each have a decimal number

that's used to represent an eight bit number.

Because these decimal numbers represent an eight bit

or eight binary digits,

this means that they can only represent a value

from zero to 255 and each of those four positions.

Now, when all four octets are combined,

we have four octets that contain eight bits each

for a total of 32 bits of total addressable space

when using an IPv4 address.

Now, for example, if I have the IPv4 address

of 192.168.1.4,

this is written in dotted decimal notation

to make it easier for us as humans to read it.

But in reality, it's actually

110000000

.10101000

.00000001

.00000100

if I put it in true binary form.

As you can see,

being able to use dotted decimal notation is a lot easier

for us to read and to type.

And therefore, it's less prone to us making mistakes

because as humans,

entering these numbers into our network devices using binary

would just be a big issue for us.

We just don't think that way.

Now, when we see an IP address like 192.168.1.4,

it's actually being broken down into two portions

by using a second 32 bit number known as a subnet mask.

Now, one part of the IPv4 address

is used to identify the network portion

and the other part is going to be used

to identify the host portion.

When you look at a subnet mask,

it's going to look a lot like an IPv4 address.

But if you convert it to binary,

you're going to see that it has a continuous strings

of ones or zeros to identify the network

and host portions of that address

being used by the client or device.

So if I have a subnet mask

like 255.255.255.0,

this is known as a default Class C subnet mask.

Now, don't worry about classes just yet.

We are going to cover them more in just a moment.

For now, I just want you to see

what a subnet mask looks like.

Now, if I converted this into binary,

each of those octets that are 255

are going to be written as 11111111.

So I'm going to get 11111111

.11111111

.11111111.

and then 00000000

This is because in binary, eight ones is going to equal 255

when I write it in decimal.

And if I have eight zeros in binary,

that is going to equal zero in decimal.

Again, don't worry too much about this conversion just yet.

We are going to do a lot of these different conversions

and math problems in a separate video lesson

when we talk about computer mathematics.

Now, when you look at this number,

if you see a one in the binary of that subnet mask,

this means it's part of the network portion

of the IP address.

If I see a zero in the binary portion of the subnet mask,

this means it's part of the host portion

of the IPv4 address.

So let's put this together by showing both an IPv4 address

and a subnet mask together.

First, we have our IPv4 address

of 192.168.1.4.

Next, we have our subnet mask underneath it

with 255.255.255.0

Now, everywhere I see that 255,

this is going to represent a one in binary.

So this becomes part of the network portion

of that IPv4 address.

If I use this example, 192.168.1.4,

the 198.168.1, that is part of the network.

Now, I can do anything that starts with 192.168.1.something

and that would all be addressable by the same local network

because they all share the same network portion.

Now, when I get to the second part,

anytime I see those zeros,

this represents all zeros in binary.

So that's going to be the host portion of that IPv4 address.

In this case, that is the .4 portion of this address,

which is going to represent the host.

That host could be a server, a desktop, a laptop, a tablet,

a smartphone, or any other network device.

It really doesn't matter.

But when we talk about this .4 device,

that is a single host.

When I talk about the 192.168.1,

that is the network that can contain up to 254 devices.

So if I have a device like 192.168.1.50

with a subnet of 255.255.255.zero,

that device is also on the same network

as our 192.168.1.4 device

and they can communicate with each other

using a switch and they wouldn't have to use a router

because they both share the same network portion,

192.168.1.

Now, on the other hand,

let's assume I have a device like 192.168.0.100

with a subnet of 255.255.255.0,

that device is on a different network.

Specifically, it's on the 192.168.0.something network.

So we cannot communicate that from our original device

at 192.168.1.4 without leaving our network

and routing our traffic over to this new network.

This 192.168.0.something network.

This is why we need to have a router.

Now, if this doesn't quite make sense yet,

don't worry too much,

we are going to dive deeper into subnetting

and you're going to actually calculate subnet maths

and things like that to learn how routing works.

Right now, we have barely just scratched the surface

in the examples we talked about.

And I just wanted to get you the idea

and introduce you to this concept.

Now, the next concept that we need to talk about

is that of IPv4 addresses

and how they're broken up into classes

or groupings of ranges

that can be used for different purposes.

Now, each class has its own default subnet mask as well.

When we talk about classes, we identify these by a letter.

These letters are A, B, C, D, and E.

Now, to identify the class for a given IP address,

you just need to look at the first octet.

If that first octet begins

with a number between one and 127,

it's going to be considered a Class A address,

and it has a default subnet mask of 255.0.0.0.

This means the network portion of that address

is just the first octet.

And the second, third, and fourth octets

are going to make up the host portion.

This means that with a Class A network,

we can have 256 X 256 X 256 hosts

on a single network, which means there are 16.7 million

possible host IP addresses

available for a single network address portion

assigned in a Class A.

Now, our second class we have

occurs when we have that first octet

beginning with a number between 128 and 191.

This would be considered a Class B address

and it'll have a default subnet mask of 255.255.0.0.

This means that the network portion of this address

is going to be the first and second octets

and the third and fourth octets

are going to make up the host portion.

This means that for a Class B network,

we can have up to 256 X 256 hosts on a single network,

and that means we get 65,536 possible host IP addresses

available for a single network address portion

assigned within a Class B.

Now, the third class we have occurs when the first octet

begins with a number between 192 and 223.

This is considered a Class C address

and it has a default subnet mask of 255.255.255.0.

This means that the network portion of the address

is going to be the first, second, and third octets,

and we save that fourth octet for the host portion.

This means that for a Class C network,

you can only have 256 hosts on a single network.

And this means there are only 256 possible host IP addresses

available for a single network address portion

that's being assigned.

Now, the fourth class we have occurs when the first octet

begins with a number between 224 and 239.

This is considered a Class D address.

Now, a Class D address

does not have any subnet mask assigned to it.

This is because Class D addresses are special

and they're reserved for multicasting or multicast routing.

Now, a multicast address is a logical identifier

for a group of hosts in a computer network

that are already going to be available to process datagrams

or frames intended to be multicast

for a designated network service.

So the actual multicast address

doesn't have to align with a single host,

but instead, it aligns with a group of hosts.

When you think about a multicast address,

I want you to think about it like a group chat on Facebook.

You might have a group chat name.

In our case, a multicast address.

And when you send a message to your group chat's name,

all of the members of that group

are going to get a copy of that message.

Well, the same thing is going to happen in multicast

when we're using IPv4.

We're going to talk more about multicasting later on

when we discuss the different types of data flows

within IPv4.

But for now, this is the idea

of what you need to understand with Multicasting.

I send it from one person

and it goes to multiple people all at the same time.

Now, the fifth class we have occurs when the first octet

begins with a number between 240 and 255.

This is called a Class E address

and it also has no default subnet mask.

This is because Class E addresses are also special.

In this case, they're reserved for experimental purposes

for research and development or study only.

This experimental range contains about 268 million addresses

that are reserved for future use.

Over the years, there have been a few proposals

to reallocate these Class E addresses for general use

because public IPv4 addresses

were becoming more and more scarce in the Class A,

Class B, and Class C ranges as more and more devices

start again connected to the internet.

That said, so far to date, these Class E addresses

still remain allocated for experimental use only.

And most IP implementations within our networks

will consider any IP in this range

from 240.0.0.0,

all the way up to 255.255.255.255

to be invalid as a source or destination

within a datagram.

And therefore, the datagram would be rejected

by the destination system.

So if you try to send something to a window server

or a workstation,

it's going to refuse to communicate with that device

if you claim to be from a Class E address.

All right, so now that we've covered

the five different classes of IPv4 addresses,

let's talk a little more about subnet masks.

Let's pretend that we have an IP address

of 182.168.1.4 again

with a subnet mask of 255.255.255.0.

This subnet mask is the default subnet mask

for a Class C network.

And since our IP address begins with a 192,

it is also a Class C address.

This means we have a Class C address

using a Class C default subnet mask.

So we consider this to be classful.

We call this a classful mask.

Now, a classful subnet mask

is just the default for a given class of IP address.

This doesn't mean though

that it's the best one for us to always use.

For example, if we're using a Class A address,

you may recall that it's default subnet mask

is 255.0.0.0.

This means we have a possibility

of having 16.7 million hosts on a single network.

Now, I don't know about you,

but I don't often come across networks that are that large

and require that many hosts.

In fact, I used to work

on one of the world's largest intranets

and we had a little over 1 million hosts

spread across six continents around the globe.

That was a really, really big network,

and we still didn't come close

to using all 16.7 million IP addresses

in a classful Class A subnet.

So often, what we'll want to do is break up these networks

into smaller networks.

So we often, instead, want to break down these large networks

into smaller networks.

To do this, we're going to use a process known as subnetting.

Now, we're not going to cover subnetting in detail

in this lesson.

Instead, I'm going to dedicate a few lessons later on

to fully dive into that concept

because you need to understand how to do it.

But for now, I just want you to remember

that we don't have to stick with a classful subnet mask.

Instead, we can use a classless subnet mask if we want to.

This is a process known as classless inter-domain routing.

This will let us borrow some of those host bits,

those zeros that I showed you in the subnet mask,

and then reassign them to the network portion.

So this lets me cut down the size of my networks

into much smaller portions with less hosts,

and this is more efficient.

This will give me at the same time,

a lot more networks that I can possibly use

because, again, if you think of it like a pie,

you can cut it many different ways,

but it's still one pie and it's a fixed amount of IPs

that we have in total.

So I can cut the pie in half

and we'd have two halves of the pie

or I can cut it in quarters

and we can have four pieces of pie.

But each of those pieces are smaller.

The same thing with our networks.

So for example,

let's say I have a classful Class C subnet mask

and I would have 255.255.255.0 as that subnet mask,

this allowed me to have 256 hosts, right?

Well, in my home network, I don't really need 256 hosts.

So maybe I want to break this down into four smaller networks.

If I take 256 and I divide it by four,

I get 64 hosts for each of those four networks.

To do this, I would change the subnet mask

from 255.255.255.0

to 255.255.255.192.

How did I do this?

Well, I borrowed two bits from the host

and I gave it to the network portion of the address.

And this is how I make four different subnetworks

or subnets out of this by using my subnet mask.

Now, later on in this section,

I'm going to teach you how to do subnetting

and understand how I got that number for that subnet mask,

including all the math involved, just like I did here.

But for now, I just want you to remember

that subnetting allows you to use a classless subnet mask

to create smaller networks

with fewer hosts in each of those networks

than you could if you had a classful subnet mask by itself.

This process is known as the classless inter-domain routing

or CIDR,

and we are going to abbreviate our IP addresses

using this CIDR or CIDR notation.

Now, when we do this,

we don't have to write out the subnet mask.

Instead, we just write the IP address,

a slash, and a number.

This is known as our CIDR notation.

So if I have an IP address of 192.168.1.4

with a subnet mask of 255.255.255.0,

I can abbreviate this using CIDR notation

as 192.168.1.4/24.

Oftentimes you'll hear this called CIDR or slash notation.

Now, if I had an IP address of 192.168.1.4,

but my subnet mask was 255.255.255.192,

I can abbreviate that

as 192.168.1.4/26

because remember, I borrowed two bits from the host portion.

So I went from 24 as my default subnet mask /24

into a /26 by borrowing two bits

and making my network portion that much larger,

bringing it from 24 up to 26.

Now, when you're trying to calculate your CIDR notation,

it's really going to be easiest

when you're dealing with classful subnet masks.

The CIDR notation here is really, really easy.

Now, for the classful subnet masks,

our CIDR notation is going to be rather simple.

If you have a Class A classful subnet mask,

you're going to have a /8 after the IP address.

This means that the subnet mask is 255.0.0.0

or it has eight bits of ones, which is the /8,

and then 24 bits of zeros inside the subnet mask.

Now, if you have a Class B classful subnet mask,

you're going to use a /16 after your IP address.

This means the subnet mask is 255.255.0.0.

Also, it means that it has 16 bits of ones

and 16 bits of zeros in that subnet mask.

Now, if you have a Class C classful subnet mask,

you're going to use a /24 after the IP address.

This means that the subnet mask

is going to be 255.255.255.0

or that it has 24 bits of ones and then eight bits of zeros

in that subnet mask.

Next, we need to talk about two different types

of IPv4 addresses.

These are called public and private IPs

and you may also hear them called routable

and non-routable IPs.

When an IP is considered a public or a routable IP,

this IP address can be directly accessed over the internet

and assigned to your network

by your internet service provider.

Routable IPs are publicly routable

across the entire internet,

and therefore they're globally managed by ICANN.

ICANN is the internet corporation

for assigned names and numbers.

So if you want a public IP address

like for running a web server for your company

or a Minecraft server for your kids,

you can buy that IP address

and ICANN is the one who will lease those out to you.

Now, ICANN has five different groups underneath them

who are responsible for actually handing out

these public IPs

based on where you actually live in the world.

There is ARIN, A-R-I-N, which is from North America.

There's LACNIC, L-A-C-N-I-C, which is for Latin America.

There's AFNIC, A-F-N-I-C, which is for Africa.

There's APNIC, A-P-N-I-C,

which is for Asia and the Pacific region,

and there's RIPE, R-I-P-E, which is for Europe.

These are the ones who are responsible for the management

of the public IP space on behalf of ICANN

in each of those different regions.

Now, since I'm located in America,

if I wanted a public IP address, I should go to Erin, right?

Well, not exactly.

Erin would be way too busy to handle all that.

They're not just going to sell me one IP address.

They're only going to sell me IPs

if I want a whole bunch of IPs.

So instead, I'm going to have to go to a reseller

who will sell me just a single IP.

In my case, I can contact my internet service provider,

somebody like Verizon or Comcast

and tell them that I want a public IP.

Then Verizon will pick me out

one out of their entire chunk from ARIN that they have,

such as a Class A with 16.7 million IPs

or a Class B with 65,536 IPs and assign one to me.

This way, Verizon can lease out that single IP

or a small subnet worth of IPs for my company

'cause maybe I need five or 10 of them.

Public IPs must be purchased before you can use them,

and you do this

through your local internet service provider.

Now, on the other hand, there are also non-routable IPs

known as private IPs because, well, they're not public.

Private IPs can be used by anyone at any time,

but only within their own local area networks.

This is why these IPs are considered not routable

because no one is controlling who's using them

and in which networks.

In fact, if you look at your IP address

of your computer right now, I bet you're using an IP address

that starts with either a 10, a 172, or a 192

as it's first octet.

Don't believe me?

Go ahead, pause this video, and check.

If you don't know how to check,

I'm going to tell you how to do it right now.

If you're on a Windows computer,

I want you to hold down the Windows key

and press the R key at the same time.

Then type CMD and hit Enter.

This stands for command.

Next, you have this black window on your screen.

Type in ipconfig for ipconfiguration,

and then hit Enter and look at your IP address.

Does it start with a 10, a 172, or a 192?

I bet it does.

Now, if you're on a Mac, don't worry,

I'm not leaving you out.

You can go ahead and look at this as well.

If you're watching this video

over a wireless network connection,

go ahead and hold down your Option key,

and then click on the wifi icon

in the top right of your menu bar.

Look down under your wireless network's name

and you're going to see your IP address.

Again, I bet it starts today with either a 10,

a 172, or a 192 as part of that IP.

Now, am I a magician

being able to tell you what your IP address is?

Well, not really.

You see, those three values

are part of what we call the private IP ranges.

This includes something like 10.0.0.something

or 172.16.1.something

or 192.168.1.something and a bunch of other IPs as well.

So if you're using one of these private IPs

and I said they're not routable,

how are you actually getting out to the internet

to watch this video?

Well, when you go out to the internet,

your router actually conducts a little trick

known as network address translation,

and it changes your private IP into a public IP.

Now, we're going to cover the concept

of network address translation and port address translation

in a separate lesson

because it's really important for you to understand that.

But for now, let's stick to talking about these private IPs.

All right, for the exam

and your life as a real world network technician,

it's going to be really important for you to understand

the ranges for private IP addresses.

Now, these are defined in something known as RFC 1918.

Now, RFC just stands for request for comments.

And request for comments is a formal publication

from the Internet Society

and the Internet Engineering Task Force or IETF.

These RFCs are authored by individuals

or groups of computer scientists

to document new technologies or standards.

In this case, the RFC 1918,

this is going to be used to document how organizations

can conduct address allocation for private intranets,

which we now call intranets or your local area networks.

Now, in RFC 1918,

there are specific ranges of private IPs described

within Class A, B, and C that anyone can use.

If you're looking at Class A addresses,

anything that starts with a 10 in the first octet

is going to be considered a private IP.

So you can have anything from 10.0.0.0

all the way up to 10.255.255.255 as your address,

and it's going to be a private IP.

This gives you a total of 16.7 million IP addresses

that anyone can use.

Now, in addition to that,

we also have some Class B addresses,

and this one's a little bit harder to memorize.

For Class B addresses,

anything that starts with a 172.16,

all the way up to 172.31

is going to be part of a private IP range.

This includes over 1 million IP addresses

because we have 16 times 256 times 256.

Now, if you're looking at a Class C address,

anything that starts with a 192.168

is considered a private IP.

This includes the range of 192.168.0.0,

all the way up to 192.168.255.255.

This gives you 65,536 IP addresses

that are there for you to use if you want them.

Now, I want you to commit these ranges to memory.

Remember, Class A is really easy.

Anything that starts with a 10.something.something.something

is a private IP in the Class A range.

Now, Class C is also pretty easy

because anything that is 192.168.something.something

is also considered a private address.

In this case, a Class C private address.

But Class B is where most people are going to struggle

because it's a little bit different.

It's going to contain any addresses

that start with 172.16.something.something,

all the way up to 172.31.something.something.

This is essentially 16 classes of Class B addresses

that are all next to each other,

and you could use any of them as a private IP.

Now, on test day,

compTIA may try to trick you and say something like,

which of these addresses is not a private IP?

And then they're going to give you something like

172.12.something.something.

Now, this starts with a 172,

so it looks like a private IP, but because it's 172.12,

it's not between 172.16 and 172.31.

So 172.12.something is actually a public IP

because it's outside of my private range.

You have to be especially careful

when you see an address that starts with 172

because it has to be between 172.16

to 172.31 to be a private IP.

All of the other 172 something addresses will be public.

Now, the next thing we need to talk about

is some specialized IPs.

And there are two big categories that we need to cover.

The loopback address and APIPA addresses.

Now, the first special IP is the loopback address.

This is assigned as 127.0.0.1.

Now, when this was created

all the way back in the early days of the internet,

the designers weren't too worried about wasting IPs

because they never envisioned

that we were going to use a lot of IP addresses

in the entire world.

So they simply dedicate an entire range

that's 127.0.0.0/8

or 16.7 million IP addresses

to be used as internet host loopback addresses.

Now, this allows any higher level protocol

to send data to the host itself

without actually going out to a switch or a router.

Essentially, this creates a loopback to the host

and it tests your networking protocols.

So it's often used in troubleshooting

and testing network protocols on a given system

to make sure your drivers are working properly.

Now, because of the way the standard was developed,

anything you see that starts

with 127.something.something.something

is considered a loopback address.

Although most people just use 127.0.0.1

as our loopback address by default.

This means the other 16.7 million IPs

are pretty much wasted because we're using this range

to be the entire loopback range,

even though most of us only use this one IP address.

Now, you may have heard the old joke

that some network technicians like to use.

There's no place like 127.0.0.1,

which means there's no place like here, home, right?

That's 127.0.0.1.

It's the local host.

Now, and during the COVID pandemic,

I also saw a couple of funny memes saying,

stay at 127.0.0.1

and wear a 255.255.255.0,

which essentially says, stay home and wear a mask.

Now, when it comes to the IP address of 127.0.0.1,

you're going to hear it called either the loopback,

or you might hear people call it the local host

like I did earlier.

Now, the word local host will always resolve to 127.0.0.1

on every computer as part of its local DNS settings.

So if you ping 127.0.0.1 or you ping local host,

you're going to get the same result.

It's going to resolve to the IP address of that loopback IP,

127.0.0.1.

Now, the second special IP address we have

is known as APIPA, A-P-I-P-A.

This is also known as the automatic private IP addresses.

These addresses are dynamically assigned

by your operating system

whenever your DHTP server is unavailable

and an IP address has not already been statically assigned.

APIPA addresses are always going to start

with 169.254.something.something.

So you're going to find them in the address range

of 169.254.0.0,

all the way up to 169.254.255.255.

So if you ever see an IP address in this range

when you look at the IP of a network device,

it means that there is something wrong with the DHCP process

and the device isn't getting a normal private IP

from one of our Class A, Class B, or Class C ranges.

Now, as your workstation boots up,

it's going to attempt to get its own IP address

using dynamic IPs, using the DHCP protocol.

This goes through a four-step process known as DORA.

This is discover, offer, request, and acknowledge.

If something goes wrong

with this DORA negotiation process with DHCP,

the system simply can't get an address

and your computer would eventually just crash

because it wouldn't know what to do

and it would keep trying over and over again.

So what the brilliant engineers

of the Internet Task Force did was create this APIPA range.

Now, basically, what it says is that if a workstation

can't get a DHCP assignment for a dynamic IP address

within some amount of time,

the workstation is simply going to pick its own address

from this special APIPA range.

Basically, any IP it wants

as long as it starts with 169.254.something.something.

So if you find a computer

that cannot connect to the internet,

the first thing you should do is check its IP address.

You can do this using ipconfig on Windows

or ifconfig on Mac or Linux machines.

Now, if you see an IP address

of 169.254.something.something.

You know you have a DHCP problem,

and you need to check your DHCP server

to ensure it's working properly

and that it's handing out private IP addresses

from Class A, Class B, or Class C assigned ranges.

Now, that's as deep as we need to go into DHCP right now,

but I promise we are going to go back to DHCP later on,

and we'll talk about all of these things

and how it provides these IP addresses

and how those are going to be used inside your networks.

Now, the last two things we need to discuss

in terms of IPv4 addressing

is the concept of virtual IP addresses and sub-interfaces.

Now, first, let's talk about virtual IP addresses,

usually abbreviate as VIP or VIPA.

Now, a virtual IP address

is an IP address that does not correlate

to an actual physical network interface.

Instead, these virtual IP addresses

are usually used for network address translation,

fault tolerance, and virtualization.

In your computer, you have a network interface card.

And normally, you can only configure

that network interface card with a single IP address

because there's only one physical network interface,

namely this network interface card.

Now, with a virtual IP address,

you can actually configure that network interface card

to respond to numerous IP addresses

and have them all resolve

back to your physical network interface

to establish the connectivity.

Routers will often use virtual IP addresses

to provide redundancy in their connectivity options as well.

By having a single virtual IP

assigned to the default gateway, for example,

we can have multiple routers

that could answer up on behalf of that virtual IP.

This would allow us to have a primary device

to use during normal conditions

and an alternate device that could be used

when the primary device fails.

So we could fail over to that secondary device.

We'll talk more about this concept

when we start digging into routers

and covering concepts like redundancy.

But for right now,

I just wanted to introduce you to the concept

that virtual IP addresses exist

and we use them for things like redundancy

and giving an interface,

one physical interface, multiple different IP addresses.

Now, the second concept we need to cover here

is the idea of sub-interfaces,

which is a virtual interface

that's created by dividing up one physical interface

into multiple logical interfaces.

This goes hand in hand

with the concept of a virtual IP address

because each of these virtual interfaces

that we're going to create

is going to be able to be a sign of virtual IP address as well.

These sub-interfaces are often used

for inter-VLAN routing of traffic

by creating two sub-interfaces for one physical interface,

and then assigning each of those sub-interfaces

with an IP address from the appropriate subnet.

This then allows the router to connect these two subnets

or VLANs virtually instead of requiring us

to dedicate two physical network interfaces

and attaching a real patch cable between them.

All right, I know that was a ton of information

in this long video.

So if I went too fast for you,

please watch this lesson a second time.

There is a lot of important information

about IPv4 addressing that you just have to know.

It is critical you understand

the different classes of IP addresses,

the public and private IPs, and when each of these are used,

as well as the different special IP addresses that we have

such as the loopback, APIPA, and others.

When we talk about the way data flows in a network,

there's really three different ways it can flow

if you're talking about IP version 4.

IPv4 allows us to have unicast, multicast,

and broadcast data flows.

First, we have unicast.

And unicast is when data travels

from a single source to a single destination.

So, if you want to think about this like

me picking up the phone and calling you,

it's a one-way conversation between you and me.

There's only two people who are talking, me and you,

and it's unicast going from me to you, or from you to me.

Second, we have multicast.

And this allows data transfer from a specific source

to multiple but specific destinations.

For example, maybe I'm at a classroom

with three different people.

I can now talk to all three of them at once,

because I'm talking to the three people

who are sitting in that classroom,

that's going from me to those three specific people.

Now, broadcast, on the other hand,

is when data travels from a single source

to all the sources on a destination network.

If you think about this, if I go and do a radio broadcast,

I might be speaking on the radio,

and I don't know who I'm talking to.

I just know that I'm talking to anybody

who happens to tune in and listen.

That's the big difference between

unicast, multicast, and broadcast.

Let's look at each of these a little bit closer

by looking at some diagrams.

First, we have unicast.

And let's say I have unicast and I have one server

who wants to send messages to PC1 and PC2.

If I put that message in an envelope,

also known as a packet, I can then send it out

based on its IP address from the server to PC1,

or from the server to PC2.

And you can see that here on the screen.

Now, when I go into multicast, I can send

just a single message and it will be directed

to whoever I want it to be into

as part of their multicast group.

So, my server, in this case, is sending out a message

to multicast group number one at 239.2.1.3.

When it gets to the switch,

the switch determines who's part of that multicast group,

and in this example, PC1 and PC2,

both get a copy of that message.

That's the great thing about multicast,

is I don't have to repeat the message twice,

even if it's the same message,

because it's going out and hitting that switch,

and getting repeated to everyone in that group.

This works very much like

when we're doing things like broadcasting video

using a service like a live stream.

And I can send the message once to my Facebook group,

and then Facebook can multicast that out

to all of the users who want it.

Next, we have broadcast.

And when you're dealing with a broadcast message,

this is when the server sends it out and says,

hey, switch, everyone who's connected to you,

tell them all about this message.

By doing that, it's going to broadcast that out

to everybody on the network.

So, if you're using a standard Class C network,

that would actually go out to 255.255.255.255.

And when it goes out to the switch,

it's going to send it to everybody who's connected to it.

And in this case, that's PC1, PC2, and PC3.

Now, some students get confused

between multicast and broadcast.

The key distinction there is with broadcast,

everybody is going to get it.

But with multicast, only those who have opted into it

are going to get that message.

And that's the key distinction there.

Assigning IP addresses.

Now when we have our networks

and they rely on IP addresses like IPv4,

how do we tell our devices

what addresses they're going to have?

Well, there are really two different methods we can use.

One is to manually or statically assign them,

and the other is to dynamically assign them.

Now, when I use a static assignment,

this is a really simple process.

As a technician,

I manually will type in the IP address for the host,

its subnet mask, its default gateway

and its DNS server.

But this can be time consuming and prone to error.

For example, let's say I have 20 devices on the network.

Now I'm going to have to go

and assign those four pieces of information

20 different times, once for each device,

that's 80 places I have to enter information.

This means there's a lot of chance for human error here,

because these numbers are very easy to mistype,

and if you mistype one,

you're going to be assigning the wrong information

to the wrong devices.

Or you might have the same information on multiple devices.

And this will also cause problems

or conflicts between two devices.

So as you start to work on large enterprise networks,

it becomes very impractical to do static assignment

of the IP addresses for all your devices.

Some of our networks in the past that I've run

have been 500 clients, 1,000 clients,

5,000 clients, 10,000 clients,

100,000 clients, or even 1 million client computers

in a large-scale intranet.

So for us to assign those all statically

and keep track of all of those different IP addresses,

would become a full-time job

for a large team of people

located at all over the world

because that large intranet spans six continents.

That would be a waste of a lot of time,

money, labor, and resources.

So instead we, simplify this process

by using dynamic allocation of IP addresses.

This is known as a dynamic assignment.

By doing this, we have a quicker,

easier and less confusing method of assigning our IPs

to all of our network clients when they join the network.

Now for larger small networks,

using dynamic IP addressing

is usually going to be your best option.

In your home, whether or not you know it,

you're already probably using dynamic IP addressing.

When you bought a new smartphone or tablet

or laptop or desktop,

you took it out of the box, you powered it on,

and then you joined your wireless network,

and you were able to go online and browse the web, right?

You didn't have to do any kind of crazy configurations.

Well, in this case,

you didn't assign your new device an IP address,

a subnet mask, a default gateway,

or a DNS server to use.

Instead, your network's DHCP server

did all that for you automatically,

without you even having to ask it to.

This is because most small office

and home office network devices like your cable modem,

fiber modem, or wireless access point

already have a running DHCP server there for you

and it's turned on by default.

You just tell the device what network to join

and your router will use DHCP

to hand out a dynamic IP address

for your network client to be able to utilize.

So what are those four components

of a fully configured client?

Well, I've said it a couple of times,

and whether you're using static or dynamic assignment,

you still need to use the same four components.

This is an IP address,

a subnet mask, a default gateway,

which is usually just the IP of your router,

and a server for either DNS or WINS.

Now, DNS is the domain name system.

DNS is going to be used to convert domain names

used by a website to the IP address of it's server

so that your computer can connect to it.

Now we're going to cover DNS more in its own video,

because there's a lot you need to know about it.

But for right now,

just realize the DNS is essentially

the internet's version of a phone book,

where we can look up a name

and get a number to connect directly to.

Now names to numbers and numbers to names,

that's what DNS is all about.

For example, when you went to diontraining.com,

you're going to be using DNS in the background

to determine what the IP address is of my server,

so you can connect to it

and access our web pages or our videos.

That is DNS at work.

Now WINS, on the other hand,

W-I-N-S, is used within a local area network,

specifically WINS is used in Windows domains,

and it is known as the Windows Internet Name Service.

It's used to enable Windows

to identify NetBIOS names on a TCP/IP network

and convert those NetBIOS names to IP addresses.

Basically, WINS is like DNS,

but it only works within a Windows domain environment.

So if I wanted to connect to my mail server

inside a Windows domain,

I could type in its IP address, if I knew it,

or I could simply type in the name of the server,

something like mailbox

or whatever I have named it.

Now, when it comes time to do the dynamic assignment

of the critical addressing information for each client,

we can use four different methods to do this.

This includes BOOTP, DHCP, APIPA, and ZeroConf.

BOOTP is by far the oldest

and least used of these four options.

BOOTP or the Bootstrap Protocol

was originally introduced in 1985

for use in disclosed Unix workstations,

because it could dynamically assign

the IP address information

and then allow the workstation

to load a copy of their boot image over the network.

Now, BOOTP used a static database of IPs and Mac addresses.

So essentially whenever a client connected to the network

to initiate the BOOTP process,

it would find its Mac address inside its database,

and then send the proper IP address that matched it

back to the requesting client as its assignment.

This wasn't as dynamic as we would like.

So in 1993, a newer updated protocol known as DHCP

was introduced to replace BOOTP.

Now DHCP, or the Dynamic Host Configuration Protocol

is going to allow the assignment of an IP

based on an assigned scope or pool of addresses,

as well as it provides the ability

for us to configure numerous other options within it.

Now, since DHCP allows me to configure my scope,

I can actually tell my DHCP server something like,

"Hey, I only want you to hand out addresses

that are from 192.168.1.100

up through 192.168.1.200."

And this gives you about 100 clients

that can now be automatically assigned.

Every time somebody connects to the network,

the DHCP is going to send out one of those IPs from that range

and assign it for a given period of time

known as a lease to a client.

Now, each IP can be borrowed for a certain amount of time

from this pool, and when that lease expires,

the DHCP server is going to pull back that address.

Now this doesn't actually cause a problem for us,

because your computer at any time can say,

"Hey, I'm still using that address.

You can't take it."

And in that case, the DHCP server would say,

"Oh, okay, you can keep it."

And reassigns it for another period by renewing its lease.

It's just like if you get a book from the library.

Say you were halfway through reading it,

and it's going to be due tomorrow,

you can take it back to the library and recheck it out.

It's the same concept with DHCP and dynamic addresses.

Now when the lease does expire

and it's no longer needed by the client,

it's going to be returned back to the scope or that pool,

and be ready to be issued to another client.

Essentially, each client can borrow that IP

during its assignment

and then return it whenever it's done with it.

Now this IP management

is going to be performed by the DHCP server on our behalf.

And it's going to use to manage all of these IPs

that are being assigned and returned over time.

This is great because we don't have to control it ourselves

or keep track of it all manually.

Instead, we have the ability to go into it at any time

and look at the logs and say,

"Hey, who was using the IP 192.168.1.132

on September 9th at 3:00 PM?"

And then we can figure it out

using DHCP's IP address management and their logs.

And so this gives us all the benefits of figuring out

who did what,

while still not having to do any of the management

and oversight of handing out these IP addresses.

Now, another great thing about DHCP

is that it gives our clients

all of these different variables

that they need to communicate.

This includes the dynamic IP address that's being assigned

as well as the subnet mask,

the default gateway and the DNS server,

and if you're using a WIN server,

you can also send that through DHCP.

This can all be done

using the DHCP protocol for us automatically,

each and every time a new client connects to the network.

Yes, I know I've repeated these four configuration options

a bunch of times now.

And you know what that means?

It means this information is really important.

So you simply have to know those four configuration items

that DHCP provides to your clients

and keep them in mind for test day.

Remember that the IP address, the subnet mask,

the gateway, and the DNS server's IP.

The WIN server is an optional component

that may or may not be sent.

All right, we're going to talk a lot more about DHCP

in a separate video, but for now,

you need to remember that DHCP

is the modern implementation of BOOTP,

and it's commonly used in our modern networks

to be able to assign automatically the IP address

and other required data

for a client to communicate on a network.

Now, the third way we can do automatic or dynamic addressing

is by using APIPA, A-P-I-P-A,

or the Automatic Private Internet Protocol Addressing.

Basically if for any reason

DHCP cannot complete the assignment process

or find an address for you to give to the client

because you ran out,

then APIPA is going to be used instead.

This kind of thing can happen if there's a problem

where the client can't reach the DHCP server

because of networking issues or something else like that.

Now in these cases,

the client is going to assign itself an APIPA address,

which is a self assigned address.

Now by default, on a Windows server or workstation,

you're going to find that APIPA is selected by default

under the TCP/IP properties

under the Alternate Configuration tab.

Now this allows the Windows machine

to assign itself an address randomly

from the 169.254 dot something dot something scope,

if it can't reach a DHCP server

or it can't contact it

and finish the negotiation process.

Now APIPA is designed

to allow for a quick configuration of a local area network

without the need of having a DHCP server.

For example, if I take 10 clients

and I connect them all to a switch without a DHCP server,

these 10 clients will default

to picking up their own IP address from the APIPA range,

which is 169.254 dot something dot something.

Again, because it's a class B address,

this is perfectly fine,

because they're all going to be on the same local area network.

So if I wanted to play Doom on this local area network

with these 10 machines, that's fine.

They'll all find each other,

and they'll talk based on their APIPA addresses

without any issues at all.

Now, the only problem that will occur

is that these are private IP addresses.

So they can't be routed outside of our local area network.

Therefore, if we need to communicate locally,

we can do that using a switch,

and everything will be fine.

But we're not going to be able to reach the internet

because we don't have the same network as the router,

because the router had a valid IP address,

not an APIPA address.

And so we don't have a default gateway

to get out of this local network we've created.

Now, this is the biggest challenge you're going to have

when you have an APIPA address assigned to your clients,

because they cannot communicate

outside the local area network

or with other devices that don't have an IP address as well.

If you ever have a computer that starts with 169.254

dot something dot something,

and you can't figure out

why it's not connecting to the internet,

well, that's your reason.

It's an APIPA address,

and APIPA addresses can't get out past the router.

And so APIPA addresses are not going to allow you

to connect to the internet.

Our last dynamic method of configuring an IP address

is known as ZeroConf or Zero Configuration.

Now ZeroConf is a newer technology

that was based on APIPA

and can provide you with a lot of the same features as APIPA

as well as some new ones.

For example, ZeroConf can actually assign

an IPv4 link-local address to a client.

This is a form of a non writeable IP

that's used on a local sub-net, just like APIPA,

but the big difference is that with ZeroConf,

this client can now have the ability

to resolve computer names to IP addresses

without the need of DNS,

by using something known as mDNS

or multicast domain name service.

Also ZeroConf can perform a service discovery on a network,

so it can find out what things are connected

and available for use.

So if there's a printer, a scanner, or show file system,

you can actually find that using ZeroConf.

There've been lots of different implementations of ZeroConf

in recent years,

and it's called different things

depending on the implementation

and the product line you're using.

For example, on Apple products,

ZeroConf is actually called Bonjour,

and it's used mostly for service discovery of other clients

end devices on the local area network.

In Microsoft Windows,

they like to call it LLMNR,

Link-Local Multicast Name Resolution.

And it's going to rely on it as an extension of APIPA

to provide for name resolution and service discovery,

in addition to providing network conductivity.

Now if you're using Linux,

ZeroConf is usually implemented using SystemD

or the System Daemon Service,

specifically the systemd-resolved background service.

So remember, there are lots of different ways

to assign IP addresses.

You can do it manually known as a static assignment

or automatically known as a dynamic assignment.

If you're using a dynamic assignment,

you can do this with one of four methods:

BOOTP, DHCP, APIPA, or ZeroConf.

Really, it all depends on the needs of your clients

and your network.

Now, at this point of the course,

we're about to jump into the area that most students dread

when it comes to networking.

And this is known as subnetting.

Now, don't worry, I'm going to get you through it.

But this is probably one

of the hardest concepts on the exam,

you may need to go through this section a couple of times

to make sure you grasp it.

Now, what we're going to do is I'm going to go through

and we're going to do the hard way first,

so you understand the theory and how it all works.

Then, I'm going to show you a shortcut

and how to get through it for the exam a lot easier.

But please don't skip the next couple of lessons

of going through the hard way first,

because if you don't understand the hard way,

the easy way will make zero sense to you.

So, as we go forward, we're going to stop and take this lesson

to talk about computer mathematics.

Now, this is important because computers do their numbering

and do math a little bit differently than we do.

Now, as a kid, you learn to count in what's known as Base-10.

This is a decimal number, right?

Decimal meaning 10.

And so, when I count, you would start at zero,

and it goes zero, one, two, three, four, five, six , seven,

eight, nine.

Now, what comes next?

Well, we go to the second column, we put a one there,

and we go back to zero.

So, now we go from nine to 10.

And then, we keep counting up,

and eventually, we get to 98, 99.

And then go, "Oh, I need another column,"

put another one there and add two zeros.

That's 100.

So I go 98, 99, 100, right?

And that's the whole idea here

when we talk about decimal, right?

When we talk about decimal,

we only have 10 choices for the numbers.

Each decimal place can only be zero, one, two, three, four,

five, six, seven, eight, nine.

Those are 10 options.

Now, computers and networks

don't actually understand decimal natively.

So, even though we've been using decimal

when talking about IPv4 addresses,

that's not how computers see them.

Computers do everything in binary,

which is base-2 numbering.

That means, every single digit is either a one

or it's a zero, that's an on or it's an off.

That is how computers understand things.

And so, when they count, they go 011011100111

and they keep going like that.

The one zero is actually the number two for us.

But for them in binary, that is actually a one zero

and one zero equates to two, because it's the third thing,

we got to zero one two.

Now, that's the way this works with a computer.

And so, we have to understand

how things are written in decimal

and how things are written in binary,

and how to convert between the two.

And that's what we're going to talk

about in this particular lesson.

Now, when we convert from binary to decimal,

we're going to do this using a table

that I have here on the screen.

Each number is a factor of two.

And so, this way, we can start going

and putting them as placeholders of one

or zero as we go through each place.

So, if we start on the right, and we go left, we have one,

and then, we have a place of one, right?

Now, if I put a one zero in the second and first columns,

this is going to become two.

And we keep going through as we keep counting.

Now, if you're not getting it quite yet,

you will by the end of this lesson, I promise.

So, let's just keep going ahead

and we're going to do some examples together.

Let's say I give you the binary number of 10010110.

And I asked you to tell me

what is this if I put it into decimal?

Well, we're going to first by populating the table

starting from the right and going to the left.

And so, we're going to put the number there

that we have from right to left

and end up with what you see here on the screen,

it turns into 1001011, and zero.

Now, there's a place where there's a one,

and I'm going to add the number above it.

So, if I see something like 128, plus 16, plus four plus two,

and I add that all together, I'm going to get 150.

That is the number 10010110 in binary,

and it translates to 150 in decimal.

You see how easy that is if you have the little chart.

So, if you make this chart on your paper,

when you sit down for the exam,

it's going to make things a lot easier for you.

Again, going from right to left on the chart,

each position starts out with the number of one, two

four, eight, 16, 32, 64, 128,

we keep multiplying by two as we go from right to left.

And so, if I added up every single one

of those with one in it,

guess what the biggest number I could get is? 255.

Which, if you remember when I talked about IPv4,

all my numbers were either zero all the way up through 255.

That's my four octets.

Well, that's because I have eight digits

in each of those octets, right?

And each digit, each binary digit,

can only hold a one or a zero.

Now, what if I gave you a decimal number,

like from an IP address

and asked you to convert that to binary?

Could you do it?

Well, the answer is yes,

it's the exact same process we just did.

So, let's take a number from 167

and convert it backwards to binary.

That'd be one of the octets inside of an IP address,

for instance.

Well, instead of adding up,

we're going to to be subtracting.

So, when I do this, I'll take 167

and I'll start from the left.

Can I take 128 out of 167?

Yes, I can.

And I'll have 39 leftover.

Then, I go to the next column.

That's 64.

Can I take 64 from 39?

No, because there's not enough there.

So, I'll put a zero under the 64 column.

Then, I'll say, Can I take 32 out of 39?

Yes, I can.

And I'll have seven left over.

So, I'll put one in that column.

Then. I look, can I take 16 from seven?

No, so I put a zero.

How about eight from seven?

No, so I put a zero.

How about four from seven?

Yes, and that leaves me with three.

Can I go ahead and take two from three?

Yes, I'll put a one in that column.

And then, can I take one from one?

Yes, and I'll put a one down for that.

And this is how you convert from decimal back to binary,

you're going to subtract all the way down.

Just like what I went from binary to decimal,

I ended up going from decimal to binary,

I'm going to go ahead and subtract.

Now, either way, if you have this chart down,

these factors of two,

it makes these problems very, very easy.

So, now that we have that example underway,

and we see that 167 is actually 10100111.

How can I check my math and make sure I got it right?

Well, I can go the other way.

And I can add up the numbers and they all come back to 167.

That means, I did the problem correctly.

So, let's check our answer,

128 plus 32 plus four plus two plus one, that equals 167.

That means, we did our math problem correctly.

Now, in the next lesson, I'm going to give you some problems.

And we're going to try these together

and see if you understand these concepts

because it is really, really important

as we go into subnetting.

Alright, now that we got those computer mathematics

under our belt, let's dive into the concept of subnetting.

Subnetting is where we can take a large network

and we split it up into smaller networks.

When we do this, this is all about logical IP addressing.

The default classful subnet mask

is rarely going to be the optimal subnet for our subnet size.

So, instead, we use subnet masks to modify the subnets

and create networks that are much better in scope.

For instance, let's say I gave you a network

of 10 dot something dot something dot something,

this is a class A address, that means

there are 16.7 million IP addresses in that range.

You may have a large network,

but your network is probably not 16.7 million clients.

So, we would want to create subnets

that allow us to borrow bits from the original host portion

and then we can add them into the network portion

to make smaller networks for us.

So, for example, if I had a network like 10.0.0.0/8,

that is a classful class A subnet,

and it has 16.7 million hosts.

But I probably need something like

a couple of hundred hosts, maybe 256 of them.

Well, if I wanted to do that,

I might use the class C subnet mask of slash 24

and use that, instead.

By using 10.0.0.0/24,

that's going to give me 256 IPs in that subnet,

and that means, those other 16 million

can be used by other people.

Then, maybe I'm going to create a second subnet,

like 10.0.1.0/24, and then I can create a third subnet

with 10.0.2.0/24, and now I've used three subnets

of 256 IPs each, and I still have millions of leftover IPs

that I can subnet out later.

This is why it's really important to use subnets,

because it's an efficient use of the IP addresses you have,

instead of just using the classful default.

Even in your home network,

if you're using a class C address

of something like 192.168.1.0/24,

this gives you 256 possibilities,

but you probably only have five or 10 devices

on your home network.

Now, by doing subnets, this is going to allow us

to create additional VLANs in subnets

and allow separation of our networks for better security

and giving us better bandwidth control.

Here on the screen you can see a chart

that has the addresses for the class A, B, and C subnets

and their default masks,

which should be a review for you at this point.

Now, on the right side, you're going to see

the assignable number of IP addresses,

which is for class A, of 16.7 million addresses,

for class B, 65,000 plus IP addresses,

and class C, 254 usable IP addresses,

that is a lot of IP addresses.

Now, we may want to start nicking those down

into usable chunks,

and that is what we're going to use subnetting for.

If we're using private IPs, this really isn't a big deal,

because they don't cost me anything.

But with public IPs, every public IP I paid for,

and that's money out of my pocket,

and you want to minimize that

to only getting the ones you need.

For example, my ISP charges $5 per month for a static IP.

If I need one, that's $5, if I need 10, that's $50,

if I need 100, that's $500,

and so, that can add up pretty quickly, right?

By subnetting, it allows me to just get the things

that I need and the fewer IPs that I need.

Now, how do we do this?

Well, that's what we use a subnet mask for.

We talked about class A, class B,

and class C default subnet masks

back when we talked about it before,

and you'll see those on the screen in red.

They are the slash eight, slash 16, and slash 24.

Now, when you convert those to binary,

you can see that they're filled by all the octets

with either all ones or all zeros, making them classful.

Now, the ones on the bottom of the screen,

from 25 down to slash 30, are going to be smaller subnets.

These are called classless subnets

because they're going to use different things

besides all ones or all zeros in that octet.

Now, as you can see, we're borrowing bits

from the host portion and making all those zeros

and taking some of those by adding them over

into the network portion, as well.

And so, this borrowing is what we're doing with subnetting.

If you don't get it yet, don't worry,

we are going to dig deeper into it

as we continue through this lesson.

Now, let's take a look at some formulas

that we're going to need to understand for calculating

the number of subnets and the number of assignable hosts.

If I'm calculating the number of subnets,

the formula is two to the s power,

where s is the number of borrowed bits from the host.

So, when I start taking those zeros from the host

and turning them to ones to make them

be part of the network portion,

that is the s that we're talking about here.

For instance, if I'm using a slash 25 as my network,

I'm borrowing one bit from the host space

and putting it into the network space.

Now, this would make it two to the one

and that would give me two, and so that tells me

I can have two subnets by using a slash 25.

Now, if I'm looking at the number

of assignable IP addresses,

this is going to be two to the number of host bits minus two.

So, in my case of a slash 25 network,

there are 32 total bits in the address.

I have 25 of those being assigned for the network portion,

leaving me with seven host bits.

So, if I take two to the seventh and then I minus two,

I get 128 minus two, or 126 available.

Now, why do I have to have that minus two there?

This is a concept that a lot of students miss,

so, really pay attention to it.

Every network out there has to have two IP addresses

to be a network.

The first is known as the network ID,

and it's the first IP in the range for that network.

And the other one is a broadcast ID,

which is the last IP in the network.

So, no matter what network you choose,

you always have to sacrifice the first one and the last one.

The first one is your network ID or network name,

and your last one is your broadcast.

Now, anytime you calculate this number of usable IPs,

it's always going to be the number minus two,

because you have to take away the network name

and the broadcast name.

Let's take a look at classful versus subnetted networks.

A classful network, if you remember,

are the ones like slash eight, slash 16, slash 24.

And if I use the example of 192.168.1.0/24,

this is a classical class C network,

there is one network here because it is two to the zero,

because I borrowed zero bits, because slash 24 is classful.

This means that there is two to the eight

minus two assignable IPs, which gives me 256 IPs,

minus two, one for the broadcast, one for the network,

and that gives me 254 usable IPs.

As you can see here on the screen,

we have all the network bits, all 24 of them,

and therefore, it is classful.

Now, we have eight bits of the host,

and let's say that I wanted to make a smaller subnet.

How can I do that?

Well, I could borrow two bits from the host

and create a subnet,

this would make it a slash 26, for instance.

So, if I borrow those two host bits,

now it becomes two to the s or two to the second power,

which gives me four networks,

four subnets that I've created.

So, if I consider those four networks,

how many IPs can I have in each one

of those four networks I just created?

Well, in the original slash 24,

I had 256 hosts total or IPs total.

Now, I borrowed those two host bits

and gave them to the network.

And so, now I have two to the six power,

that's going to give me 64 IPs per subnet.

So, if I have 64 plus 64 plus 64 plus 64 that equals two 56.

So, I took that one big network, that slash 24,

and I turned it into four smaller networks,

those slash 26s.

But remember, each of those, I have to give away two IPs,

the first one and the last one,

the first one is the network name,

the last one is the broadcast.

So, I only have 62 usable IPs that I can assign

to a server or a laptop or a phone or something like that.

And so, for each of these four networks I've created,

I now have 62 usable IPs and I have four of those

that were made up out of that original slash 24 network.

Now, if you look at the blue at the bottom of this,

you can see my original 256 possible IPs,

which does include my broadcast and my network.

If I submit that down, I'm going to have four subnets

or two to the second power.

Now, I'm going to have 62 possible IPs,

because each one was two to the sixth minus two,

because I take away the first and the last

for the network name and the broadcast.

In our case, this is going to give us the four subnets

as their name of 192.168.1.0, 192.168.1.64,

192.168.1.128, and 192.168.1.192,

those are my four network names.

Now, for the broadcast,

it's going to be the last IP of each subnet,

so, that's going to be the ones ending in dot 63,

dot 27, dot 191 and dot 255.

So, hopefully now you're starting to see how these subnets

start to work together.

If we wanted to calculate those IPs,

we're going to do that based on those subnet masks.

So, when we take those 32 bits,

which is the total number of bits

minus whatever our slash or CIDR notation is,

in this case 26,

we get that there are six host bits remaining,

that would be two to the sixth minus two,

which is 64 minus two or 62 usable, assignable IPs.

Now, I'm hoping that all of this

is starting to come together as we've gone through it.

Subnetting is a very complex topic

that a lot of students struggle with.

I want you to keep practicing it

and keep working through the problems,

because I guarantee on test day, you're going

to get some questions on subnetting.

Now, some of these questions might ask you

just to do the problem,

but most of them are going to be more about troubleshooting.

There's going to be some computers that aren't talking

to another, and when you start looking at their IP addresses

and their subnet masks, you're going to notice

they're not on the same subnet,

and if there's no router there,

data won't go from one subnet to another,

and so, this is an important concept.

Now, let's talk about how we list out these subnets.

Let's keep going with our example of the 192.168.1.0/26

and those four subnets we created of 62 usable IPs each.

Now, where does each network begin and end?

We kind of already went over this by saying

it was the dot zero, the dot 64, the dot 128, and the dot 192

as our beginning of each network.

Now, where did I come up with those?

Well, I started with the original IP, of 192.168.1.0.

Now, zero is my first IP

and it is going to be the name of that network.

Each subnet we said was 64 IPs in length,

so, if I take zero plus 64, I get dot 64,

if I add another 64, I get dot 128,

if I had another 64, I get dot 192.

Now, how do I find the last IP for each of these?

Well, that's going to be my broadcast address.

All I need to do is add dot 63 to each of these numbers,

because there's 64 IPs, the first one is used for the name,

the last one, the plus 63 gets me to that broadcast ID,

and so, I get 63, 127, 191 and 255,

that's going to be the beginning and ending of each subnet.

It goes from zero to 63, 64 to 127, 128 to 191,

and 192 to 255, and everything in between those,

for instance, I might have something like 32

is going to be a usable IP.

Those are things we're going to assign to clients

and servers and tablets and laptops

and desktops and all those things.

Now, when we do all of these,

we're going to write them out

with what's known as the slash notation.

So, we'd write 192.168.1.0/192.168.1.127,

that's pretty long to write that out

for each of the four subnets, so instead,

we have this shorthand notation called a CIDR notation,

and this stands for the Classless Inter-Domain Routing.

Instead of advertising multiple individual routes,

we summarize all of those as a single route.

We do this through CIDR notation, and in this case,

we had the slash 26.

So, I can consolidate all those underneath them

by doing a summary of the continuous networks

using route aggregation.

When I do this, and I look at the slash 26,

I noticed that the first 26 bits are all equivalent,

they're all the same for the subnet mask.

So, by consolidating those,

this allows me to put all those networks together

in one slash notation and simply write them down this way.

Next, we have a Variable-Length Subnet Mask or VLSM,

this allows subnets of various sizes to be used.

Everything I've done so far has been equal.

I took a big chunk of 256 IPs and broke them up

into four equal 64-bit chunks, but we don't have to do that,

I could actually break it up and have a 16 chunk

and a 32 chunk and a 64 chunk, and that way, I can do that

to what's right for my network.

Now, this is going to require a routing protocol

that supports this, but all modern routing protocols do,

including things like RIP, OSPF, IS-IS, EIGRP, and BGP.

Now, we haven't talked much about routing protocols yet,

but we will in the next section of this course

when we talk about routing.

Basically, Variable-Length Subnet Masking

is a subnetting of subnets.

Without VLSM, all subnets would have to be the same size

as you see here on the screen,

but by using VLSM,

I can break them up into whatever chunks I want.

And I'm going to give you the key to subnetting here.

When you look at this chart,

this is your subnetting exam tip.

Write this down in your notes.

Now, the one I want you to memorize here,

is this small chart off to the right.

You can do all the math on test day

and spend a lot of time trying to do two to the s,

and two to h minus two,

or you can memorize this small chart on the right.

If you notice, it's going to start with slash 24,

and you have that as your subnet as a classful class C,

you have 256 IPs, you know this,

because we've talked about this a lot, right?

Now, if I add one to the CIDR notation, it becomes slash 25,

I increase the number of networks,

by adding one, I borrowed a bit,

that's going to give me two subnets,

and that's going to take my 256 IPs

and divide it by two, giving me 128.

Notice, I'm always going to have 256 IPs

as I go from slash 24 downward all the way down to slash 30.

So, every time my CIDR notation goes up by one,

my number of networks goes up by one

and my number of IPs gets divided by two.

If you can remember this chart on the right,

this is going to help you on test day,

get through a lot of the questions on subnetting

and it's really nice and easy.

Now, the good thing is for Network+,

they're not going to give you a lot of difficult subnets.

Almost all of the ones you're going to get

are going to be slashed 24 or higher,

the most are going to give you something like a slash 23 maybe,

or a splash 22, but that's rare.

Generally it's going to be slashed 25, 26, 27, 28, 29, and 30,

that is the most common.

So, if they ask you something like, you have a slash 28,

how many subnets are created?

It's pretty easy, you look at this chart and go,

"Oh, the answer is 16."

And if I go from 24 up to 28, I keep doubling it

and I double it to get to 25,

I double it again to get to 26,

double it again to get to 27,

double it again to get to 16, right?

And that is where I get to slash 28 is 16,

because I went two, four, eight, 16, and there we go.

Now, you might ask, how many IPs you'd have in each subnet

if you had a slash 30?

Well, for a slash 30,

that's what we use for point to point connections,

and there's going to be four IPs,

but only two of those are usable,

because again, our first one is always the name

of the network and the last one is always the broadcast.

So, you have four IPs, two of which are usable.

And if you remember this chart, you're going to be able

to do great on subnetting on the exam.

Now, before the test,

I want you to take lots of practice subnetting problems,

especially in your class C range.

Everything from slash 24 to slash 30

is really fair game for the Network+ exam.

Memorizing these slash notations and the CIDR

is going to help you answer any question they throw at you.

But as we go through this section,

I'm going to show you a shortcut

that you can do just using your hands, as well,

and you're going to really like this,

but having this chart memorized

is going to help with that, as well.

Now, for the exam, something you may see is they might

have you do a variable-length subnetting problem.

Now, if they do a VLSM problem for you,

it might be they gave you an IP address with 256 IPs

because they're going to be a slash 24

and then, they might say,

"Hey, you have an it department that needs X number of IPs,

you have HR department that needs X amount of IPs,

you have a sales department that needs Y amount of IPs."

And you'll have to go through and do the slash notations

for each one based on the number of IPs they give you.

That is a totally fair question on the exam

and how you'd put this subnetting stuff to work

in the real world.

In the last lesson,

we covered a lot of details about subnets.

Now, what I want to do in this lesson

is take some of these problems and go through them,

because you may experience them come test day,

sort of walk through how to tackle them.

Now, before we get to the first problem,

the first thing I want to do is put my chart back

on the screen.

Now, why am I doing this?

Well, because I already told you,

you should commit this chart to memory.

And when you get down and you sit for the exam,

the first thing they're going to do

is give you a piece of paper and a pencil.

And once they start the clock,

you can write whatever you want on that dump sheet.

And so, one of the things I put down on my dump sheet

or my note sheet, is that I put down from my memory,

this chart, and that way,

as I get subnetting questions throughout the exam,

I can go back and look at those,

and so, it makes it really easy.

So, I'm going to give it to you here,

because I'm assuming you've memorized this before the exam,

and you've already written down

as you're sitting at your computer.

All right, let's take our first question.

Let's pretend you're a network administrator

for diontraining.com.

We decided we're going to open up a small branch office

in another city.

And so, to support that,

I need a subnet for a private IP address

that's given to you, for several different networks,

based on the different sizes of the departments

in this new location.

Now, the new office

has been given the range of 10.10.10.0/24,

which tells you, you have 256 total IP's to work with.

Now, when you set up your new network,

I want you to configure separate subnets

for each of the departments that I'm going to give you.

And you're going to tell me which CIDR notation

is going to be the summary for each one in each place.

Now, I want you to resist the urge

of going big for each subnet.

I want you to tell me the minimum subnet

that would support my requirements.

Now, in my new company,

we're going to have this number of users that we have.

So, first, I have 54 new people in IT,

I have 32 new instructors, I have five people in sales,

three in administration,

and then there's going to be an unused portion that's leftover,

and you're going to give me a subnet for that as well.

Now, each of these five areas,

I need you to tell me what the CIDR notation is for it,

and that's how we're going to answer this question,

so, first, we're going to take it step by step.

Again, you've memorized your table,

so this is going to be a pretty simple problem,

because we don't have to do the two to the S,

or the two to the H minus two calculations,

because we already have all that done in our table here.

Now, we're going to round up our department numbers

into multiples of two, or factors of two.

So, it's either going to be a one, a two, a four, an eight,

a 16, a 32,

64, or 128.

Now, remember for each of these,

I told you how many clients needed to be on the network.

For example, I say I had 54 people that work in IT,

but, you need to still account for the network address,

and the broadcast address, for each of those networks.

That's always going to be your first and last IP.

A lot of students forget that,

and they mess up the problems that way.

So, if I had 54, what do I need?

I really need 56.

Now, if I move that up

to the next closest range of two to the something power,

I get up to 64.

So, if I look at IT,

I'm going to end up having 64 IP's for that subnet,

and we have to figure out the slash notation for it.

If I look at the instructors, we had 32 instructors,

and we had to get two more, which puts us at 34,

which, rounding up, would get me back up to 64,

the same as IT.

Notice that trick there,

I would expect to see something like that on the exam,

where they put, it's something,

where they tell you I need 16, or 32,

because that looks like you'd just be able

to give them that slash notation,

but if you forget that first and last,

you're going to mess that up, so don't do that, right?

Make sure you pay attention.

Now, the next one we had is sales,

and for sales, there was five people,

so you add two to that, and that gives me seven,

and now, I look in and go,

"Oh, seven is close to eight,

so that's going to be eight people."

Next, we're going to look at admin or administration.

Administration has three people,

plus two more gets me to five,

round that up, that gets me to eight as well.

And so, that's the way we do this, right?

Now, lastly, I have to calculate the unused amount.

Now, this is going to be a little bit different,

because instead of rounding up,

we're going to have to actually round down.

So, first, we're going to start

with the total number of IP's that we had,

and we were given a slash 24 subnet,

so that means we had 256 total IP's.

Then, I'm going to take away the 64 I had for IT,

and, the 64 I had for instructors,

then I'm going to take away the eight I had for sales,

and the eight I had for administration.

And what I get left is 112 left,

but, I can't round up here,

because I'm talking about what was left.

So, I have to round down to the closest power of two,

which, in our case, is going to be a 64 again.

So, while this problem originally looked really tough,

it really isn't,

because we only have to calculate two things.

We need to figure out what's the CIDR notation for 64,

because that's what we're going to use for IT,

the instructors, and unused.

And then, we had sales and administration

that each had eight,

so I need to figure out the CIDR notation for eight.

And so, if I figure those two out, it's pretty easy,

and I have my chart,

so if I look there for 64, that's a slash 26.

So, if I have slash 26,

that's going to give me 64 for the instructors, 64 for IT,

and 64 for the unused.

Then, I have sales, which had eight,

and that's going to be a slash 29.

And administration again is eight, so that's a slash 29.

As you can see, this really complex problem

became really easy, as we broke it down step-by-step.

All right, for our next question,

we're going to make it a little bit simpler.

The last one may have been something

you might see in more or less of a simulation or a PDQ.

We have to calculate lots of different things,

and then pick the right ones.

But, during the test itself, it's going to be multiple choice.

So, you're going to get a problem

that looks something more like this.

Let's say, how many assignable IP addresses exist

in the network of 172.16.1.0/27?

Now, before we answer this, let me ask you this.

Does it even matter what the IP address is?

No, it really doesn't.

The only thing that matters here is that this is a slash 27.

The IP address itself is just there to mess with you,

so, you can ignore it in this particular problem.

The question becomes how many assignable IP's

are there in a network that uses a slash 27?

Well, I'm going to have to think about that here for a second.

Oh, wait, I really don't have to think about too hard,

'cause I have my chart, right?

Is it 30, 32, 14, or 64?

Well, looking at my chart,

I know it's either going to be 32, or it's going to be 30,

because a slash 27 shows me there's 32 IP's.

Now, which one is it?

The keyword here is assignable.

The question said how many assignable IP's.

Assignable IP means that I can give it to a client.

I can assign it to a computer.

I can't give away my network ID,

and I can't give away my broadcast ID.

So, out of those 32 in the subnet,

I only have 30 that are assignable.

And so, my answer here is 30.

Now, even if you couldn't remember that, using your chart,

you'd still get it down to either of those two options,

30 or 32, and you have a 50/50 shot of guessing right.

So, even if you couldn't remember,

the fact that you have the network and broadcast ID,

which I really hope you do,

because I've been beating that into your head

a couple of times now,

you still would at least get a 50/50 shot on this.

Now, let's assume you forgot to write down your chart.

How could you calculate this using your subnetting formulas?

Well, we would do two to the H minus two.

And that's going to give you your number of usable hosts.

Since we had slash 27 as our CIDR notation,

it becomes,

32 minus 27, giving me the number of host bits,

which is five.

So, two to the number of host bits, two to the fifth,

is going to give me 32 minus two, gets me back down to 30.

Now, whether you want to do it from the formula,

which is a lot more work, or memorizing my table,

either way will get you to the right answer

of 30 usable IP's, or 30 assignable IP's.

Now, just remember,

are they asking about assignable or usable IP's?

Or, are they asking about total IP's?

Depending on how the wording is in the question,

that's going to determine, if you're going to minus two or not.

If they said, "how many total IP's exist in this network?",

the answer would have been 32,

but because they use the word assignable,

you have to take into account the network ID,

and the broadcast ID.

And this is going to make sure

that you get to the right answer.

Now, let's go ahead and look at our third practice problem.

How many assignable IP's exist in the network?

If I have a network of 192.168.1.0/28,

this is just like the last problem.

If it's slash 28, we're going to look at our chart,

and come up with 16, but they said assignable,

so we take off two,

one for the network, one for the broadcast,

and this gives us 14.

And so, our answer is 16 minus two, which gives us 14.

In this video, I'm going to show you

how to do subnetting by hand.

This way, we can eliminate all of the math

and all of the pen and paper

and be able to get the answers we need really quickly.

It's really just a shortcut.

In the last couple of videos,

I showed you how everything worked

so that you understand the theory behind it all,

but now, we're going to get to the quick way of doing it.

So, when I subnet by hand to do this demonstration,

I'm going to wear my subnetting gloves.

Now, my subnetting gloves are special

because I have little numbers

written on the backside of them

that I'm going to show you here.

You may see here on my subnetting gloves

that I have numbers written on them.

Now, in the real world, you're not going to have your gloves on

but you can memorize these numbers fairly easily.

They're just multiples of two.

Now if I'm going to start on my right,

which on your screen is going to look like the left,

you'll see I have two, four, eight, 16,

32, 64, 128, and 256 written down.

This tells me how many networks there are.

This goes back to our two to the nth power,

which was our formula that we used when we did this by hand.

So, if I asked you if you had a /26 network,

how many subnets can be put into a /26

if you were given a /24 to start with?

Well, if I have a /24 to start with,

I'm going to start counting up from there.

So, my first finger is going to be /25 and then /26.

Since I asked for /26, you can see the number is four.

Now, we can do it the other way

and we can go from my left to right.

In your case,

it's going to be your right side of the screen to your left.

And this is going to tell me how many hosts or IPs I can have

inside of a network.

So, if I start from the other side

and I start with a /24, I have 256.

/25, I have 128.

If I have a /26, that's 64.

/27, that's 32.

/28, that's 16.

/29, that's eight.

/30, that's four.

/31, that's two.

So, you can see how this works

going from right to left or left to right

to be able to figure out how many networks

or how many IPs are available.

Now, we're going to do a couple of examples using this method.

The first one I'm going to use is this one.

It's 192.168.0.85.

Now, the numbers really don't matter.

It's really what's important is that /29 at the end.

So, if you want to figure out

how many subnets that were in a /29,

how would you do that?

Well, we're going to count up from right to left,

upwards in the multiples of two.

So, if I start counting on my fingers,

I go 25, 26, 27, 28, 29,

and what's on my 29?

It's 32.

There are 32 subnets if I'm using a /29 network

within a normal class C network.

Now, if I'm going to look at this from an IP perspective,

I'm going to do it the other way.

How many IPs are there in a /29?

Well, we're going to start with /24, 256.

/25, 26, 27, 28, 29,

and there, you'll see I have eight IPs.

Now, if you remember from our lessons,

you always have a broadcast

and you always have a network name.

So, really, I have six usable IPs for host.

I have the first IP which is my network

and my last IP which is my broadcast,

giving us a total of eight.

Again, that's at /29, that's eight.

The next problem we have is 192.168.1.25 /28.

So, from doing /28, how many subnets and how many IPs?

Well, let's pull out our gloves again and we'll start.

We have /25, /26, /27, /28,

and the /28 has 16 on it.

Now, if I go the other way for how many IPs,

I'm going to go /24, /25, /26, /27, /28.

It's also 16.

So, we have 16 subnets and 16 available IPs.

You could see how this works.

It makes it really easy.

So, I think you're getting the idea here

but let's try two more.

This next one has a /30.

So, how many subnets and how many IPs?

Well, let's pull out our gloves again

and so, we have /25, /26, /27, 28, 29, and 30,

and on 30, you see 64 because we counted up,

two, four, eight, 16, 32, 64.

And so, by doing that,

we now know we have 64 subnets.

That's what a /30 is.

I could take 256 IPs in a class C

and I could break it up 64 different times.

Now, how many IPs is that going to be?

Well, that's 256 divided by 64

but if we use our gloves, we can go the other way,

24, 25, 26, 27, 28, 29, 30, and it's four, four IPs.

Because we have four IPs and 64 subnets,

that's going to give us 256 total.

Let's try another one.

This one's going to be a /31.

Now, a /31, that's not even on the table

you told me to memorize, Jason.

That's right.

That's because it's a nonstandard one

but Cisco does support it.

It's usually used for point-to-point connections.

So, if I'm going to connect one router to another router

and I don't need to have a network

and a broadcast in that case with Cisco devices,

I can use a /31 so I'll only use two IPs

instead of the four that we had in the last /30 example

which is normally used for point-to-point.

But again, if we pull out our hands,

we could start going

and figuring out how many subnets and how many IPs.

So, let's do that.

/25, 26, 27, 28, 29, 30, 31,

so, that tells me it's on my seventh finger.

So, if I didn't have my gloves on,

I would just do powers of two.

Two, four, eight, 16, 32, 64, 128,

and so, that seventh finger is 128.

There's 128 subnets.

Now, I'm sure you can do the math here.

If there's 256 total IPs in a class C

and we just took up 128 subnets,

that means, there's going to be two IPs per subnet.

So, let's do it with our hands anyway

just because that's what we're doing in this video

and so, here we go.

We're going to go 24, 25, 26, 27, 28, 29, 30, and 31,

and on 31, we have a two, right?

We can do that if we went and counted up or counted down.

256, 128, 64, 32, 16, eight, four, and two,

and that gets us to that same number of two.

So, it's all about memorizing these multiples of two

as we go through this.

Again, a /31 is something that's Cisco supported.

Some of the other routers are starting to support it

but it is something that is mainly a Cisco thing.

Let's go ahead and do one last one here

and this one is going to be a /32.

So, when we're doing a /32,

you may go do I even need to use my hands?

Well, you really don't

because a /32 just denotes a single host.

If I have your IP address and put /32,

it means it's only that IP

and so, that really is just one IP

and there can be 256 subnets.

But if you wanted to use your hands,

you could do that, right?

Because we have 25, 26, 27, 28, 29, 30, 31, and 32,

and you see on 32 it's 256.

I think you get the idea here.

So, this is how you can do this

in a very quick and easy manner.

Now, before we finish this video,

I do want to put this to practice

to some of the things you're going to see on the Network+ exam

because you're likely not going to get a question

that just says how many subnets or how many IPs.

It's going to be a little bit more complicated than that.

So, let's go to the next question

and you're going to be able to see what we have.

So, in this case, I have 171.129.67.160 /25

and the question I have for you is,

what's the network IP?

What's the first host?

What's the last host?

And what's the broadcast?

So, let's pull out our gloves and take a look at this.

So, if I pull out my gloves and I go for the first one,

a /25 is two subnets.

Now, with two subnets,

that means if I'm using 256 IPs,

remember, I told you on the Network+ exam,

we're always going to assume that we're doing this

in terms of a /24 or greater

because that's usually what you're going to see on the exam.

So, in this case, we're going to start with /24, is 256 IPs.

Because this is a /25, I'm going to have two subnets.

So, I take that 256 and I break it in half, right?

So, if I have two subnets, what's the first subnet going to be?

Well, it's going to start at .0.

What's the second subnet going to be?

It's going to start at 128 because we have 128 IPs.

So, the first one goes from zero to 127

and the second one is going to go from 128 to 255,

which gives us our network ID.

Our network ID is 171.129.67.128

and our broadcast ID is going to be .255.

So, what's our first host and our last host?

Well, it's going to be .29

and our last host is going to be .254

because our first host is always

one above what the network was

and our last host is always

one below what the broadcast was.

That's how this works

when you're just doing it with your hands.

It makes it very quick and easy.

Let's try another one.

The next one we have is 56.187.210.21 /28.

So, if I want to take a /28,

we want to ask how many subnets and how many IP addresses

cause that's going to help us

write down what our network and broadcast are

and that helps us figure out

our first host and our last host.

So, if we have a /28, we'll pull out our gloves

and we'll go 25, 26, 27, 28,

and 28 is my pinky which is a 16 so there are 16 subnets.

Now, if I go the other way, how many IPs are there?

Well, 24, 25, 26, 27, 28.

There's also 16 IPs.

So, 16 IPs and 16 subnets.

Now, if you take those 16 IPs,

you're going to go zero through 15, that's the first range,

16 through 31, that's the second range.

If we look at our IP address we were given,

it ended in .21.

Where does that fall?

Well, it falls in that second range from 16 to 31.

So, our network is 56.187.210.16.

Our broadcast is 56.187.210.31.

And then, we take our last host,

which is one less than the broadcast .30,

and our first host, one more than the network .17.

That's how we can do this

just using our hands with no pen and paper

and we can get these answers very quickly as we go through.

I hope this subnetting by hand was helpful

and if you didn't get it the first time,

go ahead and rewatch the video again cause once it clicks,

it makes all of your subnetting problems so easy

and so quick to get through

and you'll have no issues on the Network+ exam.

IPv6 addressing.

In this lesson, we're going to introduce the concept

surrounding IPv6 or Internet Protocol version six.

So up to this point, we've really just talked

about IPv4 including how to subnet it,

but one of the problems that we have with IPv4

is that it's limited in its address space.

This is because there are only 32 bits

that make up an IPv4 address,

giving us only 4.2 billion possible address combinations.

Now I know 4.2 billion sounds like a whole lot of IPs,

but when we took out entire portions of things

for things like IP put addresses,

local host addresses, private IPs

and then there was a huge amount of waste

before you even started to use subnetting,

this led to a big issue and we began to start running

out of network addresses inside of IPv4.

This is known as address exhaustion and it is a real thing.

In fact, in November of 2019,

RIPE NCC, the Regional Internet Registry for Europe,

West Asia and the former USSR,

announced they have already exhausted

its entire pool of IPv4 addresses.

Luckily though, the Internet Engineering Task Force or IETF

had already started looking into the future

and they developed IPv6 as a standard

all the way back in 1995

with an RFC that documented their vision for IPv6

which they termed IP Next Generation or IPng.

Now you see IPv6 is actually a huge improvement over IPv4

in terms of the number of addresses available.

Instead of using a 32-bit address like we do in IPv4,

IPv6 going to use 128-bit address.

This is going to give you a much larger address space.

In fact, it's going to give you a possibility

of 340 undecillion IP addresses.

That is enough IP addresses for every man,

woman and child on the planet.

This is two to the 128th power.

In fact, there are many, many IP addresses

for each man, woman and child on the planet

because there are so many IP addresses available.

Now you might be wondering to yourself,

hey, we went from IPv4 to IPv6.

What happened to version five?

Why did we jump straight to version six?

Well version five was created,

but it was never fully adopted as an official protocol

or standard and therefore it never went into production.

Instead, a lot of those concepts that were developed

under version five because it was an experimental protocol,

were then brought into IPv6

when it became an official standard.

So let's talk about the benefits of IPv6.

One of the biggest benefits

is that much larger address space

because of the 128-bit addresses.

In addition to that,

IPv6 also increased the efficiency of our networks

by removing IPv4's broadcast data flow type.

Now IPv6 is also more secure

because there is no packet or datagram fragmentation

within the IPv6 standard.

There's also no maximum transmission units

for discovery within each session

unlike IPv4 which contained an MTU

with a certain size for each packet.

In IPv4, if I sent you a packet that was larger

than your maximum transmission unit size,

it would actually fragment that

and send it over the network.

And then when it reached its destination,

it would be reassembled and read.

This was actually a security risk.

It also required extra processing

and it could actually slow down your networks

because it become a very inefficient way of doing things

in our modern networks

with higher internet connection speeds.

So with IPv6, they decided to do away

with fragmentation completely.

Now in addition to providing all of these new benefits,

the creators of IPv6 were also very smart

and realized that IPv6, to be embraced and accepted fully,

it would have to be backwards compatible with IPv4

and allow both IPv6 and IPv4 to coexist on the same network.

After all, it was already late in 1990s

when IPv6 was being developed and released

and lots of computer networks

were already feel it all over the globe.

So it would not be feasible

for us to simply change over everything in a single day.

Think about it like the current migration

we're going through

from gas powered to electric power vehicles.

This is happening throughout the entire 2020s

and into 2030s.

Now this would be an impossible thing for us to say,

hey, everybody on January 1st 2025,

no one can use gas powered vehicles anymore.

All of them we'll replace with electric vehicles

as of that date.

If a government tried to do that,

they'd probably have a revolution on their hands

because so many people already own gas powered cars

and spent a ton of money and investing into those cars

and infrastructure to support them.

Now for that reason, we're not going to simply replace

all gas powered cars overnight.

Instead, there's going to be this slow transition

from gas power to electric that goes on by 2030

or maybe 2040 as more and more of the newer cars

sold in the world will be sold as electric

and they'll stop selling gas power vehicles.

Well the exact same thing is going on with IPv6.

So IPv6 allows for both IPv4 and IPv6

to co-exist on the same networks

and the equipment that runs these networks

becomes known as dual stack which simply means

they can run both IPv4 protocols and IPv6 protocols

on the same network devices simultaneously.

With dual stack devices, if a client supports IPv6,

the router or switch would prefer to use IPv6

and we'll talk under that method.

Now if a device is not able to support IPv6,

it flips itself back and says,

okay, I'll talk to you using the older IPv4 protocol.

This way I can still support you.

Another method that we use is known as tunneling.

This is where IPv6 is going to be tunneled over an IPv4 device.

This allows your older IPv4 routers

to still carry IPv6 traffic.

IPv6 is essentially going to be tunneled as a mechanism

for encapsulating the IPv6 packets within IPv4 headers

and carrying this IPv6 data over those IPv4 routers

and other infrastructure that already exists.

It does this by creating a point-to-point tunnel

between the source and destination

and then encapsulating that information.

This allows isolated IPv6 clients and servers

to be able to communicate without needing to upgrade

all of the routers and switch infrastructure

that still uses IPv4 that may exist between them.

Now one day we may eventually see IPv4 retired fully,

but so far it hasn't happened and personally,

I'm not holding my breath.

From some articles I've read, predictions are that IPv4

will remain with us until at least 2040

so you're going to have to know how to work with both

IPv4 and IPv6 for the foreseeable future

as a network technician.

Another benefit of IPv6 is that it has a simplified header.

So instead of those 12 fields that we had in IPv4,

we only have five fields in IPv6

making it a slimmed down header

that is a lot more efficient to send over our networks.

So you may be wondering what does an IPv6 header look like.

Well I'm going to show it to you, but please realize

you don't need to memorize this for the exam.

Instead, this is just to show you the different fields

that were in IPv4 which is on top

versus IPv6 which is on bottom.

And now you can see how much more simple

IPv6 really is over IPv4.

Alright, let's get back to some things

that you do need to understand for the exam.

Like what does an IPv6 address actually look like.

Well I already said that it's 128 bits in length

so that means it would have 128 ones or zeros

if we wrote it out in binary

and that seems like a really bad idea to me.

So we're not going to do that.

Now we could use dotted decimal notation

like we did an IPv4,

but that still will be a lot of octets to write out

because we would need 16 octets to represent all 128 bits.

So to solve this problem, the IETF

decided that we should use hexadecimal digits instead.

You see, hexadecimal is base 16

which you may or may not remember from your high school

algebra classes.

Now in hexadecimal, each hexadecimal digit

is actually four bits and this is going to allow us

to represent an IPv6 address

by combining four hexadecimal digits together

to make up what we call a segment.

Now a segment is going to have 16 bits in it.

This is represented by those four hexadecimal digits

and then we're going to add a colon

and then we're going to keep adding segments

until we get up to 128 bits

which is going to take eight segments,

each of those having four hexadecimal digits each.

This gives me a total of 32 hexadecimal digits

which is still pretty long.

Now with 128 bits being represented in an IPv6 address,

this means we will have no more than 32 hexadecimal digits

inside of all these segments.

Now why did I say no more than 32 digits in length

for the total of these eight segments?

Why wouldn't it just be 32 hexadecimal digits

because 32 digits times four bits per digit

would give us 128 bits?

Well this is because IPv6 actually allows us to use

a shorthand to be able to simplify

our very long IPv6 addresses.

Now the rules of shorthand are really important

because you could see exam questions on these.

So if you have four zeros for a segment,

you can actually put one zero there instead

and drop those leading zeros.

For example, let's pretend I have a really long IPv6 address

of 2018:0000:0000:0000:0000:0000:4815:54ae.

Using this simple rule, I can replace all those segments

that have multiple zeros with a single zero.

This would give me 2018:0:0:0:0:0:4815:54ae.

Alright, this reduced my number of hexadecimal digits

down from 32 to just 17 so it's about half the length.

We're getting better, but I'm not going to stop there.

There's another rule I can use

in the world of IPv6 shorthand.

This rule says that if there are multiple segments

that all have zeros in them

and no other hex digits are being represented there,

I can summarize that by using a double colon

and take out all those zeros.

Now this rule is special because you can only do

the double colon thing one time inside an IPv6 address.

So using my double colon rule, I can summarize

2018:0:0:0:0:0:4815:54ae

into removing all those five sets of zeros

and replacing them with a double colon

and getting myself 2018::4815:54ae.

So I went from 32 hexadecimal digits

down to 17 hexadecimal digits.

and now I'm down from 17 digits

all the way down to 12 digits.

Much smaller, much easier to work with.

You can see how this shorthand is really helpful.

So how are you going to recognize an IPv6 address

versus an IPv4 address?

Well the first way is by looking at what IPv4 is.

IPv4 is always going to use the dotted decimal notation

using four octets.

Now IPv6 on the other hand

is going to use colons between its numbers

and it's going to be written in hexadecimal.

Alright, now one of the questions you might see on test day

is to identify an IPv6 address when you see one.

For example, you might get a question like

which of the following is an IPv6 address?

This would be a fair question to ask you.

You're going to get some options like 192.168.1.1

which we know isn't it because that's an IPv4 address.

You're going to get 12:34:56:78:90:AB or 1234::5678:90AB.

So wait a minute, those last two I just said,

they're really similar, aren't they?

Yes, but only one of those is a valid IPv6 address.

Do you know which one it is?

Because most students get confused here.

Now the second option here is actually not an IPv6 address.

Instead it's a MAC address.

Remember, MAC addresses which are a layer two

physical addresses are always going to have

12 hexadecimal digits and separated by colons.

Usually they're going to be written as six groups

of two digits each and each of those are going to be separated

by a single colon.

An IPv6 address on the other hand,

should always be written in segments of four digits each

and they should always have 16 segments

unless you see a double colon.

In this example, we have a double colon

between the first and second segment in our third option.

So this is a good shorthand that we can use

and we identify that it was an IPv6 address

because we removed all the zeros

between the first and second segment inside this address.

So if you count up something that looks like an IPv6 address

and it has 12, exactly 12 hexadecimal digits

separated by single colons

and you don't see a double colon anywhere,

that is a MAC address, not an IPv6 address.

Otherwise if it looks something like this

and includes hexadecimal digits,

it's going to be an IPv6 address on exam day.

For the exam, you just need to be able to recognize

what an IPv6 address looks like

and you should be able to summarize one down

by taking out zeros and consolidating them

using that double colon trick.

If you can do these two things, you're going to be fine

for IPv6 addressing on exam day.

Now when it comes to IPv6 addressing,

there are three different address types that you can use,

unicast addresses, multicast addresses

and anycast addresses.

One of the interesting things about IPv6

that really distinguishes it from IPv4

is that we can assign multiple IPv6 addresses

to a single interface on a client

and these assignments can be a mixture

of any of these three different types,

unicast, multicast and anycast.

So even if you only have one network interface card

on your workstation or laptop,

you could have multiple IPv6 addresses

and different types of IPv6 addresses

assigned to that one card.

Unicast addresses are going to be used to identify

a single interface.

These are broken down into globally-routed unicast addresses

and link-local addresses.

A globally-routed unicast address

is similar to what we have as a public address with IPv4

using unicast A, B and C class addresses.

Now in IPv6, a globally-routed unicast address

is always going to start out with its first segment

containing 2000 to 3999.

Now if you see 2000 to 3999 as your first segment,

this means it's a globally-routed unicast address.

For example, the IPv6 address

of 25A40DB885A3123456788A2E03707334

would be globally-routable as a unicast address

because its first segment contains 25A4

which is between 2000 and 3099.

Now a link-local address on the other hand,

also called a local use address,

is used like a private IP address was in IPv4.

A link-local address in IPv6 can only be used

on a local area network and it always is going to begin

with FE80 as its first segment within an IPv6 address.

Now whenever an IPv6 system starts up,

it's going to actually create a link-local address

for each IPv6 interface on that system

even if a globally-routable address

was already manually configured or obtained

through a configuration protocol like DHCP.

To do this, it's going to use something known as SLAAC,

the Stateless Address Auto Configuration or SLAAC.

With stateless auto configuration,

the host does not need to obtain addresses

or other configuration information

from a centralized server like DHCP.

Instead, it can actually independently assign itself

a link-local address, test the uniqueness

of that link-local address,

assign the link-local address to itself,

contact the router and provide direction to the node

about how to proceed with the auto configuration

and it can even configure the global unicast address

that it wants to use.

We're going to come back to this concept in just a few minutes

as we dive a bit deeper into it as we start to talk

about EUI-64 and the Neighbor Discovery Protocol

since both of these processes

are used with the stateless address auto configuration

protocol known as SLAAC.

Next we have multicast addresses.

Now multicast addresses are used to identify

a group of interfaces so that a packet can be sent

to a multicast address and then be delivered

to all the interfaces within a group.

In IPv6, a multicast address will always contain FF

as the first two digits within the first segment.

If you see FF at the beginning of an IPv6 address,

remember, it's multicast.

The final type of address we have

is known as an anycast address.

Anycast addresses are used to identify a set of interfaces

so that a packet can be sent to any member of a set.

We're going to talk more about how anycast works

when we cover IPv6 data flows in a separate video.

Anycast addresses are actually allocated

from the unicast address space

so there's really no way to determine if an IPv6 address

is unicast or anycast just by looking at the IPv6 address.

Now when you're looking at multicast or link-local,

you have a very easy way to do this,

but you don't have an easy way

to figure out unicast versus anycast.

Alright, let's go back and talk a little bit more

about SLAAC,

the stateless address auto configuration process.

Now as I said, in IPv6,

there is an auto configuration process known as SLAAC

and we use this to discover the current network

that the interface is located on

and then allow it to select its own host ID

based on its MAC address using a process known as EUI-64.

Now this EUI-64 or Extended Unique Identifier process

will allow a host to assign itself a unique 64-bit

IPv6 interface identifier called EUI-64.

Now this EUI-64 format address

is obtained by using the interface's 48-bit MAC address.

The MAC address is first separated into two 24-bit portions.

The first half of the MAC address is going to contain

the OUI or the Organizational Unique Identifier

and the second half is going to contain

the specific network interface card.

Now in between those, we're going to shove a 16-bit

hexadecimal value of FFFE.

This way I can take 24 bits, 16 bits and 24 bits

and put together to get a 64-bit EUI address.

Now this gives you the 64 bits that you're going to need

to identify your interface on that network.

Then the interface will use auto discovery

to determine the network it is on

and add the network portion of the IPv6 address

which is going to be the first 64 bits

inside of our addresses.

Now we're going to put that first 64 bits

to represent the network in front of the 64 bits

from the EUI-64 address we created from our MAC address

to create a unicast globally-routable IPv6 address

that we can now use.

So you can see how all this works together

using that MAC address

to create this globally-routable address.

Now DHCP can also be used within IPv6

if you prefer to use it.

If you do, you're going to have to use

the DHCP version six protocol.

This would allow you to have DHCP

automatically assign things from a DHCP version six server.

But since the auto configuration process with EUI-64

is already built into the IPv6 protocol by default,

you really don't need to use DHCP version six.

But if you do want to use DHCP version six,

you can and it will allow you to assign

what addresses each interface is going to get

instead of allowing them

to use the auto configuration protocol of SLAAC.

Now as I said, IPv6 will choose its own address

based on its MAC address by default.

Then it's going to use this thing known as NDP

or the Neighbor Discovery Protocol

to learn about the other layer two addresses on the network

based on their MAC addresses

and then it will pick its own host ID.

Let's talk a little bit more about NDP,

that Neighbor Discovery Protocol.

NDP is used to perform router solicitation,

router advertisement, neighbor solicitation,

neighbor advertisement and redirection.

Router solicitation is when your client

is going to send a message

to locate the routers on its network

because it has to figure out what the default gateway is

that everyone's using so it can get out of that network

and onto the internet.

Routers can also do advertisements over NDP.

Essentially they're going to say,

hey, I'm the router, I'm over here.

You guys can solicit me and ask me questions.

Neighbor solicitation is where your IPv6

starts going around and saying,

hey, what other nodes are on this network

and they ask their neighbors or other devices.

This allows your interface to try to determine

what link layer addresses or layer two addresses

are out there so it can learn from them

and figure out how to talk to them directly.

Now neighbor advertisement is like router advertisement,

but it happens with your neighbors.

Those clients start saying, hey, hey, I'm over here.

These are all the different services that I offer.

You should come talk to me.

Finally, we have redirection

which is where routers can inform the hosts

that there are better first-hop router options out there

to increase the efficiencies of your network.

For example, if I'm sitting in San Juan, Puerto Rico

and my computer thinks the best first-hop router

is for me to get to New York and then go over to California,

this may not be the most efficient route for me,

but it probably works.

Now another router in Florida might go, hey, you know what?

I'm actually a better first-hop router for you.

I can save you some time. So I'm closer.

Come talk to me and I'll send your information

all the way over to California faster for you

than that guy up in New York can.

This is the idea of redirection.

Essentially the router says I'm better and faster.

Use me instead.

Now for the exam, you don't need to know NDP in depth,

but you should understand that NDP,

this Neighbor Discovery Protocol,

is used in IPv6 and it takes a lot of the functions

from layer two and layer three of the OSI model

for router advertisement and neighbor discovery

and handles them for you.

We first talked about data flows back

in IPv4 earlier in this section.

But they changed a little bit here in IPv6.

Now, there's still three data flows,

but one of them has changed.

We still have unicast, we still have multicast,

but now we've done away with broadcast

and we've introduced something called anycast instead.

Now, unicast works just like it did in IPv4,

but instead of using IPv4 addresses,

we're going to use IPv6 addresses.

Now, in this case,

let's say my server wants to send a message

to both PC1 and PC2.

Well, it's going to have to send out two copies

of the same message, one intended for each host,

because it's going from one server to one host

and then back from one server to the other host.

Notice the only difference here is their IP addresses.

We're using ABCD::1

and all that space in between those two colons

will be filled up with zeros

to get us back to 32 hexadecimal digits.

Next, we have multicast.

And in multicast, just like IPv4, we use multicast groups.

The server is going to address the message

to the multicast group, in this case, FF00::A.

Remember, anything that starts with FF

is a multicast address.

And then it's going to get to the switch.

And when it gets to the switch,

it distributes the additional copies

to each of the recipients

in the multicast group number based on this number one

that we're using.

Data is going to travel from the single source,

our server, to multiple but specific destination devices.

The third type of data flow we have is known as anycast.

And this is the different one from IPv4.

This is the one that's unique to IPv6

because we did away with broadcast in IPv4.

Now, anycast is designed to let one host initiate

an efficient updating of router tables

for a group of other hosts.

IPv6 is able to determine which gateway the host

is closest to and send the packets to that gateway

as though it was a unicast communication.

The gateway is then going to broadcast that

through the anycast to any other hosts in the group until

all of the routing tables are updated.

Now, what does this really look like?

Well, let's take a look at this diagram on the screen.

Here we have a server and the router there

is gone on the left

is going to send things out through anycast.

It sends it out to the internet

and it's going to go to whatever the closest place is

that has a router.

Now in this case, it's the router

that's at the top of the screen here.

Now, once it gets to that router on top,

it's then going to go to the DNS server

and it will reroute down through the routers onto the bottom

and keep pushing that around,

but the original router on the left side

of the screen only had to send out that message once.

And then from there,

all of the other routers will figure out the best way

to update everybody else's tables.

That's the benefit of using something like anycast,

as opposed to using another way to update your routers.

And this becomes very efficient

and it's one of the benefits of using IPv6.