Hub, Bridge, Switch, Router - Network Devices - Networking Fundamentals - Lesson 1b

Hello. Welcome to the second part of our

lesson on network devices.

This is the first lesson from the first

module of my new course
on networking fundamentals.

The purpose of this module is to teach you

how data flows through the internet.

In part one of this lesson, we discussed

the concepts of a host,

an IP address, and a network. If you

haven't watched that video, go ahead and

pause this video right now
and watch the first video.

There'll be a link in the description.

In this video, we're simply going to

continue right where we left off.

Now, the main idea we want to teach in

this video are these last two devices:

switches and routers.
But we can't really understand those

until we understand
where we've come from.

So we have to start there.

In the last video, we unpacked the idea

of a network. We identified that a

network is created anytime you connect

two computers to each other using a wire.

One thing to understand about sending

data across a wire is that it decays as

it travels greater and greater distances.

If the two computers you're connecting

are in the same room, then you don't

really have to worry about it.

The decay will still occur, but the

signal will still get through,
and therefore,

connectivity between these
hosts is still attained.

If, however, these hosts span greater

distances, maybe you're
connecting two computers on

opposite sides of a building, or even in

two different buildings,
then you might have a problem.

If the signal decays before
it gets the other side,

then these two hosts cannot share data.

In those cases, what you need is a

repeater. A repeater is a device
whose sole purpose is to

regenerate signals.
Anything that comes in on one end

simply gets regenerated
out the other side.

This allows you to connect devices
together which span greater distances.

So far, we've been talking
about networking

from the perspective of
connecting one host

directly to another host. Well, if you add

a third host, you now have
to connect that host

to all the other hosts which
you've already established.

And if you add a fourth host,

you now have to connect this fourth host

to all the hosts that already exist.

And again, if you add a fifth host,
you now have to connect this fifth host

to every host that
has already been connected.

As you can see, connecting
hosts directly to each other

simply doesn't scale.

Instead, we created devices
which we could put

at the center of every network and connect

all the hosts to those devices.

And these devices would then handle

funneling communication
between these different hosts.

The benefit to these types of devices is

that if a sixth host gets spun up,

it's very easy to simply connect it once

to that device,
and now it has connectivity

to every host that has already existed.

That's what all of these are. And the

first of these types of devices
that we're going to discuss

is known as a hub. A hub

is nothing more than
a multi-port repeater.

Earlier, we discussed repeaters, and we

said all they do is regenerate signals.

Hubs do the same thing, except they do it

across multiple ports.

For example, if these two hosts over here

need to communicate,

one of them sends a packet to the other,

it'll hit the hub, and the hub will

simply duplicate that packet and send it

out all remaining ports.

That'll allow what this guy
sends to arrive over here.

This fixes the scale problem. A hub is the

first device that allows us to connect

multiple devices in the center, and now

all of them have connectivity to each other.

But, as you can probably see, the problem

with the hub is that everybody receives

everybody else's data.

These two hosts over here, which are

uninvolved in the communication
between these two hosts,

are receiving a copy of
everything they send.

Which brings us to bridges.

Here, we have two sets of hosts, all

interconnected using a hub.

And a bridge is meant to sit in between

hub-connected hosts.

Bridges, by definition, only have two

ports: one port facing one set of

hub-connected devices and another port

facing the other set of
hub-connected devices.

Bridges will also then learn
which hosts are on which

side of the bridge. This would allow the

bridge to contain communication

to only the side that is necessary.
For example,

if these hosts again need
to speak to each other,

when they send data to each other

through that hub, the hub is of course

going to simply regenerate that signal

at all ports

and notice that the bridge can be

getting a copy of that packet.

But the bridge knows that the other

green host is on this side of the bridge,

and therefore, the bridge isn't going to

bring that packet to the other side.

The bridge is the first type of device

that helps contain packets
only to their relative networks.

On the other side, if these hosts
need to speak to each other,

they can also send packets to each other

through their hub, and once again, the

bridge will not let those packets bleed

into the other side, because it knows

the yellow devices exist on the right hub.

And of course, if this device needs to

send something to this device, the bridge

is going to know that that traffic

is going to have to cross the bridge,

and the bridge will allow that packet to

traverse to the other side.

The main takeaway is understanding that

bridges can learn
which hosts are connected

on either side of the two
ports of the bridge.

Now this finally brings
us to switches.

Switches are sort of like a combination

of hubs and bridges. They are like hubs

in the sense that many devices can

connect to the switch, and they are like

bridges in the sense
that they can learn which hosts

are connected to each port.

The main difference is that they're

doing it on a per-port basis,

which means if these two
hosts want to speak to

each other, the switch
will know that the only ports

that need to receive this traffic

are the two that are connected to those

green hosts, and will keep that communication

contained to just those ports.
Moreover, if these

two hosts want to speak to each other,

the switch will again make sure that

that communication only flows
between the relative ports.

So this is how a switch is like a

combination of a hub and a bridge.

The formal definition of a switch that

we want to use is that
a switch is a device

which facilitates
communication within a network.

Earlier, we defined a network as a

logical grouping of hosts which require

similar connectivity. Which means

all of these devices over here all

belong to the same network.

Moreover, networks all share the same

IP address space,

which means this network owns all the IP

addresses which start with

192.168.1.anything,

and this host's identity is the IP

address 192.168.1.33

And this host would be 192.168.1.66.

And this set of devices could
very easily represent

all the different hosts on
your home wi-fi network.

Maybe this device is your printer, and

this device is your laptop,
and this device is your mobile phone,

and so on.

Or maybe this network and these devices

represent all the PCs that might exist

within a particular classroom

of the school network. Or maybe, even

further, all these devices represent

hosts that exist in the sales team

of the London office of the
ACME corporation.

One way or another, since all these

devices are connected with a switch,

they all belong to the same network.

Now, let's go back to that
example of the school network.

we said that the school
likely has many different classrooms,

and each of those classrooms
belong to their own network,

which means this would be a more

accurate representation of the school

network. We would have Classroom Two

owning that IP space, and Classroom Three

owning that IP space.

Now, the reason you might want to

separate these two sets of devices into

their own network is
because they might have

different connectivity requirements.

For example, maybe
these computers over here

all belong to the biology classroom and

all they need is simple
internet connectivity,

but maybe these computers over here

belong to the computer science classroom,

and they not only need internet connectivity,

but also access to various cloud resources

to do their studies. Well, since these

computers have different
connectivity requirements

than these computers, it's a good idea to

separate those out into separate networks.

Now, in both cases, we can still use

switches to facilitate

all the communication within the networks,

meaning this switch can handle all the

communication between these three

hosts, and this switch can
handle all the communication

between these three hosts.

But what happens if this host down here

wants to speak to this host
on a different network?

Well, if a switch can only facilitate

communication within a network,

we would need another type of

device to handle the communication

between networks, and that device would be

a router. A router is a device whose

primary purpose is to
facilitate communication

between networks. At the very least,

you're going to need that router to

connect you with the
ultimate network of networks

known as the internet.

So let's unpack this further.

Routers provide traffic
control points between

networks. Let's say we wanted to limit

the traffic that could go
from this PC to this PC.

Well, since these two PCs
aren't separate networks,

all that traffic has to flow through the

router, creating a great place

to add security policies or traffic

filtering, or even redirecting that

traffic elsewhere entirely.

Since routers sit on the boundary between

networks, they provide a logical
location to apply security policies.

This type of security filtering isn't

traditionally available on switches.

These days, there are modern switches

that can do such filtering, but it is

generally accepted that
the devices sitting

on the same network don't
typically need filtering for

traffic traveling within the network.

If you had devices that needed different

types of connectivity, you'd want to

place them in different networks.

The network boundary is
what is meant to be

the logical separation of devices.

The way routers work is that they learn

which networks that they are

attached to. Meaning, this router is going

to learn that on this interface, it's

connected to the 172.16.20 network

And on this interface, it's
connected to the 172.16.30 network.

And out here is the direction
to go to the internet.

The knowledge of each of these different

networks is known as a route,

and all these routes are stored in what

the router calls a routing table.

A routing table is therefore

all the networks that a router knows

about, and the router is going to use

this routing table in
order to funnel traffic

out the appropriate interface.

Now, when we say a router learns which

networks they are attached to,

what we mean is that
a router has an IP address

in every network that they're attached to.

For example, when this
router is attached to this

network, it is given an
IP adress in that network.

This interface's identity
is the IP address 172.16.20.1,

and this interface's identity
is the IP address

172.16.30.254.

This IP address is going to serve as

what's known as a gateway.

A gateway is a host's way
out of their local network.

For example, this host over here has the

IP address 172.16.20.33

But if that host wants to
speak to something on a

different network, it knows
it's going to have to go

through a router, and the IP address for

that router is stored as
that host's default gateway.

Notice, this host has a
default gateway of 172.16.20.1.

That's this interface
IP address of that router.

Now, if we go a step higher than that,

routers are actually what create

the hierarchy in networks and IP

addresses that we discussed in the prior

sections of this lesson. For example, the

New York office of the ACME corporation

that had all the different teams that

each had their own IP networks?

Well, each of those networks
are connected to different

routers, and each of
those routers are then

connected to another router.

And if a host in the sales team wants to

speak to a host on the marketing team,

it's going to use its gateway, which is

its closest router IP address,

which is then going to send the packet

to the next router, to the next router,

and then finally to the
host on the marketing team.

The Tokyo office of the ACME corporation

is likely going to have a similar setup,

and both of these routers are then

likely going to connect to the internet.

The internet is nothing more than a

bunch of different routers itself.

Meaning, if a host on the marketing team

wants to speak to a host on the

engineering team in Tokyo,

that host will send the data to the

router, which will send
the data to the next router,

which will send it through all
the routers on the internet,

which will finally send it
to the Tokyo router, and

finally to the engineering team.

That is how data is going to
flow across the internet,

and that is the role that routers play

in making that possible.

Now, the last idea I want
to leave you with

actually involves pulling back
the definition of switches as well.

There's something important
you have to understand

about what we've defined
as routers and switches.

Routing is the process
of moving data between

networks. A router,
as we have described it,

is simply a device whose primary purpose

is to perform routing. In the same way,

switching is the process
of moving data within networks.

And a switch, as we have described it, is

a device whose primary purpose

is switching. The reason I bring that up

is there are many other types of network

devices that exist out there.

Access points, firewalls, load balancers,

Layer-3, switches, proxies,

and there's even devices that only exist

in the cloud, like
virtual switches and virtual routers.

One way or another,
all these devices are going to

perform routing, or
switching, or both. So later on

in this module, when we describe

what a router does, or what a switch does,

what we are actually describing is what

any device does that implements routing,

or any device does
that implements switching.

And with that, we close our lesson on

network devices.
In part one of this lesson,

we unpacked hosts,
IP addresses, and networks,

and in part two,

we continued that discussion
by illustrating repeaters,

hubs, bridges, switches, and routers.

In the next lesson, we're going to give

you a practical perspective
on the OSI model.

This will lay the foundation to

understand what all of these devices do

to enable data flowing
through the internet.

But that wraps up this lesson. Your main

takeaways are on the slide right now.

I hope you enjoyed this lesson.
I want to thank you for watching,

and we'll see you in the next one.

Hey, YouTube. I hope you enjoyed that free

lesson for my new course on
networking fundamentals.

I'll be releasing the entire first

module for free here on YouTube.

I want this course to be the ultimate

networking fundamentals course, and since

I'm still scoping out the outline, you

could have a say in what topics will be covered.

Let me know in the comments below what

subjects you want included in this course.

Otherwise, remember to like and subscribe,

and of course, if you learned something

from this video, the best way to thank me

is to share this video.

It's a small act of gratitude, but one I

appreciate greatly.

I hope you enjoyed this lesson. I want to

thank you for watching, and we'll see you

in the next one.