Network devices
There are three key devices you need to be aware of in order to understand internet-working:-
-HUB
Used as the central connection point within the star topology.
-SWITCH
Used to reduce collision domains within a star network.
-ROUTER
Used to link network segments together

HUB
All devices on a star network plug into the hub.
Hence it forms the central connection point of the whole network.
If a hub receives a signal on a port it will propagate that signal to the whole network.
This means that collisions can result.
The more collisions on a network the slower the network will run.

SWITCHES
Switches are sophisticated hubs.
They learn what machines are connected to each port.
They only propagate the signal on the port the device is connected to.
Switches reduce collision domains.

How do switches learn?

Using Switches
Switches are too expensive for most networks to be entirely switched.
This means they get used in combination with hubs.
The idea behind this is that of a compromise. In a completely switched LAN all computers would be on their own collision domain.
When used in combination with hubs we have more pcs in the collision domain but less than if we simply had hubs.
Broadcasts are sent to the whole LAN.

A switched network

Routers
Routers are used in two instances
-If you want to reduce the broadcast domains on your network
-If you want to link LANs together via some kind of telephone network
Unlike switches, routers have to be programmed before they can function.
-The basic configuration includes things like IP addresses, routes to remote networks etc.
This will typically be done via a direct cable connection or across the network via telnet.

How do routers work?
The aim of a router is to link separate networks together.
It is only interested in the destination network. The host portion of the address is irrelevant.
If a host computer wants to send to another computer it will work out if the destination is local or remote.
-if the destination is LOCAL it will send directly to that machine
-if the destination is REMOTE it will send to its default gateway.
The default gateway is the local address of the router.
The router then calculates which network the packet is destined for.
It then examines its route tables for the route to that network.
If a route is present it will list the address of the next router (hop) in the route.
The router will then forward to that router.
If the route wasn’t present it will either be discarded or forwarded to the gateway of last resort. (This is the router version of the default gateway)

TCP/IP
Stands for Transport Control Protocol \ Internet Protocol.
Is an example of a routable protocol.
This means that it can be routed between networks.
Its configuration is made up of two parts
-an address
-a subnet mask
The address is made up of four octets.
This is because each section contains eight binary digits.
Typically we write the address in decimal.
-192.36.27.1
The address contains the network address and the address of the machine.
We therefore have to tell the system how the address is broken down.
We do this by specifying the subnet mask.
There are three main subnet masks. – one for each class of network
The sections that are 255 tell the systems that the corresponding octet in the address is part of the network address.
-192.36.27.1 mask 255.255.255.0
-Therefore network address is 192.36.27.0
-This is a class c network
A class A network would have a subnet mask at 255.0.0.0
-10.0.0.0
A class B network would have a subnet mask at 255.255.0.0
-128.37.0.0
A class C network would have a subnet mask at 255.255.255.0
-192.36.27.0

You can tell the class of network by looking at the first octet of the address.
-1-127 = Class A (127 is reserved for testing)
-128-191 = Class B
-192-223 = Class C
Although eight bits allows us up to 255 some cannot be used because of how the system works in binary.