Learn IT with Davo
  • Mr Davidson's Blog
  • Twitter
  • A Level CS - H446
    • A Level Exam Technique
    • Lessons
    • Unit 3 - Coursework Guidance
  • OLD GCSE CS - J276
    • All GCSE Questions
    • GCSE Exam Technique
    • Glossary of Terms
    • Unit 1 Revision >
      • 1.1 - Systems Architecture
      • 1.2 - Memory
      • 1.3 - Storage
      • 1.4 - Wired and Wireless Networks
      • 1.5 - Topologies, Protocols and Layers
      • 1.6 - System Security
      • 1.7 - Systems Software
      • 1.8 - Ethics and Law
    • Unit 2 Revision >
      • 2.1 - Computational Thinking
      • 2.1 - Searching and Sorting Algorithms
      • 2.1 and 2.2 - Writing Algorithms/Programming Techniques
      • 2.2 - SQL and Database Structure
      • 2.3. Robust Code
      • 2.4. Logic
      • 2.5. Translators and Facilities
      • 2.6. Data Representation
  • NEW GCSE CS - J277
    • Glossary of Terms
    • Exam Technique
  • GCSE Business - J204
    • Lessons >
      • Unit 1 - Business Activity, Marketing and People
      • Unit 2 - Operations, Finance and Influences
    • Exam Technique
  • Contact

1.5 - Topologies, Protocols and Layers

In this section (click to jump):
  • Topologies - An overview
  • The Star Topology
  • The Mesh Topology
  • Ethernet and Wifi
  • IP Addresses
  • MAC Addresses
  • Protocols and Layers
  • Packet Switching
 

Topologies

Picture
Look at them. Real men, with real side burns. Grrr. Wouldn't it be awesome to still live in a time where the height of your hat dictated how important you were.
So far we've seen how networks are organised or categorised in two different ways:
  • Geographically - how spread out they are in terms of LAN or WAN.
  • Control - who or what controls the users, data, backup etc on the network in terms of Client - Server or Peer to Peer.

The final piece of our logistical networks puzzle is about how the computers are physically connected together, this is called the topology of the network. 

So our definition is: in networking, the word topology refers to how things are physically laid out, how they are connected and the relationship between devices on a network.

There are two types of topology we are interested in for this exam and they are:
  • Star
  • Mesh

As usual, you'll need to know what they are and some advantages and disadvantages. A top exam tip for these questions is always to remember to discuss network topologies in terms of:
  • Cost - how much cable are we using?
  • Expertise/Employees - how many people would be needed to look after it and how complex is it?
  • Ease of adding new devices - how easy is it to stick a new machine in your network?
 

The Star Topology

Picture
If you've been going through this in order then right about now your bat senses should be tingling and you're going to start to notice some similarities here. This is because a Star topology looks pretty much identical to the diagram we looked at for client-server networks and, er, the reason is that this is usually exactly the topology a client-server network uses.

The star topology has the following features:
  • All devices are connected by a single cable
  • Those cables are all connected to a central point, usually a switch
  • The switch is responsible for managing traffic on the network
  • There may be one or more servers connected to the switch
Picture
Looks like a star, smells like a star.
The star topology has several advantages:
  • It is easy to connect new devices to the network - simply plug a cable in to the switch
  • If a single device fails, it does not affect the rest of the network
  • The switch should manage traffic on the network so that bandwidth is used efficiently
  • It is simple to set up

The disadvantages are:
  • Star networks can use a lot of cable (think each device needs its own cable to the switch which may be far away.)
  • If the central point (the switch) fails, then the whole network will fail.
  • Can require expert staff to configure switches and manage any servers on the network.
 

The Mesh Topology

Picture
A full mesh topology
Picture
A partial mesh topology
Does this look a lot like a peer to peer network? It should do, because peer to peer networks are mesh networks.

A mesh network is where nodes are connected to multiple other nodes in the network - there is no one central point. The connections can be wired, wireless or a mixture of both. There are two types:
  • Full mesh - every device is connected to every other device
  • Partial mesh - each device is connected to one or more other devices in the network.

Each node in a mesh network is responsible for receiving and forwarding packets on the network.

Advantages:
  • Reliability - A fully connected mesh has lots of redundancy, meaning if a connection breaks the network can continue working
  • The burden of managing traffic is split between devices on the network

Disadvantages:
  • Not suitable for large networks - as the network grows the cost becomes prohibitive
  • Uses a lot more cable than other topologies, again increasing cost
 

Ethernet and WiFi

Picture
WiFi, WiFi, where for art thou, WiFi?
Ethernet and WiFi are both standards. This means that there are a set of defined rules and methods for sending and receiving data on a network and if you adhere to these standards, your device will be able to communicate with any other device which also adheres to these standards.

Ethernet is IEEE standard 802.3. Now you know that, you can wow your friends, parents, pet trifle with your new found knowledge.

What does this mean?
Ethernet

Ethernet is:
  • A set of standards
  • Which dictate how computers should be physically connected...
  • ...and exactly how data will travel between those devices.
  • More secure than wireless because it uses physical connections.

Ethernet is responsible for network devices having MAC addresses, for data being split in to packets, error correction and even the type of cables that must be used to connect devices together.

Ethernet has several advantages:
  • It is the standard for network communication and as such virtually all devices that are networked use these standards
  • Speed - it's silly fast - speeds of up to 100 Gigabits can be achieved
  • Security - you have to physically connect to the network to intercept packets
  • Reliability - the use of packets and error correction methods means data can be sent quickly and reliably on the network

Disadvantages:
  • Ethernet can be affected by interference on or around cables which may disrupt data transmission
  • Distance - Ethernet cables have a maximum sensible length of 100m before the signal degrades too much to be useful.

WiFi

WiFi is another set of standards, IEEE 802.11 to be precise and weirdly it stands for Wireless Fidelity. Whatever that means.

So again, WiFi is:
  • A set of standards
  • Which dictate how computers should be connected without wires
  • It includes methods of encrypting and securing data which is sent
  • Also states which frequencies should be used for transmission of data (2.4Ghz and 5Ghz bands)

WiFi Advantages:
  • No need to connect using cables
  • Can connect anywhere with signal - flexibility
  • Easy to connect new devices

WiFi Disadvantages:
  • Signals are badly affected by distance
  • and solid objects
  • and interference from other devices/electronics
  • Security - anyone can intercept wireless data/traffic

A note on transmission bands:
  • 2.4Ghz is generally a more stable signal and can be sent longer distances
  • 5Ghz is generally faster but is more affected by objects, distance etc

Each band is split into a number of channels to try and separate out traffic on the network, especially in areas where there may be multiple wireless networks (think about how many networks you can pick up around your house from your neighbours). However, those channels do overlap so one big problem in wireless networking can be poor performance in areas where there is simply too much activity on each channel/band.

WiFi is inherently less secure than Ethernet. Why? Because you can easily intercept packets of data by just sticking an antenna in the air, whereas with Ethernet you'd need to physically plug yourself in to the network.

As a result, there is a standard for encrypting and securing wireless data called WPA2.

WPA2 Provides:
  • An implementation of the 802.11i standard (you can't get enough 802.11 - look it up its truly fascinating reading.)
  • Per packet encryption
  • 256 bit encryption key

In case you were wondering how secure 256 bit encryption was, if you were to try and guess the key by brute force (trying different combinations until you happened to be successful) then you would potentially have to try...

115792089237316195 42357098500868790785 32699846656405640394 57584007913129639936

... different combinations.
Picture
"Need more input!"
"Read this fascinating book..."
"oooh, all combinations of a 256 bit number. Interesting..."
...different combinations before you were successful. Now, I don't know about you but I don't even know how to say such a large number. What I would be confident in saying is that by the time you and the fastest computer on earth had tried all of those combinations you would almost certainly be dead, the universe would probably have ended too and you would go down in history as one of the least interesting people ever to have lived.

​ "Ah yes, he died watching a  progress bar that hardly ever moved, trying to pointlessly crack a wifi password that he could've just social engineered out of someone. Still, kept him out of trouble. Kids today..."
 

IP Addresses

IP addresses are not scary. Indeed, if you can find your way home at the end of the day then you already have the knowledge required to understand what they are and how they work.

You should already have heard the term IP address when learning about Packets, DNS and how the different types of networking hardware work. If you don't understand packets yet, please go back and learn this before you go any further otherwise this won't be as clear as it should be.

So.... We know that:
  • Devices are connected to networks 
  • Either wired or wireless
  • Data that is sent is split in to packets
  • These packets are delivered to devices by switches...
  • ...or moved closer to their destination by routers
Picture
Postman Pat plays the didgeridoo. This is why no one in Greendale ever gets their post on time.
In exactly the same way that a packet sent in the post needs an address on it, a packet sent through a network needs a destination address so that it can be successfully delivered to the right place.

Each device connected to a network is given an IP (Internet Protocol) address. An IP address looks something like this:

10.154.16.9

Things we need to know:
  • There are always 4 numbers, separated by dots
  • Each number can be anything from 0 to 255
  • These numbers are.... 8 binary bits! Remember the biggest number we can store in binary with 8 bits is 255. 
  • So, therefore, an IP address is 32 bits
  • This means IP addresses are in the range 0.0.0.0 to 255.255.255.255

Every packet sent through a network will have a destination and source IP address written in to the header data of each packet. This is so any router or switch knows where the packet should be delivered to and also where it came from.

Because the internet is a very fluid and dynamic place (devices connected to it change all the time) there are a few things to be aware of:
  • There are approximately 4.2 billion possible IP addresses. This means there are not enough for all the devices that are/could be connected to the internet!
  • IP addresses are dynamically assigned. This means automatically given out by Routers and Switches when you connect to them
  • Because they are dynamic they can change.
  • This means you can't use an IP address to identify an actual piece of hardware - you only know that an IP address is a destination on a network.

So, to summarise:
  • When you connect to a network, a switch or router gives your device an IP address
  • This IP address is not a unique identifier for your device and can change - they are dynamic
  • The IP address is used to deliver packets to a device connected to a network
  • Packets contain a source and destination IP address in their headers
  • IP addresses are 32 bit, written in decimal, separated by dots - e.g. 192.168.0.1 (which will probably take you to your router log in page if you type it in to a browser...)

So now you know!
 

MAC Addresses

PictureDo you know why you can't hear anything?
Because there are no sounds to hear.
The eagle eyed amongst you, when reading about IP addresses, should've asked the question "if IP addresses don't uniquely identify a device, what does?"

More to the point, you should also have asked "...and how do we know one device from another?"

Or "How come if I have a JesusPhone 10xPlusNegative and my friend has the exact same one, how does a network know which is which?!"

And you would have a point.

Before we answer those questions, we need to learn what MAC stands for and what a MAC address actually is.

MAC = Media Access Control

MAC Address = Media Access Control Address.

We're still none the wiser are we?

A MAC address is simply a unique identifier given to any device which is capable of connecting to a network. This literally means anything that has wireless or wired capability has a MAC address. 

A MAC address looks something like this:

A3:F1:00:3B:FF:11

This looks significantly different to an IP address and that's because... it is. A bit...

MAC addresses consist of:
  • 6 "octets" separated by colons ":"
  • Each octet may be any number from 0-255 (like an IP address)
  • Each octet is written in Hexadecimal (which explains all the letters). If you're not sure what that is, look at number systems here.

Some facts about MAC addresses that you might find really useful in the exam:
  • They are totally unique and never change. If you throw away a device, the MAC address dies with it.
  • It is used to uniquely identify each device on a network
  • There are 2^48 possible MAC addresses (281,474,976,710,656 possible MAC addresses) so we are not going to run out of them in the near future.
  • A MAC address is given to a device when it is manufactured - unlike IP addresses which can be assigned every time you join a network.

When a device is connected to a network, it is assigned an IP address so that data can be routed to it through the internet. However, a Router or Switch will then translate this IP address into a MAC address to ensure that packets are delivered to their final destination.

This explains why your home internet can be shared so easily:
  • You have only one internet connection at home
  • Which means you are only given one IP address for your house
  • Your Router uses MAC addresses to ensure that data is directed to the right device when it arrives

And that's all you need to know about that.

 

Protocols and Layers

Picture
​The first recorded use of the word "Hello" being used as a greeting is thought to have occurred around 1833, which is quite weird when you think about it. What did people say to each other before then?  It gets stranger still when you think that instead of hello it could've been any word that we use. Imagine if we lived in a world where every time you answer the phone you say "Wibble!"

I used to go to University with a Greek guy named Lucas. Whenever he answered the phone he would say "Neh! Ella!" Which, he told us at the time, apparently means "Yes, speak." I knew a French girl once who would answer with an elongated "Ouiiiiiii" or "Coucou!" We just accept these conventions because we're used to them, but actually we shouldn't be surprised when we come across a culture or situation where the convention we are used to  is completely different or ignored entirely!

And so it was that when Alexander Graham Bell was inventing the telephone in 1876 he had to decide what people should say when they pick their phone up. His preferred greeting was "Ahoy-hoy!" which has its roots in the nautical world. Inevitably, however, the general population decided they would rather use hello and ahoy-hoy just fizzled out, sadly. Can you imagine how lovely the world would be if we were all ahoy-hoying each other? That'd be great it really would.

Why on earth am I telling you this?

Picture
Richard Ashcroft of The Verve. Musical and lyrical genius.
Without you realising (most of the time) we are constantly following unwritten rules or codes of ethics in our day to day lives. There are expected ways of doing things - for example when you walk down the street you make an effort to go around people rather than just pushing them to one side like you're pretending to be Richard Ashcroft in the Bittersweet Symphony music video (as an aside if you have never seen this nor heard the song then your life is quite incomplete - listen to it now before you do anything else, feel uplifted and then feel ashamed that you've never listened to The Verve before).

Going back to our phone call example from earlier - I hope you noticed that the way we answer the phone and the way it's done in other countries is quite different. This is fine in most circumstances - how often do you find yourself phoning someone from another country? Certainly not often. However, the internet is a very, very different place.

The internet is entirely unique in that it is one of the few examples in human history where we have connected the entire world together in one, single standard way. The internet works in the exact same way whether you're in England, Japan or Brazil. Every single device will communicate in a set, standard way. Even TV isn't this standardised (you can use a TV from the US in England for example).

Even electricity supplies aren't this standardised!

The internet works like this because of something called Protocols - and you've heard of them before without realising it. Everything we do on the internet will use one or more protocols to make it happen.

So what is a protocol?

Definition: A Protocol is a set of rules which establish how communication between two devices should happen.


A protocol will define the rules for:
  • How devices establish a communication link
  • How data should be sent and received (including how it is broken up and re-assembled)
  • How data is routed around a network
  • How devices identify themselves and connect to a network
  • How errors and data corruption should be handled

A protocol may be implemented (carried out) by either hardware, for example a network card, or in software, for example a web browser.
Types of Protocol

Some examples of common protocols are detailed below.

Picture
 An important note about TCP/IP:

TCP/IP is actually a "suite of protocols" and usually referred to as the "TCP/IP Stack" or "TCP/IP Model."

What does that mean in English? It means that it is a protocol which is made up of, er... lots of other protocols. In fact, all of the protocols in the table above "live" inside the TCP/IP Model

TCP/IP is split into layers: 

The TCP/IP Stack

Picture
This stack describes how data is presented, prepared and transmitted through a network (the internet). When you send data, if follows the stack from the top to the bottom and is then sent. When data is received, it goes through the stack from bottom to top to be reconstructed and displayed.

The diagram above makes it fairly obvious what each layer does and is about. For your exam I doubt you need to learn the stack, nor know the contents of each layer. BUT... I would want to know:
  • What each protocol is in each layer
  • A little bit about how each works

For example, if we take "Ethernet" you would want to be able to say:
  • Ethernet is a protocol
  • Within the TCP/IP Stack (obviously!)
  • Which describes how computers are physically connected together
  • It describes how data is transmitted between computers on a network.

Definition of layers:
  • Layers divide the design of a set of related protocols into small, well defined pieces.
  • Each layer adds to the services provided by the lower layers
  • Each layer is  independent of other layers and clearly defines the services it provides. No layer needs to know HOW another layer works, only what to pass to it or expect from it.
The big advantage of layers is that they allow us to make changes in a layer without affecting other layers!

Considering we've gone into the others in such detail in previous  sections, or they're just so, so obvious (HTTP?!) you should be able to describe the others yourself. 

Shouldn't you?
 

Packet Switching

You already know this! Packet switching is the job carried out by Switches and Routers. But what do we mean by packet switching exactly?

When two devices want to talk to each other you have two choices of how to do this in a network. 

Option number 1 is:
  • Find a route from one device to the other and keep this connection exclusively until either device is finished.

You can imagine how this might work on the road network. Say I want to travel from Burntwood to London, I would first find a route I wanted to drive, then ask for all the roads to be closed that I need. I'd then drive to London, having a great time on empty roads. When I arrive, the roads could re-open for someone else to use.

You probably have realised that this is "generally a bad idea", because:
  • Once that connection is made, it cannot be shared until the devices have finished communicating
  • This basically then blocks out one potential path that other data may wish to use
  • There is no sharing of cables, therefore it is an incredibly inefficient way of networking.

Believe it or not, this is how the phone network used to work.

Therefore, Option number 2 is:
  • Split your data in to bits (packets) so you don't need to send lots of continuous data all at once. Then, let devices on the network find a path for packets to take to their destination along shared routes.

Straight away you can tell that this is "generally a much better idea," because:
  • Cables and connections can be shared!
  • We do not block a connection simply by sending data somewhere
  • This is a far more efficient way of networking
  • Packets may travel in any order to reach their destination
  • Packets may travel down many different routes to reach their destination (quick and efficient use of the network)

If you want to know more about Packets, Switches and Routers (go on, you know you want to) then you need to visit 1.4 - Wired and Wireless Networks. It's all in there! Go on, click it, I've even made a link for you...
Powered by Create your own unique website with customizable templates.
  • Mr Davidson's Blog
  • Twitter
  • A Level CS - H446
    • A Level Exam Technique
    • Lessons
    • Unit 3 - Coursework Guidance
  • OLD GCSE CS - J276
    • All GCSE Questions
    • GCSE Exam Technique
    • Glossary of Terms
    • Unit 1 Revision >
      • 1.1 - Systems Architecture
      • 1.2 - Memory
      • 1.3 - Storage
      • 1.4 - Wired and Wireless Networks
      • 1.5 - Topologies, Protocols and Layers
      • 1.6 - System Security
      • 1.7 - Systems Software
      • 1.8 - Ethics and Law
    • Unit 2 Revision >
      • 2.1 - Computational Thinking
      • 2.1 - Searching and Sorting Algorithms
      • 2.1 and 2.2 - Writing Algorithms/Programming Techniques
      • 2.2 - SQL and Database Structure
      • 2.3. Robust Code
      • 2.4. Logic
      • 2.5. Translators and Facilities
      • 2.6. Data Representation
  • NEW GCSE CS - J277
    • Glossary of Terms
    • Exam Technique
  • GCSE Business - J204
    • Lessons >
      • Unit 1 - Business Activity, Marketing and People
      • Unit 2 - Operations, Finance and Influences
    • Exam Technique
  • Contact