A computer network is a group of devices connected together so that they can communicate.
Devices can be computers, laptops, phones, printers, servers and smart devices.
Links can be wired (copper or fibre cables) or wireless (radio signals).
Networks only work because every device follows the same protocols (sets of rules).
Advantages of networking
Disadvantages of networking
Share hardware: many users can print to one printer or share one Internet connection, so fewer devices need buying.
Set-up cost: servers, cabling, switches and routers must be bought and installed.
Share files and data: staff can open the same files from any computer, without copying them on USB sticks.
Management: a large network needs skilled staff to run it, which costs money.
Communicate and collaborate: email, messaging and editing shared documents at the same time.
Malware can spread from one infected device to others on the network.
Central management: software can be installed or updated on every computer from one place, and backups can be done centrally.
More ways in for attackers: every connected device is a possible point of attack.
Log on anywhere: users can reach their own files from any computer on the network.
Dependence: if a key server or link fails, many users can lose access at once.
Exam tipMake each point specific: not ‘it’s cheaper’ but ‘one printer can be shared by 30 computers, so fewer printers are bought’. In a ‘discuss networking’ question, stay on advantages and disadvantages for the people in the question: don’t drift into network types.
Computer network
A group of devices connected together so that they can communicate.
AQA 8525 · 3.5
PAN, LAN and WAN
Networks are named by the geographical area they cover and by who owns them.
PAN: for AQA you only need to know Bluetooth. It links personal devices over a short range.
LAN: one owner can buy, run and secure the whole network.
WAN: no single organisation owns all of it. Links are often leased from telecoms companies, and many organisations own different parts.
A WAN usually joins many LANs together.
Type
Area covered
Ownership
Example
PAN (Personal Area Network)
A small distance around one person (about 10 m)
One person
A phone linked by Bluetooth to earbuds and a smartwatch
LAN (Local Area Network)
A relatively small geographical area: one building or site
Often owned and managed by a single person or organisation
A school’s network, or a home network
WAN (Wide Area Network)
A wide geographical area: across towns, countries or the world
Often under collective or distributed ownership
The Internet, the biggest WAN; a company linking offices in several cities
PAN, LAN and WAN
Exam tip‘A WAN is bigger’ is too vague: say it covers a wider geographical area. And don’t say a LAN is wired and a WAN is wireless: either can use either.
Exam tipIn a ‘describe two differences’ question, give both sides each time: ‘A LAN covers a small geographical area, whereas a WAN covers a wide one.’
PAN
Personal Area Network: devices connected and communicating over a small distance around one person.
LAN
Local Area Network: devices connected and communicating over a small geographical area, often owned and controlled by a single person or organisation.
WAN
Wide Area Network: devices connected and communicating over a large geographical area, often under collective or distributed ownership.
AQA 8525 · 3.5
Wired and wireless networks
Networks can be wired (data sent along cables) or wireless (data sent by radio signals). Many use both.
Wireless
Wired
Advantages
Users can move around with laptops and phones and stay connected. No cables to run through walls, so it’s quicker to install in an existing building. Easy to add new devices, including visitors’ own devices.
Higher data transfer rates, because each device has its own cable. A more stable connection with fewer dropouts, as cables suffer less interference. Data is harder to intercept: an attacker needs physical access to a cable.
Disadvantages
Interference from walls, other devices and nearby networks can slow or drop the connection. Limited range: the signal weakens with distance from the access point. Signals can be picked up by anyone in range, so data must be encrypted. Bandwidth is shared by every device in range, so it slows as more connect.
Devices are tied to a socket, so users can’t move around. Cables, sockets and switches cost more and take longer to install, especially in old buildings. Cables can be damaged or trip people up.
Exam tipIf asked for disadvantages of wireless, write about wireless (‘the signal can be blocked by thick walls’), not about wired advantages. ‘Easier’, ‘cheaper’ and ‘greater range’ need a reason: ‘no cables need to be run through walls’.
Wired network
A network in which data travels along cables, such as copper or fibre-optic cables.
Wireless network
A network in which data travels by radio signals between devices.
Interference
Unwanted signals or obstacles that disrupt the data being sent.
AQA 8525 · 3.5
Cables: fibre and copper
Wired networks use copper cable (data sent as electrical signals) or fibre-optic cable (data sent as pulses of light through glass or plastic fibres).
Fibre-optic cable
Copper cable
Strengths
Higher bandwidth: data is transmitted faster. Data can travel much longer distances before the signal weakens. Not affected by electrical interference. Harder to tap, so data is harder to intercept.
Cheaper to buy per metre. Easier to install: easy to cut, bend and join, with no specialist tools. Fast enough for short runs inside a building.
Weaknesses
Costs more to buy and install; joining fibres needs specialist kit. More fragile if bent sharply.
Signal weakens over longer distances. Affected by electrical interference. Lower bandwidth than fibre.
When it’s appropriate
Long links: between buildings, towns or countries; and where a lot of data must move fast, e.g. to a server room.
Short runs: from a wall socket to a desk, or around a small office or home, where cost matters more than top speed.
Exam tip‘The cable is faster’ earns nothing. It’s the data transmission that is faster, and fibre is more reliable over long distances. Always say what is faster or more reliable, and why.
Fibre-optic cable
A cable that carries data as pulses of light along glass or plastic fibres.
Copper cable
A cable that carries data as electrical signals.
Bandwidth
How much data can be sent over a connection each second.
AQA 8525 · 3.5
Network protocols
A network protocol is a set of rules that allow devices/networks to communicate/transfer data.
Devices on a network are made by many different companies and run different software. They can only understand each other if they all follow the same rules.
A protocol can set rules for things such as: how data is split up and addressed, the format of a message, how errors are found, and how two devices start and end a conversation.
Different protocols do different jobs. You need six: TCP, IP, HTTP, HTTPS, SMTP and IMAP.
Protocol
Stands for
Used for
TCP
Transmission Control Protocol
Setting up communication between two end points and making sure it is error free
IP
Internet Protocol
Addressing packets and routing them across networks
HTTP
Hypertext Transfer Protocol
Sending and receiving web pages between a web server and a browser
HTTPS
Hypertext Transfer Protocol Secure
The same as HTTP, but the data is encrypted, for more secure web transactions
SMTP
Simple Mail Transfer Protocol
Sending email to an email server
IMAP
Internet Message Access Protocol
Receiving email from an email server and managing it on the server
Exam tipIf asked to define a protocol, give the definition: a set of rules (not ‘instructions’) that allow devices to communicate. Don’t describe what one particular protocol does.
Not on the courseOlder books, videos and past papers also teach network topologies, Ethernet, Wi-Fi, UDP, FTP and other names for the link layer. They’re no longer on the AQA course, so you can skip them.
Network protocol
A set of rules that allow devices/networks to communicate/transfer data.
AQA 8525 · 3.5
TCP and IP
TCP and IP work together to get data from one device to another, in small pieces called packets.
TCP (Transmission Control Protocol) establishes communication between two end points and ensures that this is error free.
TCP splits a message into packets and gives each one a sequence number. The receiver uses the numbers to put the packets back in the right order.
The receiver checks each packet arrived without errors. If one is missing or damaged, it is sent again.
IP (Internet Protocol) is responsible for routing: it adds the source and destination IP addresses to each packet so it can be sent across networks to the right device.
Packets from one message may take different routes and arrive out of order. That’s fine: TCP puts them back together.
An IP address is given to a device when it connects to a network. An IPv4 address is four bytes, written as four numbers from 0 to 255 separated by dots, e.g. 172.16.254.9.
TCP splits the message into numbered packets; IP routes each one
Why 0 to 255?Each part of an IPv4 address is one byte (8 bits). The largest 8-bit number is 11111111 = 255, so 192.168.300.4 can’t be a valid address.
TCP
A transport layer protocol used to establish communication between two end points and ensure that this is error free.
IP
An internet layer protocol responsible for routing packets; it uses IP addresses.
Packet
A small chunk of a message, sent separately across a network.
IP address
An address given to a device when it connects to a network; IPv4 uses four bytes, each 0 to 255.
AQA 8525 · 3.5
HTTP, HTTPS, SMTP and IMAP
These four protocols are used by the apps people use every day: web browsers and email.
HTTP is used for sending and receiving web pages between a web server and a client (browser). The browser requests a page; the server sends it back.
HTTPS is HTTP with encryption. The data sent between the browser and the server is encrypted, so if it is intercepted it can’t be understood. It’s used for logins, payments and anything private.
SMTP is used to send email: from your email app to your mail server, and from that server on to the recipient’s mail server.
IMAP is used to receive email from a mail server and to manage it on the server. Emails stay on the server, so a phone and a laptop show the same inbox: delete an email on one and it’s gone from the other.
SMTP sends email on; IMAP reads it from the server
Exam tipHTTP sends web pages, not ‘websites’. HTTPS doesn’t ‘make a website secure’ or ‘encrypt the website’: it encrypts the data sent between the browser and the web server.
Memory hookSMTP = Send. IMAP = Inbox kept on the server.
HTTP
An application layer protocol for sending and receiving web pages between a web server and a client (browser).
HTTPS
HTTP with encryption added, allowing for secure data movement.
SMTP
An application layer protocol used to send emails to another email server.
IMAP
An application layer protocol used to receive emails from an email server and to manage emails on the server.
AQA 8525 · 3.5
The 4-layer TCP/IP model
The TCP/IP model is a protocol stack used when connecting a device to the Internet. It has four layers. Each layer does one job and passes data to the next.
Learn the order from the top: Application, Transport, Internet, Link. One way to remember it: ‘All Teachers Inspect Laptops’.
Layering means each layer only has to do its own job. One layer can be changed (e.g. new network hardware) without changing the others.
Layer
Main function
Application (top)
Where the network applications, such as web browsers and email programs, operate.
Transport
Sets up the communication between the two hosts, and they agree settings such as the size of packets. Splits data into packets with sequence numbers.
Internet
Addresses and packages data for transmission (adds source and destination IP addresses). Routes the packets across the network.
Link (bottom)
Where the network hardware, such as the NIC (network interface card), is located. OS device drivers also operate here.
Data goes down the layers to be sent and up the layers when it arrives
Exam tipUse the TCP/IP names exactly. ‘Data Link’ and ‘Network’ layer are from a different model and were marked wrong.
TCP/IP model
A protocol stack used when connecting a device to the Internet, with four layers: Application, Transport, Internet and Link.
NIC
Network interface card: the hardware that connects a device to a network. It sits at the link layer.
AQA 8525 · 3.5
Which protocol works at which layer
Every protocol on the course belongs to one layer.
Layer
Protocols you need
Application
HTTP, HTTPS, SMTP, IMAP
Transport
TCP
Internet
IP
Link
No named protocol: the NIC and device drivers
Worked exampleA student types a web address into their browser. Follow the request for the web page down the layers on the student’s computer.
Application layer: the browser makes an HTTPS request for the page.Browsers work at the application layer; HTTPS because the page is requested securely.
Transport layer: TCP sets up communication with the web server, agrees settings such as packet size, and splits the request into numbered packets.TCP works at the transport layer.
Internet layer: IP adds the source and destination IP addresses to each packet.IP addresses and routes packets.
Link layer: the NIC and its driver send the packets out onto the network as signals.The hardware is at the link layer.
At the web server the packets go back up the layers: the NIC receives them, IP checks the address, TCP puts the packets in order, and the web server software reads the request.Receiving reverses the order.
Application (HTTPS) → Transport (TCP) → Internet (IP) → Link (NIC), then back up at the server.
Exam tipIf you can’t remember a layer’s job, use its protocol: the transport layer is where TCP works, so it deals with packets and setting up the connection; the internet layer is where IP works, so it deals with addresses and routing.
Genious