Network models & topologies, cloud computing, wired & wireless media, LAN & internet hardware, Ethernet & CSMA/CD, bit streaming, IP addressing, and URLs & DNS.
🎯 By the end of this topic you should be able to: compare network models and topologies and justify a choice for a given scenario; describe the hardware supporting a LAN and the internet; explain Ethernet/CSMA/CD, IP addressing, and how a URL is resolved via DNS.
Arrow keys or buttons to navigate | F11 for full screen
📌 2 Communication has a single sub-topic: 2.1 Networks including the internet — this presentation covers all of it.
A network is two or more devices connected together so they can communicate and share resources, rather than operating as isolated, stand-alone machines.
Expensive peripherals (printers, scanners) and storage can be shared by many users instead of duplicated at every device.
Files can be accessed from multiple devices; software licences and updates can be managed centrally rather than per machine.
Enables email, messaging and collaboration between users, and shared access to a single internet connection.
📌 Networking also enables centralised management — security policies, user accounts and backups can be controlled from one place rather than device-by-device.
| Feature | LAN (Local Area Network) | WAN (Wide Area Network) |
|---|---|---|
| Geographic area | Small — a single site or building | Large — across cities, countries, or globally |
| Ownership | Typically owned and managed by a single organisation | Typically relies on third-party infrastructure (leased lines, ISPs, satellite links) |
| Data transfer speed | Usually faster — shorter distances, dedicated infrastructure | Usually slower — greater distance and more intermediate devices |
| Cost | Lower cost to set up and maintain | Higher cost — needs leased lines, routers and long-distance links |
| Example | A school or office network | The internet; a bank's network linking branches nationwide |
A central server provides services (files, authentication, printing) to requesting client computers.
✅ Centralised security/backups/management; only one copy of shared data needed; scales well as clients are added
❌ Expensive server hardware/software; server is a single point of failure; needs IT expertise to maintain
No central server — every peer is equal and can act as both client and server, sharing resources directly.
✅ Cheap and simple to set up; no single point of failure; suited to small networks
❌ Harder to secure/manage centrally; performance degrades as more peers join; backups must be done on each device
📌 Subnetwork models: a larger network can be divided into smaller subnetworks (e.g. separate staff and student subnets in a school) for better organisation, performance and security — each subnet can still use a client-server or peer-to-peer model internally.
📌 Justify: a large school with hundreds of accounts needing centralised login and security → client-server. A small office of 4–5 staff informally sharing files → peer-to-peer (cheaper, no IT staff needed).
Minimal local hardware/software — relies almost entirely on the server for processing and storage; mainly handles input/output.
✅ Cheaper devices, easy centralised updates/maintenance, lower power use, more secure (data stays on server)
❌ Useless without a working network/server connection; server must be powerful enough for all clients
Has significant local processing power, storage and installed software — can operate largely independently of the server.
✅ Can keep working if the network/server connection drops; better performance for demanding applications
❌ More expensive hardware per device; updates/patches must be applied to every device individually
Devices attach to one shared cable; red end-caps are terminators that prevent signal reflection.
All devices connect to a single central cable (the "bus"), terminated at each end.
How it works: a signal travels along the whole length of the bus; every device "sees" it, but only the intended recipient accepts it — the rest ignore it.
✅ Cheap and easy to install; uses the least cabling of any topology
❌ A break in the main cable brings down the whole network; performance drops as more devices are added, since bandwidth is shared and collisions become more likely
Every device has its own dedicated cable to a central switch (gold).
Every device connects individually to a central switch (or hub) via its own cable.
How it works: a device sends data to the switch, which forwards it directly only to the intended recipient device.
✅ One cable or device failing doesn't affect the rest of the network; fewer collisions than a bus; easy to add or remove devices
❌ If the central switch fails, the whole network goes down; needs more cabling than a bus, so higher installation cost
In a full mesh, every device connects directly to every other device.
Devices connect directly to several/all other devices. Full mesh = every device interconnected; partial mesh = only some direct links.
How it works: data can take multiple possible routes to its destination; if one link fails, data is simply rerouted via another path.
✅ Highly resilient — no single point of failure; handles high traffic volumes well
❌ Expensive and complex — many cables/connections are needed, and it becomes hard to maintain as the network grows
Two star segments (gold hubs) joined by a backbone link — a common "star-bus" hybrid.
A combination of two or more topologies, designed to suit the needs of different parts of a network.
✅ Flexible — cheaper topologies can be used where less critical, and more resilient ones where they're needed most
❌ Complex to design, install and manage; can still be costly overall
📌 Justify: a bank's data centre linking mission-critical servers → mesh (maximum resilience — downtime is unacceptable). A small office → star (good balance of cost, performance and ease of adding devices). A school with several buildings, each with its own star network → hybrid (each building's star is cheap/simple, linked by one resilient backbone).
Cloud computing means storing data and running applications on remote servers accessed via the internet, rather than on local hardware.
Infrastructure owned and run by a third-party provider (e.g. AWS, Google Cloud, Microsoft Azure) and shared among many customers over the internet.
Infrastructure dedicated to a single organisation — hosted on-site or by a third party, but not shared with other customers — giving more control and security.
✅ Benefits
❌ Drawbacks
| Medium | Advantages | Disadvantages |
|---|---|---|
| Copper cable | Cheap, easy to install | Susceptible to interference; signal degrades over distance; easier to intercept |
| Fibre-optic cable | Very high speed; long range without signal loss; immune to electromagnetic interference; more secure | Expensive; fragile; needs specialist installation |
| Radio waves (incl. WiFi) | No cables — flexible, mobile, easy to add devices | Limited range; blocked by walls/obstacles; less secure (signal can be intercepted) |
| Microwaves | Long distance without cables (e.g. building to building) | Needs clear line-of-sight; affected by weather |
| Satellite | Coverage in remote areas with no other infrastructure | High latency; expensive; affected by weather |
📌 Overall implication: wired connections are generally faster, more reliable and more secure but limit mobility and require physical installation. Wireless connections offer flexibility and easy installation but are typically slower, less reliable and less secure.
| Device | Function |
|---|---|
| Switch | Connects multiple devices on a LAN; forwards data only to the intended device (using MAC addresses) rather than to everyone |
| Server | A powerful central computer that manages network resources and provides services to clients |
| NIC (Network Interface Card) | Hardware allowing a device to physically connect to and communicate over a network; has a unique MAC address |
| WNIC (Wireless NIC) | A NIC that connects a device to a network wirelessly, e.g. via WiFi |
| WAP (Wireless Access Point) | Allows wireless devices to connect to a wired network, extending coverage to a wireless area |
| Cables | The physical medium (copper, fibre-optic) carrying data between wired devices |
| Bridge | Connects two separate LAN segments so they function as a single network |
| Repeater | Receives and retransmits a signal at full strength, extending the distance it can travel without degrading |
A router connects multiple different networks together — e.g. a home or school LAN to the internet, or two separate WANs.
📌 Inside the dashed LAN boundary, the switch wires together PCs, a server and a WAP (which serves wireless laptops/phones). The switch's uplink leaves the LAN and reaches the router, which sits at the boundary between the LAN and the wider WAN/internet — examining each packet's IP address to forward it correctly beyond the local network.
Ethernet is a widely-used set of standards defining how data is formatted and transmitted over a wired LAN. Where a transmission medium is shared, two devices transmitting at once causes a collision. CSMA/CD (Carrier Sense Multiple Access with Collision Detection) manages this.
Carrier Sense: a device "listens" to the medium to check whether it is currently free (no other device transmitting).
Multiple Access: any device may attempt to transmit once it senses the medium is free — there is no central controller granting permission.
Collision Detection: if two devices transmit at exactly the same time, a collision occurs and is detected (the signal becomes corrupted).
Transmitting devices stop immediately, send a jam signal to alert others, then each waits a random back-off period before retrying — reducing the chance of an immediate repeat collision.
Bit streaming is the continuous transfer of data (typically audio/video) as a steady stream of bits, allowing playback as it arrives rather than waiting for the whole file to download.
Data is streamed live, as it is generated — e.g. a live video call or live sports broadcast. Must be received and played with minimal delay.
Pre-recorded data is streamed whenever the user chooses — e.g. a saved video on a streaming service — and can typically be paused or rewound.
📌 Bit rate & broadband speed: the bit rate is the amount of data streamed per second. If the connection speed is lower than the required bit rate, the stream will buffer/stutter, or the service must reduce quality (a lower bit rate/resolution) to match the available bandwidth. Higher-quality audio/video needs a higher bit rate and therefore a faster, more reliable connection.
| Feature | The Internet | The World Wide Web (WWW) |
|---|---|---|
| What it is | A global network of interconnected computer networks | A service that runs on the internet — webpages linked by hyperlinks |
| Includes | Email, file transfer, streaming, gaming, the WWW, and more | Just one of many services that use the internet |
| Access | Requires infrastructure & protocols (e.g. TCP/IP) to connect devices | Accessed with a web browser, using protocols such as HTTP/HTTPS |
📌 A useful analogy: the internet is the road network; the WWW is just one type of traffic using specific roads for a specific purpose. The internet is the underlying infrastructure — the WWW is one application that runs on top of it.
| Hardware | Role |
|---|---|
| Modem | Converts digital data into a signal suitable for the connection type used (e.g. phone line, cable), and converts incoming signals back into digital data |
| PSTN (Public Switched Telephone Network) | The traditional telephone network, which can also carry internet data (e.g. dial-up, DSL broadband) |
| Dedicated (leased) lines | A private, permanent connection leased from a provider, giving guaranteed bandwidth not shared with other customers — common for businesses |
| Cell phone network | Mobile infrastructure (masts/cell towers) providing internet access to mobile devices via cellular data (e.g. 4G, 5G) |
An IP address uniquely identifies a device on a network, allowing data to be correctly routed to it.
| Feature | IPv4 | IPv6 |
|---|---|---|
| Format | 32-bit; 4 decimal numbers (0–255) separated by dots, e.g. 192.168.1.1 | 128-bit; 8 groups of 4 hex digits separated by colons, e.g. 2001:0db8:85a3::8a2e:0370:7334 |
| Total addresses | ≈ 4.3 billion (2³²) | ≈ 340 undecillion (2¹²⁸) — vastly more |
| Why it exists | The original internet addressing system | Introduced because IPv4 addresses were running out as more devices connected |
📌 How an IP address is associated with a device: when a device connects to a network it is assigned an IP address — either configured manually, or automatically by a DHCP server — which routers then use to correctly direct packets to and from that specific device.
Subnetting divides one larger network into smaller sub-networks (subnets) — improving performance (less congestion per subnet) and security (subnets can be isolated, e.g. separate staff and guest WiFi).
Public: a unique address visible/routable on the internet, assigned to a router by an ISP.
Private: only valid within a local network (LAN); not directly reachable from the internet — hidden behind the router (via NAT), which adds a layer of security since external attackers cannot target it directly.
Static: manually assigned, never changes — used where a consistent address matters, e.g. a web/email server or network printer.
Dynamic: automatically assigned (usually by DHCP) and may change on reconnection — simpler to manage for everyday client devices and reuses addresses efficiently.
A URL (Uniform Resource Locator) is a text address used to locate a specific resource on the World Wide Web.
https://www.csexamready.com/notes/unit1.html
■ protocol — method used to access the resource · ■ domain name — identifies the web server · ■ path — the specific file/location on that server
The user types a URL; the browser needs the numerical IP address of the web server (computers route data using IP addresses, not domain names).
The browser sends the domain name to a DNS server, which acts like a "phone book" for the internet.
The DNS server looks up (or asks other DNS servers to look up) the IP address matching that domain name.
The matching IP address is returned to the browser, which uses it to request the resource directly from the correct web server.
| Term | Definition |
|---|---|
| LAN / WAN | A network covering a small local site / a network spanning a large geographic area |
| Client-server | A model where a central server provides services to requesting client computers |
| Peer-to-peer | A model where all devices are equal and share resources directly, with no central server |
| Thin / thick client | A client relying mostly on the server for processing / a client with significant local processing power |
| Topology | The physical/logical layout of how devices are connected in a network (bus, star, mesh, hybrid) |
| Router | Connects different networks together and forwards packets based on destination IP address |
| CSMA/CD | An Ethernet method for detecting and recovering from data collisions on a shared medium |
| Bit streaming | Continuous transfer of data allowing playback as it arrives, rather than after a full download |
| IP address | A unique numerical address identifying a device on a network (IPv4: 32-bit; IPv6: 128-bit) |
| DNS | The service that translates a human-readable domain name into its numerical IP address |
| URL | A text address used to locate a specific resource on the World Wide Web |
State two benefits of connecting computers in a network rather than using them as stand-alone machines.
[2]Any two (1 each): shared hardware (e.g. printers); shared files/data; shared internet connection; centralised software/security management; enables communication between users.
A small dental practice with 4 computers wants to share patient files informally between staff. Justify whether a client-server or peer-to-peer model would be more appropriate.
[4]Peer-to-peer is likely more appropriate (1). It is cheap and simple to set up, with no expensive server hardware/software required (1). With only 4 computers, the network is small enough that centralised management is not essential (1). No dedicated IT staff are needed to maintain a server (1). [Accept a justified answer for client-server if patient data security/confidentiality is used as the justification.]
Explain how CSMA/CD is used to manage data collisions on an Ethernet network.
[4]A device listens to the medium and only transmits once it senses it is free (carrier sense) (1). Any device may transmit once the medium is free, with no central controller (multiple access) (1). If two devices transmit simultaneously, a collision is detected (1). The devices stop, send a jam signal, and each waits a random back-off period before retransmitting, reducing the chance of an immediate repeat collision (1).
Describe how a router uses an IP address to forward a data packet across the internet.
[3]The router examines the destination IP address contained in the data packet (1). It determines the best available path/route towards that destination (1). It forwards the packet on towards the next router/network in that path, repeating until the packet reaches its destination device (1).
Explain the difference between a static and a dynamic IP address, and give one example of a device that would typically use a static IP address.
[3]A static IP address is manually assigned and never changes (1); a dynamic IP address is automatically assigned (e.g. by DHCP) and may change each time the device reconnects (1). Example of a static IP use: a web server / email server / network printer (1).
Explain the role of DNS when a user types a URL into a web browser.
[3]Computers communicate using numerical IP addresses, not domain names, so the domain name in the URL must be translated (1). The browser sends the domain name to a DNS server, which looks up the matching IP address (1). The IP address is returned to the browser, which uses it to contact the correct web server (1).
Explain why fibre-optic cable is often preferred over copper cable for a network's backbone connection.
[3]Fibre-optic transmits data as light rather than electrical signals, allowing much higher transmission speeds (1). It suffers far less signal loss over long distances than copper (1). It is immune to electromagnetic interference and harder to physically tap, making it more secure (1).