2.1 Networks – The Internet
2.1.1 Purpose & Benefits of Core Networking Devices
- Network Interface Card (NIC) – provides the physical link between a computer and the network medium; converts data between digital signals and the electrical/optical signals used on the cable.
- Switch – connects multiple devices within a LAN, forwards frames based on MAC addresses, and creates a separate collision domain for each port, improving bandwidth utilisation.
- Router – interconnects different networks (e.g., LAN to WAN), routes packets using IP addresses, and can provide services such as NAT, DHCP and basic firewall filtering.
- Access Point (AP) – adds wireless capability to a wired LAN, allowing devices to connect via Wi‑Fi while still being managed by the switch/router.
- Hub (legacy) – simple multi‑port repeater; repeats incoming signals to all ports, creating a single collision domain (generally replaced by switches).
2.1.2 LAN vs WAN
| Aspect | Local Area Network (LAN) | Wide Area Network (WAN) |
|---|
| Geographic scope | Single building or campus (≤ several kilometres) | Regional, national or global (hundreds to thousands of kilometres) |
| Typical bandwidth | 10 Mbit s⁻¹ – 10 Gbit s⁻¹ (Ethernet, fibre) | 1 Mbit s⁻¹ – 100 Mbit s⁻¹ (leased lines, MPLS, satellite) |
| Ownership of infrastructure | Usually owned/managed by a single organisation | Often provided by telecom operators; shared among many users |
| Latency | Low (microseconds to a few milliseconds) | Higher (tens to hundreds of milliseconds) |
| Typical devices | Switches, NICs, Wi‑Fi APs, routers (gateway to WAN) | Routers, modems, satellite dishes, MPLS switches |
2.1.3 Network Topologies
Physical layouts used in LANs and WANs. The choice influences cost, fault‑tolerance and performance.
| Topology | Key Features | Pros | Cons |
|---|
| Bus | All devices share a single coaxial cable. | Simple, inexpensive cabling. | Single point of failure; limited scalability. |
| Star | Each device connects to a central hub or switch. | Easy to add/remove devices; fault isolated to individual links. | Central device failure disables whole network. |
| Ring | Devices connected in a closed loop; data travels in one direction. | Predictable performance; efficient for token‑passing protocols. | Break in the ring stops all traffic; more complex cabling. |
| Mesh | Multiple redundant paths between devices. | High reliability and fault tolerance. | Expensive and complex to install. |
| Hybrid | Combination of two or more basic topologies (e.g., star‑bus). | Flexibility to balance cost and performance. | Design and troubleshooting can be more involved. |
ASCII Example – Star Topology
[PC1]
|
[PC2]---[Switch]---[PC3]
|
[PC4]
2.1.4 Wireless vs Wired Networks
- Wired (Ethernet) – uses copper (Cat 5e/6/6a) or fibre cables; offers deterministic latency, higher security, and typically greater bandwidth.
- Wireless (Wi‑Fi – IEEE 802.11) – provides mobility and ease of deployment; performance depends on signal strength, interference and channel congestion, which can affect thin‑client responsiveness.
- Hybrid deployments – a wired backbone for switches/routers combined with wireless APs for end‑users gives the best of both worlds.
2.1.5 Fundamental Network Services
IP Addressing (IPv4 & IPv6)
IPv4 uses four octets (e.g., 192.168.1.10) with a subnet mask (e.g., 255.255.255.0) to define the network portion. The mask can also be written in CIDR notation: 192.168.1.10/24.
IPv6 expands the address space to 128 bits, written as eight groups of four hexadecimal digits, e.g., 2001:0db8:85a3:0000:0000:8a2e:0370:7334. Leading zeros can be omitted and consecutive groups of zeros can be compressed using :: (only once per address). Example:
Full: 2001:0db8:0000:0000:0000:0000:1428:57ab
Compressed: 2001:db8::1428:57ab
IPv6 also uses CIDR notation, e.g., 2001:db8::/48, where the prefix length tells how many bits belong to the network part.
DNS (Domain Name System)
Translates human‑readable domain names (e.g., www.example.com) into IP addresses. A typical query flow is:
- Client sends a request to a recursive resolver (often provided by the ISP).
- The resolver checks its cache; if not found, it contacts the root name server.
- From the root it is directed to the appropriate TLD server (.com, .org, …).
- The TLD server points to the authoritative server for
example.com, which finally returns the IP address.
URL (Uniform Resource Locator)
Identifies a resource on the Internet:
protocol://host[:port]/path?query#fragment
Example: https://www.cambridge.org/subjects/computer-science
2.1.6 Ethernet, CSMA/CD and Bit‑Streaming
- Ethernet frame – preamble, destination MAC, source MAC, EtherType/length, payload (46–1500 bytes), and CRC checksum.
- CSMA/CD (Carrier Sense Multiple Access with Collision Detection) – devices listen to the medium before transmitting; if a collision occurs they back off for a random period and retry. Modern switched Ethernet gives each port its own collision domain, so collisions are rare, but the principle is still examined for exam purposes.
- Bit‑streaming vs. packet‑switching – real‑time services (VoIP, video) may use continuous bit streams, whereas most data traffic is packet‑switched, allowing many flows to share the same link.
Binary vs. Decimal Prefixes
Network specifications often use decimal prefixes (kilo = 10³, mega = 10⁶, giga = 10⁹). Operating systems and some utilities report binary prefixes (kibi = 2¹⁰, mebi = 2²⁰, gibi = 2³⁰). Example:
- 1 Gbps (decimal) = 1 × 10⁹ bits s⁻¹
- 1 Gibps (binary) = 2³⁰ bits s⁻¹ ≈ 1.074 Gbps
Understanding the difference prevents mis‑interpretation of advertised speeds.
2.1.7 Client‑Server and Peer‑to‑Peer (P2P) Models
Client‑Server – a central server provides resources (files, applications, databases) to multiple client devices that request services. Thin and thick clients are both forms of the client in this model.
Peer‑to‑Peer – each node can act as both client and server, sharing resources directly with other peers (e.g., file‑sharing applications, BitTorrent).
ASCII Diagram – Client‑Server vs. P2P
Client‑Server: Peer‑to‑Peer:
[Client] [Peer]---[Peer]
| | |
[Server]-------------------[Peer]---[Peer]
2.1.8 Thin‑Client vs Thick‑Client (Expanded)
Definitions
- Thin client – a lightweight device that provides only input, output and a minimal operating environment. The bulk of processing, data storage and application execution is performed on a central server (often via Virtual Desktop Infrastructure – VDI or cloud‑based SaaS).
- Thick client (fat or rich client) – a fully‑featured computer that runs its own operating system and applications locally, while still being able to communicate with servers for additional services.
Typical Architecture
- Thin client
- Keyboard / mouse / touch input
- Display (often thin‑client‑specific OS or firmware)
- Network interface (Ethernet or Wi‑Fi)
- Minimal CPU/RAM (e.g., ARM Cortex‑A7, 256 MiB RAM)
- Relies on a server running VDI, Remote Desktop Services, or cloud‑based SaaS.
- Thick client
- Full‑spec CPU (x86‑64), ≥ 4 GiB RAM, SSD/HDD storage
- Complete OS (Windows, macOS, Linux)
- Locally installed applications (IDE, CAD, games)
- Network interface for optional server access (file shares, web services).
Advantages & Disadvantages
| Thin Client | Thick Client |
|---|
| Hardware cost | Low (often <$300) | Higher (>\$600–\$1500) |
| Maintenance | Centralised updates; minimal on‑site work | Updates required on each device |
| Security | Data stays on server; easier policy enforcement | Data stored locally; needs endpoint protection |
| Performance | Depends on server capacity and network latency | Local CPU/GPU deliver high performance |
| Offline capability | Very limited; must be online | Full functionality without network |
| Power consumption | Low (≈ 5–10 W) | Higher (≈ 30–100 W) |
When to Choose Which?
- Use a thin client when:
- Centralised control, security and easy maintenance are priorities.
- Applications are web‑based, virtualised, or delivered as SaaS.
- Budget constraints limit hardware upgrades.
- Users operate in a stable, high‑speed LAN/WAN (e.g., corporate office, school computer lab).
- Use a thick client when:
- High‑performance, graphics‑intensive or specialised software is required (e.g., CAD, video editing, scientific simulations).
- Users need to work offline or in environments with unreliable connectivity.
- Custom OS configuration or local software installation is essential.
- Latency‑sensitive tasks cannot tolerate network delays.
2.1.9 Cloud Computing (Relation to Thin Clients)
Cloud computing delivers computing resources (servers, storage, applications) over the Internet on a pay‑as‑you‑go basis. Thin clients are a natural endpoint for cloud services because most processing occurs remotely.
Example: Google Workspace – users access Docs, Sheets and Slides through a web browser on a thin client; all documents are stored in Google Drive on Google’s data centres.
2.1.10 Network Security Basics (AS‑Level Requirement)
- Firewalls – filter traffic based on IP address, port number and protocol; can be hardware (router‑based) or software (host‑based).
- Intrusion Detection/Prevention Systems (IDS/IPS) – monitor network traffic for suspicious patterns; IDS alerts, IPS can block the traffic.
- Encryption – protects data in transit (e.g., TLS/HTTPS, VPNs) and at rest (e.g., full‑disk encryption on thick clients).
- Authentication & Authorisation – verifies identity (passwords, 2‑FA, certificates) and determines what resources a user may access.
- Secure network design – segmentation with VLANs, DMZs for public services, and least‑privilege principles.
2.1.11 Network Management Tools (AS‑Level Requirement)
- SNMP (Simple Network Management Protocol) – enables monitoring of devices (CPU load, interface traffic, error rates) and remote configuration.
- Network monitoring software – tools such as Wireshark (packet capture), Nagios/Zabbix (service health), and PRTG (bandwidth utilisation).
- IP address management (IPAM) – tracks allocation of IPv4/IPv6 addresses, subnet usage and DHCP scopes.
- Logging & Auditing – centralised log servers (e.g., syslog) help detect security incidents and diagnose faults.
Suggested Diagram (for classroom hand‑out)
A side‑by‑side schematic showing the processing location for each client type.
ASCII Diagram – Thin vs. Thick Client
Thin client Thick client
[Thin] [Thick]
| |
Network ------------------- Local CPU, RAM, Storage
| |
[Server] <--- Processing & Data ----------------> [Internet/Cloud]
Arrows indicate the direction of data flow. The thin‑client side highlights “Server‑side processing”, whereas the thick‑client side highlights “Local processing”.