Show understanding of the bus, star, mesh and hybrid topologies

2.1 Networks – The Internet and LAN Topologies

Learning Objective

Show understanding of the bus, star, mesh and hybrid topologies, the devices that support them and the wider networking concepts required by the Cambridge International AS & A Level Computer Science syllabus (2026).

1 Key Concepts

• Network topology – physical layout of cables (physical topology) and the way data moves through the network (logical topology).
• Physical vs. logical topology – e.g. a star‑wired LAN can have a logical bus (shared‑medium) or logical star (switch‑based).
• Network devices – NIC, hub, switch, router, bridge, gateway, firewall, access point, modem.
• LAN vs. WAN – Local Area Network (single building/campus, high bandwidth) vs. Wide Area Network (city‑wide or global, lower bandwidth, often uses public infrastructure).
• Client‑server & peer‑to‑peer (P2P) models
  • Client‑server – centralised servers provide resources (e.g. web server, file server).
  • Peer‑to‑peer – each node can act as both client and server (e.g. BitTorrent, Skype).
• Thin‑client vs. thick‑client – thin client relies on a server for processing; thick client performs most processing locally.
• Cloud computing models – IaaS, PaaS, SaaS; affect the choice of internal LAN topology.
• Wired vs. wireless media – copper (Cat5e/6), fibre‑optic, Wi‑Fi, Bluetooth, satellite; compare speed, range, cost and security.
• Ethernet & CSMA/CD – frame structure, carrier‑sense multiple access with collision detection; relevant for shared‑medium topologies (bus, hub‑based star).
• Protocol stack & layers – OSI‑like 7‑layer model and the TCP/IP suite (Link, Internet, Transport, Application). Understanding which protocols operate at each layer (e.g. Ethernet at Link, IP at Internet, TCP/UDP at Transport, HTTP/FTP at Application).
• Circuit‑switching vs. packet‑switching – circuit‑switched networks reserve a dedicated path (e.g. traditional telephone); packet‑switched networks break data into packets and route each independently (e.g. Internet).
• Network performance metrics – latency, bandwidth, jitter, throughput, packet loss; how topology influences each metric.
• IP addressing – IPv4 (32‑bit) and IPv6 (128‑bit), public vs. private ranges, static vs. dynamic allocation, basic subnetting.
• URL & DNS – structure of a URL and the recursive resolution process that maps a hostname to an IP address.
• Security threats & counter‑measures
  • Threats: phishing, malware, denial‑of‑service (DoS/DDoS), man‑in‑the‑middle, eavesdropping.
  • Counter‑measures: firewalls (packet‑filtering, stateful), IDS/IPS, VPN (IPsec, SSL/TLS), encryption (WPA2/WPA3, TLS), authentication (passwords, 2‑FA, certificates).

2 Overview of Common Network Topologies

2.1 Bus Topology

• Structure – a single backbone cable with each node tapped onto it.
• Logical behaviour – shared medium; all nodes see every frame.
• Medium access – CSMA/CD; only one node transmits at a time, collisions are detected and retransmitted.
• Typical use – small legacy LANs, temporary setups, 10 Base‑2/5 Ethernet.
• Advantages
  • Very low cabling cost.
  • Easy to add a new node by attaching another tap.
• Disadvantages
  • Backbone failure disables the whole network.
  • Performance degrades as traffic grows (more collisions).
  • Length and node limits (≈30 nodes for 10 Base‑2).
• Real‑world example – early university computer labs using thin coaxial cable.

2.2 Star Topology

• Structure – each node connects to a central device (hub, switch or router) with its own cable.
• Logical behaviour
  • Hub: logical bus (half‑duplex, collisions).
  • Switch: logical star (full‑duplex, separate collision domain per port).
• Typical use – modern Ethernet LANs, Wi‑Fi access‑point networks, office and school environments.
• Advantages
  • Failure of a single link affects only that node – easy fault isolation.
  • Scalable – add nodes by running a new cable to the hub/switch.
  • Higher aggregate bandwidth; simultaneous transmissions on separate links.
• Disadvantages
  • More cabling than a bus.
  • Central device is a single point of failure (mitigated with redundant switches or link aggregation).
• Real‑world example – a corporate office where each workstation plugs into a managed Gigabit switch.

2.3 Mesh Topology

• Structure – a dedicated point‑to‑point link between every pair of nodes (full mesh) or between selected nodes (partial mesh).
• Logical behaviour – multiple paths exist; routing algorithms (e.g. OSPF, BGP) select the optimal route.
• Typical use – Internet backbone, data‑centre interconnects, mission‑critical SCADA networks.
• Advantages
  • Highly reliable – redundant paths mean a single link failure rarely isolates a node.
  • Scalable bandwidth – many simultaneous transmissions.
• Disadvantages
  • Cost grows quadratically: L = n(n‑1)/2 links for n nodes.
  • Complex cabling, installation and management; requires sophisticated routing.
• Real‑world example – a regional ISP’s core network where each router is linked to every other router.

2.4 Hybrid Topology

• Definition – a combination of two or more basic topologies to meet specific performance, cost and reliability requirements.
• Common variants
  • Star‑bus – several star segments connected by a bus backbone; typical in large campuses.
  • Star‑mesh – core network is a mesh, edge devices connect via star; used in modern data‑centre “top‑of‑rack” designs.
• Advantages
  • Flexibility – topology can be tailored to the physical layout of a building.
  • Balanced cost and reliability – critical sections can be meshed, peripheral sections star‑wired.
• Disadvantages
  • Design, documentation and troubleshooting are more complex.
  • Overall cost depends on the mix of constituent topologies.
• Real‑world example – a university campus where each department uses a star LAN, all departments are linked by a high‑speed fiber bus.

2.5 Choosing a Topology – Decision Tree

1. Is the network small (< 30 devices) and cost‑sensitive? → Bus.
2. Do you need easy expansion, fault isolation and higher traffic capacity? → Star (use a managed switch).
3. Is maximum reliability and bandwidth essential (e.g., data‑centre, ISP backbone)? → Mesh (full or partial).
4. Does the site have distinct zones (different buildings, floors) with varying requirements? → Hybrid (combine star, bus, mesh as needed).

3 Supporting Concepts for the Syllabus

3.1 Networking Devices

Device	Primary Role	Typical Use in Topologies
Network Interface Card (NIC)	Provides physical/electrical connection to the network	Every node
Hub	Repeats incoming signals to all ports (half‑duplex)	Bus‑style star (legacy)
Switch	Intelligent frame forwarding; full‑duplex per port	Modern star, hybrid core
Router	Inter‑connects different networks; performs IP routing	WAN edge, Internet gateway
Bridge	Connects two LAN segments at the data‑link layer	Partial mesh, segmenting
Access Point (AP)	Provides wireless connectivity	Wireless star or hybrid
Firewall	Filters traffic based on security policies	Perimeter of LAN/WAN, between VLANs
Modem	Modulates/demodulates signals for transmission over telephone or cable lines	WAN entry point

3.2 LAN vs. WAN Comparison

Aspect	LAN	WAN
Geographic scope	Single building / campus	City, nation, global
Typical bandwidth	100 Mbps – 10 Gbps	1 Mbps – 1 Gbps (often shared)
Media	Ethernet (copper/fibre), Wi‑Fi	Leased lines, MPLS, satellite, public Internet
Management	Single organisation, direct control	Multiple providers, service‑level agreements
Addressing	Private IPv4/IPv6 ranges	Public IPv4/IPv6, NAT frequently used

3.3 Client‑Server vs. Peer‑to‑Peer

• Client‑Server – Centralised servers host resources. Example: A school’s file server stores documents; each workstation (client) accesses them via SMB.
• Peer‑to‑Peer (P2P) – Every node can both request and provide resources. Example: Students share lecture recordings via BitTorrent; each laptop acts as a seed and a leecher.
• When to use
  • Client‑server for controlled, secure environments and when data integrity is critical.
  • P2P for distributed content distribution, collaborative editing, or when a central server is impractical.

3.4 Thin‑Client vs. Thick‑Client

• Thin‑client – Minimal local processing; relies on a server (e.g., Virtual Desktop Infrastructure). Benefits: lower hardware cost, easier updates.
• Thick‑client – Full‑featured PC with local CPU, storage and applications. Benefits: high performance, works offline.

3.5 Cloud Computing and Topology Impact

• IaaS – Virtual machines, storage, networking are provided; the underlying data‑centre uses a high‑capacity mesh backbone.
• PaaS – Development platforms run on the same mesh infrastructure; internal LAN can remain simple.
• SaaS – End‑user applications accessed over the Internet; the local network often only needs a star or hybrid LAN to reach the ISP.
• Design implication – When most services are cloud‑based, a simple star LAN with a reliable WAN link is sufficient; critical internal services may still use a partial‑mesh core for redundancy.

3.6 Wired vs. Wireless Media

Media	Typical Speed	Range	Security	Cost
Twisted‑pair Ethernet (Cat5e/6/6a)	100 Mbps – 10 Gbps	Up to 100 m per segment	Physical security; optional MAC filtering	Low
Fiber‑optic (single‑mode)	10 Gbps – 100 Gbps+	km‑scale	Very secure (no EM leakage)	High
Wi‑Fi (802.11ac/ax)	600 Mbps – 10 Gbps (theoretical)	30‑100 m indoor	WPA2/WPA3, hidden SSID, 802.1X	Moderate
Bluetooth	1‑3 Mbps	≤10 m	Pairing, encryption (AES‑CCM)	Very low

3.7 Ethernet & CSMA/CD (Shared‑Medium Topologies)

• Frame format – preamble, destination MAC, source MAC, type/length, payload, CRC.
• CSMA/CD steps
  1. Listen to the medium.
  2. If idle, begin transmission.
  3. Continuously monitor for a collision.
  4. On collision, abort, send a jam signal, wait a random back‑off time, then retry.
• Modern switched Ethernet eliminates collisions; CSMA/CD is only relevant for legacy bus or hub‑based star networks.

3.8 Circuit‑Switching vs. Packet‑Switching

• Circuit‑switching – A dedicated path is established for the duration of a session (e.g., traditional telephone network). Guarantees bandwidth but is inefficient for bursty data.
• Packet‑switching – Data is divided into packets, each routed independently (e.g., Internet). Allows efficient use of bandwidth and supports many concurrent sessions.
• Most computer networks (LANs, WANs, the Internet) are packet‑switched; some specialised services (e.g., VoIP over MPLS) may reserve a virtual circuit for QoS.

3.9 Network Performance Metrics

Metric	Definition	Influence of Topology
Latency	Time for a packet to travel from source to destination	Fewer hops (e.g., mesh) generally reduce latency.
Bandwidth	Maximum data rate that a link can carry	Parallel links in a mesh increase aggregate bandwidth; a single bus limits it.
Jitter	Variation in latency between packets	Stable, dedicated paths (circuit‑switched or full‑mesh) minimise jitter.
Throughput	Actual data rate achieved	Depends on collisions (bus), congestion (star with hub), or load‑balancing (mesh).
Packet loss	Percentage of packets that never reach the destination	High collision rates in bus/hub, or link failures in partial mesh, increase loss.

3.10 IP Addressing Basics

• IPv4 – 32‑bit address, dotted‑decimal (e.g., 192.168.10.25). Subnet mask separates network and host portions.
• Subnetting example – 192.168.10.0/24 → network 192.168.10.0, host range 192.168.10.1‑254, broadcast 192.168.10.255.
• IPv6 – 128‑bit address, hexadecimal groups (e.g., 2001:0db8:85a3::8a2e:0370:7334). Uses prefix length (e.g., /64).
• Public vs. private IPv4 ranges
  • 10.0.0.0/8
  • 172.16.0.0/12
  • 192.168.0.0/16
• Static vs. dynamic allocation – static: manually assigned; dynamic: provided by DHCP server.

3.11 URL and DNS

• URL structure – protocol://hostname[:port]/path?query#fragment (e.g., https://www.example.com:443/articles?id=12#section3).
• DNS resolution process
  1. Client asks local resolver (often the ISP’s DNS server).
  2. If not cached, resolver queries a root server.
  3. Root server directs the query to the appropriate TLD server (.com, .org, etc.).
  4. TLD server points to the authoritative server for the domain.
  5. Authoritative server returns the IP address; resolver caches it and returns it to the client.

3.12 Security Threats & Counter‑Measures

Threat	Impact	Typical Counter‑measure
Phishing	Credential theft	User education, email filtering, SPF/DKIM
Malware (virus, ransomware)	Data loss or system compromise	Antivirus/anti‑malware, application whitelisting, regular patches
Denial‑of‑Service (DoS/DDoS)	Service unavailability	Firewalls with rate‑limiting, IDS/IPS, CDN/DDoS mitigation services
Man‑in‑the‑middle (MITM)	Data interception/modification	TLS/SSL, VPN (IPsec), certificate pinning
Eavesdropping on wireless	Unauthorised data capture	WPA3, strong pre‑shared keys, 802.1X authentication

4 Detailed Topology Review (Condensed)

4.1 Bus

Single backbone, shared medium, CSMA/CD, inexpensive but vulnerable to single‑point failure and collisions.

4.2 Star

Each node links to a central hub/switch; modern networks use switches for full‑duplex operation, offering fault isolation and scalability.

4.3 Mesh

Multiple point‑to‑point links; provides high reliability and bandwidth at a high cost and complexity.

4.4 Hybrid

Combines two or more basic topologies (e.g., star‑bus, star‑mesh) to balance cost, performance and reliability.

5 Comparison Table of Topologies

Topology	Typical Use	Advantages	Disadvantages	Scalability (nodes)	Cost (relative)
Bus	Small legacy LANs, temporary setups	Very low cabling cost; easy to add nodes	Backbone failure disables network; collisions limit performance	Up to ~30 nodes (10 Base‑2)	Very low
Star	Modern office, school, Wi‑Fi networks	Fault isolation; high bandwidth; easy expansion	More cabling; central device is a single point of failure (mitigated with redundancy)	Hundreds to thousands (limited by switch port count)	Low‑moderate
Mesh	Data‑centre core, ISP backbone, mission‑critical SCADA	Redundant paths; high reliability and aggregate bandwidth	Quadratic cost; complex installation and routing	Scales well for high‑performance cores; impractical for large edge networks	High
Hybrid	University campus, large corporate sites	Tailored to physical layout; balances cost and reliability	Design and troubleshooting more complex	Depends on the mix of constituent topologies	Variable (moderate‑high)

6 Key Take‑aways for the Cambridge AS & A Level Exam

• Be able to sketch and label each topology (bus, star, mesh, hybrid) and state whether it is a physical or logical description.
• Explain how CSMA/CD works and why it is only relevant to shared‑medium topologies.
• Compare the impact of topology on latency, bandwidth and reliability.
• Identify appropriate networking devices for each topology and justify their selection.
• Describe the client‑server and peer‑to‑peer models with concrete examples.
• Outline the TCP/IP layers and give one protocol that operates at each layer.
• Distinguish circuit‑switching from packet‑switching and relate the concepts to real‑world networks.
• List common security threats and the corresponding protective technologies (firewall, IDS/IPS, VPN, encryption).
• Perform a simple IPv4 subnetting calculation (e.g., determine the network address for 192.168.12.45/26).
• Explain the DNS resolution process using a step‑by‑step example.