Network topologies
A topology is the arrangement of devices and the connections between them. AQA focuses on two key topologies: star and mesh (full and partial).
Star topology
All devices connect to a central node — typically a switch or router. Every device communicates by sending data through the centre.
Device A
|
Device B — Switch — Device C
|
Device D
Pros
- Easy to add devices — plug into the switch.
- Failure of one device doesn't affect others — they each have an independent link.
- Simpler to manage — central point to monitor.
- Good performance — modern switches give each device dedicated bandwidth.
Cons
- Central failure brings down the network — if the switch fails, no device can talk to any other.
- Cost of the central device — bigger switches for many devices.
- Cable per device — a lot of cabling for large networks.
Used in: most home and office LANs.
Mesh topology
Every device is connected to many (full) or several (partial) other devices, with multiple paths between any two endpoints.
A — B
|\ /|
| X |
|/ \|
D — C
Full mesh
Every node connects to every other. For n nodes, you have n(n-1)/2 links — grows fast.
Partial mesh
Each node connects to several others, not necessarily all.
Pros
- High redundancy — multiple paths between any two devices. If one link fails, traffic re-routes.
- High reliability — no single point of failure.
- Scalability for large networks (especially partial mesh).
Cons
- Expensive — many cables/connections.
- Complex — routing decisions, more equipment.
- Hard to install — full mesh impractical beyond small networks.
Used in: telecoms backbones, the internet itself, mission-critical military and emergency networks, mesh Wi-Fi systems.
Comparison
| Feature | Star | Mesh |
|---|---|---|
| Layout | All connect to central | Devices connect to each other |
| Single point of failure | Yes (central) | No (multiple paths) |
| Cost | Lower (one switch + cables) | Higher (many links) |
| Reliability | Vulnerable centrally | Highly reliable |
| Scalability | Up to switch capacity | Excellent (with partial mesh) |
| Common use | LANs | Backbones, internet, IoT mesh |
✦Worked example
A school computer room with 30 PCs and a printer. Recommend a topology.
- Star — connect all PCs and printer to a switch B1.
- Cheap, easy to manage, performance per device good B1.
- If a PC fails, others continue B1.
- Single point of failure: the switch — but a school will tolerate occasional downtime more readily than a backbone B1.
When mesh wins
A police communications network must keep working when nodes go down or links are damaged. Mesh ensures every officer's radio can find an alternative path. The cost is justified by the criticality.
⚠Common mistakes— Pitfalls
- Calling the internet a "star". It's a vast partial mesh of routers.
- Drawing star without a central device. The switch/hub is the central node.
- Confusing a "star" with a tree. A tree is a hierarchical extension.
- Saying "mesh has no failure points". Mesh has no single point of failure — multiple simultaneous failures still hurt.
- Underestimating mesh cost. Full mesh of 10 nodes = 45 links — expensive.
Other topologies (extension)
- Bus — all devices on one shared cable. Cheap but failures cascade. Obsolete on LANs.
- Ring — devices in a loop, data passes round. Used in some industrial networks.
- Tree / hierarchical — stars connected to a higher star (a school's computer rooms each starring to a backbone switch).
GCSE focus is star and mesh.
➜Try this— Quick check
For each, choose star or mesh:
- A 4-PC home network: star (one router).
- A backbone connecting 5 city offices needing redundancy: mesh (partial).
- A small library's 12 PCs: star.
- The Internet: (partial) mesh of routers.
AI-generated · claude-opus-4-7 · v3-deep-computer-science