Theory
The shape of a network
When you connect several devices, you have a choice about the shape: do they all hang off one cable, form a loop, or all plug into a central box? This arrangement is the network's topology, and it is not just cosmetic. The shape decides how much cabling you need, how fast data flows, and, crucially, what happens when a cable or device fails.
There are five classic topologies: bus, ring, star, mesh, and tree. This lesson compares them, because choosing a topology is really about trading off cost, performance, and reliability.
At a glance
| Topology | Shape | Key trade-off |
|---|---|---|
| Bus | All devices share one backbone cable | Cheap and simple, but the whole network dies if the backbone breaks |
| Ring | Each device links to two neighbours, forming a loop | Orderly data flow, but a single break can disrupt the ring |
| Star | All devices connect to a central hub or switch | One cable failing loses only that device, but the central device is a single point of failure |
| Mesh | Every device connects to many or all others | Very reliable (multiple paths), but expensive (many links) |
| Tree | A hierarchy of stars connected together | Scalable for large networks, but depends on the top-level nodes |
Theory
Why star dominates today
In modern wired networks, the star topology is by far the most common. Every device connects by its own cable to a central switch. This is what a typical office or lab looks like.
The appeal is fault isolation: if one device's cable fails, only that device drops off, everyone else keeps working, unlike a bus, where a break in the shared backbone takes down the whole network. The catch is that the central switch is a single point of failure: if it dies, the whole star goes with it. But switches are reliable and easy to replace, so the trade is well worth it, which is why the star won.
Formula
The real question: what happens when something fails?
The clearest way to compare topologies is to ask: if one link or device fails, how much breaks?
In a bus, a break in the shared backbone can take down everything. In a star, a failed cable loses just one device, but a failed central switch loses all. In a mesh, because there are many paths between devices, one failed link barely matters, traffic simply reroutes, which is why mesh is the most fault-tolerant (and the most expensive). Reliability versus cost is the core tension, and 'what fails when a link dies' is how you weigh it.
Quiz
In which topology does a single break in the shared central cable bring down the entire network?
- Star, because everything depends on cables
- Bus, because all devices share one backbone cable, so breaking it disconnects everyone
- Mesh, because it has the most links
- Ring, because it has no cables
Show the answer
Bus, because all devices share one backbone cable, so breaking it disconnects everyone
In a bus topology, every device shares a single backbone cable, so a break in that backbone splits or disconnects the whole network, this shared-cable vulnerability is the bus's defining weakness. Option A is wrong: in a star, each device has its OWN cable to the central switch, so one broken cable loses only that device, not the whole network (the central switch, not a shared cable, is the star's weak point). Option C is the opposite of the truth: mesh has MANY redundant links, so one break barely matters, it is the most fault-tolerant, not the most fragile. Option D is wrong: a ring absolutely uses links between neighbours; a single break can disrupt it, but the 'shared central cable' described is specifically the bus. Match the failure to the shape: one backbone, total failure means bus.
Think first
If mesh is the most reliable, why is not everything a mesh?
Mesh survives failures beautifully with its many paths. So why do most networks use star instead? Then tap.
Show the answer
Because mesh's reliability comes at a steep COST in cabling and complexity that most networks do not need. In a full mesh, every device connects directly to every other device, so the number of links grows very fast: connecting just 10 devices in a full mesh needs 45 separate links (each of the 10 connected to the other 9, divided by 2), and 100 devices would need thousands. That is a huge amount of cable, ports, and configuration, expensive to build and to maintain. For an ordinary office or lab, this is massive overkill: a star gives you good-enough reliability (one cable failure loses only one device) at a fraction of the cost, with simple wiring, one cable per device to a central switch. So mesh is reserved for places where failure is truly unacceptable and budget allows, backbone links between core routers, critical infrastructure, parts of the internet's core, where the many redundant paths are worth every rupee. Everywhere else, the star's balance of low cost, easy management, and adequate fault isolation wins. It is the classic engineering trade-off: you buy exactly as much reliability as the situation justifies, and for most, star is the sweet spot. Reliability is not free; you pay for it in links.
Summary
Key takeaways
- A topology is the shape of a network: how devices are wired together, which affects cost, speed, and fault tolerance.
- Bus: all devices share one backbone cable, cheap but a backbone break kills the whole network.
- Ring: each device links to two neighbours in a loop; a break can disrupt the ring.
- Star: all devices connect to a central switch; a failed cable loses one device, but the switch is a single point of failure.
- Mesh: every device links to many others, very reliable via multiple paths, but expensive in links.
- Tree: a hierarchy of stars, scalable for large networks; star is the dominant modern topology.
- Memory hook: to compare topologies, ask what breaks when one link or device fails.