Network Topologies: Star, Tree, Bus, Ring and Mesh Explained

What is a network topology and why does it matter?

A network topology describes how devices and links are arranged and, in a logical sense, how traffic relationships are organized. Topology affects failure domains, redundancy, scalability, cabling, operational complexity, and sometimes performance.

Real LAN

Most modern Ethernet LANs look like a star at the access layer: endpoints connect to switches, and switches connect upward into a hierarchy.

Scale

As a network grows, simple stars become trees or hierarchical fabrics. Redundant links may create partial-mesh characteristics between key devices.

Internet

The Internet is not one clean topology. It is a network of networks with many interconnections, policies, and redundant paths.

First distinction: physical vs logical topology

Physical topology

The physical topology is what you can trace through cables, fiber, radio links, racks, and ports. It answers questions such as which switch a host connects to and which uplinks join buildings.

Logical topology

The logical topology describes traffic relationships and segmentation. VLANs, routing adjacencies, overlays, VPNs, and virtual networks can create logical paths that differ significantly from the physical wiring.

A useful network diagram often needs both views.

Topology and performance: what really changes?

Collisions

Classic shared Ethernet topologies had real collision-domain implications. Modern switched full-duplex Ethernet isolates links, so the more relevant concerns are congestion, oversubscription, and failure domains.

Shared medium such as Wi-Fi

Wireless clients share airtime, so adding clients and overlapping cells affects capacity in ways that differ from dedicated switched Ethernet links.

Single points of failure

A central device simplifies management but can also become a single point of failure. Redundancy reduces this risk at the cost of more links, protocols, and operational complexity.

Scalability

Hierarchical designs scale better than connecting every device directly to every other device. Aggregation points keep cabling, routing, and policy manageable.

Quick mental checklist

• What happens if the central device fails?
• Can the design grow without rewiring everything?
• Are there loops, and which protocol controls them?
• Where are the bandwidth bottlenecks?
• Does the logical segmentation match the physical design?

1. Star topology: the modern LAN default

In a star, endpoints connect to a central device, typically an Ethernet switch or wireless access point. Modern office LANs are usually built from multiple stars.

How does it work?

Each endpoint has its own access link to the central device. A failure of one endpoint cable usually affects only that endpoint.

Strengths and weaknesses

✅ Main advantage: isolation

Individual access-link faults are easy to isolate and troubleshoot, and switched Ethernet provides dedicated collision domains.

❌ Main weakness

The central switch is critical. If it fails and there is no redundancy, every attached endpoint loses connectivity.

2. Tree topology: the expanded star

What is it?

A tree connects several stars in layers: access switches connect to distribution devices, which may connect to a core. This is common in multi-floor and campus networks.

Why is it used?

Hierarchy lets the network scale while keeping roles and traffic aggregation understandable. Redundant uplinks can be added where availability requirements justify them.

3. Bus topology: the old shared-medium model

How did it work?

Devices shared one common transmission medium. Classic coaxial Ethernet is the familiar example.

The bus problem

The shared medium created one large collision domain, faults were harder to isolate, and physical problems could affect many users. Switched star Ethernet replaced this model in modern LANs.

4. Ring topology: controlled paths and resilient variants

In a ring, each node or network segment connects to two neighbors, forming a closed path. Historical token-ring technologies used this idea directly, while modern transport and industrial systems may use ring structures for predictable redundancy.

The loop that keeps traffic moving

A ring can offer an alternate direction when one path fails if the protocol and equipment support fast protection. Without appropriate control, however, Layer 2 loops can be destructive.

5. Mesh topology: many possible paths

Full connectivity

In a full mesh, every node connects directly to every other node. A partial mesh connects only selected pairs and is much more common at scale.

Why not fully mesh every office?

The number of links grows rapidly as nodes are added. Cabling, ports, routing adjacencies, cost, and troubleshooting become difficult. Mesh is therefore most useful where redundancy justifies the complexity.

How to choose the right topology

Home or small office

A simple star around one or two switches and wireless access points is usually enough.

Several floors or a campus

A hierarchical tree with access and distribution roles provides cleaner scaling and fault isolation.

High availability

Add redundant uplinks and devices selectively. The goal is not maximum connectivity; it is removing unacceptable single points of failure without creating unmanageable complexity.

Quick summary: which one should I choose?

Conclusion: from cables to architecture

Topology is useful because it turns a list of devices into a structure. Choose a topology based on failure impact, growth, capacity, physical constraints, and operational simplicity. Then layer VLANs, IP addressing, routing, and security on top of that physical foundation.

FAQs about network topologies