Network topologies explained for CompTIA Network+

What is a network topology, and why does Network+ test it?

A network topology is the pattern that connects your devices and carries data between them. Network+ tests it because the shape of a network decides how fast it runs, how much it costs, and how it fails when something breaks. Get the shape wrong and every later decision, from where you run cable to how you chase down an outage, gets harder.

On the N10-009 exam, topologies sit in Domain 1.0, Networking Concepts, which CompTIA weights at 23% of the test. Objective 1.6 asks you to compare and contrast network topologies, architectures, and types. The shapes it names directly are star or hub-and-spoke, mesh, hybrid, spine-and-leaf, and point-to-point, plus the three-tier hierarchical model and the collapsed core design.

Before we walk through each one, there is a distinction that is easy to trip on: the difference between how a network is wired and how its data actually moves. That is physical versus logical topology, and it is the right place to start.

Physical topology vs logical topology: what is the difference?

Physical topology is where the cables actually run. Logical topology is the path your data takes once it is on the wire, and the two do not have to match. A single box can be a physical star while behaving like a logical bus.

The classic example is old Ethernet. Workstations plugged into a hub looked like a star, one cable per device back to a central box. But a hub repeated every signal out every port, so logically all of those devices shared one line and one collision domain, which is a bus. Token Ring did the reverse: devices cabled into a central unit in a physical star, while the data passed device to device in a logical ring.

Modern switched networks are physical stars, and switching plus VLANs shape the logical paths on top. If the OSI layers are still fuzzy, the OSI model in plain English pairs well with this, because logical topology is really a Layer 2 and Layer 3 idea. And the way VLANs work is the clearest modern case of a logical topology that ignores the physical wiring entirely.

Star, mesh, and point-to-point: the topologies you will actually build

Star is the default LAN today, mesh buys you fault tolerance, and point-to-point is the simplest link there is. These three cover most of what you will actually wire in the field.

In a star, or hub-and-spoke, every node connects back to one central device, normally a switch. It is easy to manage and easy to grow: add a node, run one cable. The trade-off is fault tolerance. A single cable fault isolates only that one node, which is good, but the central switch is a single point of failure, so if it dies the whole segment goes down. Star is a LAN topology with a clear advantage: one central point makes it convenient to manage.

Mesh connects nodes directly to each other. A full mesh links every node to every other node, which gives you the most redundant paths and the most fault tolerance, but the cabling and cost climb quickly as you add nodes. A partial mesh adds redundant links only where they matter. You see mesh in WAN cores and in wireless mesh networks, where alternate paths keep traffic moving when one link fails.

Point-to-point is a single dedicated link between two endpoints, like a leased line between two offices or a wireless bridge across a campus. There is nothing to share and nothing to contend with, which makes it simple and predictable.

Bus and ring: the legacy shapes still worth knowing

Bus and ring are mostly history, but they teach failure modes worth understanding. N10-009 objective 1.6 focuses on star, mesh, hybrid, spine-and-leaf, and point-to-point, so treat bus and ring as foundational background rather than the headline.

In a bus topology, every device taps into one shared backbone cable that is terminated at both ends. It was cheap and used little cable, but the weaknesses are brutal: one break in the backbone, or one missing terminator, and the whole segment goes down. Only one device could transmit at a time, and collisions were managed with CSMA/CD.

In a ring, each device connects to exactly two neighbors, forming a loop that data travels around. A single break can split the ring unless it is a dual-ring design like FDDI, which can wrap and heal itself. Token Ring passed a small token around the loop to decide who was allowed to transmit, which avoided collisions but added its own complexity. Both shapes lost out to switched star networks, which are cheaper to run and far easier to troubleshoot.

Hybrid, spine-and-leaf, and the three-tier model: how real networks scale

Real networks rarely use one pure shape. Hybrid mixes topologies together, the three-tier model organizes campus networks into layers, and spine-and-leaf is the fabric modern data centers run on.

A hybrid topology is any deliberate mix, and it describes most enterprise networks. A common form is a tree, or star-of-stars, where several access switches each form a star and all connect up to a distribution switch. That layering is formalized in the three-tier hierarchical model: an access layer that connects endpoints, a distribution or aggregation layer that handles routing and policy between segments, and a high-speed core that ties it all together. Smaller sites often use a collapsed core, which merges the core and distribution layers into one to save cost.

Spine-and-leaf is the modern data-center answer. Every leaf switch connects to every spine switch, and leaves never connect to other leaves, nor spines to spines. The payoff is that any server is the same number of hops from any other server, so latency stays low and predictable. That matters because data-center traffic is now mostly east-west, meaning server to server, rather than the north-south, client to server, pattern the three-tier model was built around.

• Star or hub-and-spoke: central switch, easy to manage, one switch failure takes the segment down. The default LAN.
• Full mesh: every node linked to every other, maximum fault tolerance, cost grows fast. Seen in WAN cores and wireless mesh.
• Partial mesh: redundant links only where they matter, a practical middle ground.
• Point-to-point: one dedicated link between two endpoints, simple and predictable.
• Bus: one shared backbone, cheap, a single break kills the segment. Legacy.
• Ring: each node linked to two neighbors, a break can split it unless it is dual-ring. Legacy.
• Hybrid: a mix of the above, which is what most real networks actually are.
• Spine-and-leaf: every leaf to every spine, low predictable latency, built for east-west data-center traffic.
• Three-tier (core, distribution, access): the layered campus model, with a collapsed core for smaller sites.

How to make network topologies stick

The fastest way to own this objective is to stop memorizing definitions and start drawing. Sketch each topology, then break it: erase the center of a star and watch the segment die, cut one link in a full mesh and see that everyone still has a path. Pair each shape with its fault tolerance and its typical use, because that is exactly how the exam frames its compare-and-contrast questions. When you are ready to drill it with quick reps and feedback, run through it in Study mode until star, mesh, spine-and-leaf, and the three-tier model feel obvious. Topologies are foundational, and once they click, cabling, VLANs, and troubleshooting all get easier.