Network Topologies Explained: Star, Bus, Ring, Mesh & Hybrid [N10-009 Guide]

Part 3 of our N10-009 study path: every physical network topology drawn, compared, and mapped against cost, scalability, and fault tolerance.

What is a network topology? Just as a topographical map is a type of map that shows the shape of the terrain, the physical topology of a network is a map of the network itself — it defines exactly where every workstation, server, and device is located, and the precise arrangement of all the physical media, such as cables, that connect them. Learn to read that map and every network you ever touch — from a two-PC peer-to-peer setup to a multi-branch enterprise WAN — becomes a diagram you can troubleshoot, defend, and redesign on demand.

This is Part 3 of our complete N10-009 study series and Part 2 of 3 of Chapter 1. In Part 002 you learned what a network is and what it's built from; now we arrange those components into the seven physical topologies the exam expects you to recognize — bus, star, ring, mesh, point-to-point, point-to-multipoint, and hybrid — and grade each one against the three criteria that decide every real-world design: cost, scalability, and fault tolerance. Master these and Part 004's backbones and segments will feel like a natural next step instead of new vocabulary.

I'm Mostafa Amaan, and I've spent over 16 years cabling exactly these shapes into existence — running structured cabling and wireless networks through office ceilings, administering point-to-multipoint WAN links across multi-branch enterprise environments, and managing the servers and imaging systems those topologies exist to reach. In this guide I'll translate the official CompTIA N10-009 Chapter 1 objectives into plain language, redraw the classic topology diagrams as original artwork, and flag every exam trap before it catches you.

Quick Answer: What Is a Network Topology?

Network Topologies Explained: Physical vs. Logical

Now that we've defined the term, let's expand it. A network's topology refers to exactly where every workstation, server, and device is located, and the precise arrangement of all the physical media, such as cables, that connect them. When you look at any network diagram — whether it's a floor plan of an office or a rack elevation in a data center — you are looking at a topology. And the N10-009 exam expects you to do more than name the shapes: it expects you to know why a designer would choose one over another, and what breaks when a cable in that shape fails.

CompTIA splits the discussion into two viewpoints that beginners constantly confuse, so we'll separate them before drawing a single diagram:

What Is a Physical Topology?

As the name implies, physical topology is the physical layout of the network — how the cables and devices are actually arranged and wired together. In other words, the physical topology is the map of the network itself. It is hardware-oriented: if you could see through the ceiling tiles and under the raised floor, the pattern of cables you observed would be the physical topology. The seven shapes we cover in this article — bus, star, ring, mesh, point-to-point, point-to-multipoint, and hybrid — are all physical topologies.

What Is a Logical Topology?

Logical topology, by contrast, is the arrangement of how the data flows through the network. It is software-oriented: a network's data path may not match its physical wiring at all. The classic example — and a favorite exam question — is Token Ring over a star-wired physical layout: the cables run in a star to a central MAU/switch, but the logical data path circulates as a ring, passing from station to station in turn. Two networks can share one physical topology and behave completely differently because their logical topologies differ.

The table below summarizes the distinction — memorize it, because the exam loves to swap the two terms:

The Bus Topology

Now let's walk through each topology in turn, starting with the simplest shape ever built. A bus topology consists of a single cable (called the bus or backbone) to which every device on the network connects. Only one device can send at a time — when one station transmits, its signal travels down the cable in both directions and every other station hears it, but only the addressed recipient keeps the data. This "everyone hears everything" behavior is the defining characteristic of the bus: it's a shared medium in the truest sense.

If you look at a bus network diagram, every workstation hangs off that one central line — imagine clothes pegged onto a single washing line. Both ends of the cable are capped with terminators, small resistors whose entire job is to absorb the signal when it reaches the end of the cable. Without terminators, the signal would bounce back down the wire (an "echo"), corrupting live traffic — which is why a missing or failed terminator could take down an entire bus segment.

Figure 1.9: In a bus topology, a single cable — the bus — connects every device. Terminators at both ends absorb the signal so it cannot echo back down the wire.

Bus Topology Variations

The bus concept comes in two flavors the exam may ask you to distinguish. In a linear bus, all of the network's nodes connect directly to one common half-duplex link — the single cable runs from one terminator to the other, and every workstation taps directly into it. A linear bus may have minimal cabling compared with other topologies, but it also carries a single point of failure: any break in the main cable splits the network into two isolated islands and, worse, the unterminated ends reflect signals until the whole segment stops working.

In a distributed bus, the nodes are connected to multiple branching links branching out from the common (half-duplex) link — several bus segments are joined together so the "trunk" reaches more areas of the building. Legacy 10BASE2 (thinnet) and 10BASE5 (thicknet) Ethernet networks were the textbook examples of both forms; you will rarely meet a pure bus in production today, but you must still recognize the shape because the exam draws it constantly.

So how does the bus score against the three evaluation criteria — cost, scalability, and fault tolerance? The table below gives the answer the exam expects:

The Star Topology

Next is the shape you are almost certainly sitting inside right now. A star topology is a physical topology in which every device on the network connects to a central device — and remember this next line, because the exam does: in a star topology, the central device is a hub or a switch. Every cable radiates from that central point out to an individual workstation, server, or printer, exactly like spokes from the center of a wheel. No device ever talks directly to another device's cable — everything passes through the center.

The behavior at the center is what separates the two implementations. When a star topology's central device is a hub, the hub simply repeats every incoming signal out of every other port — electrically, the network still behaves like a shared bus folded into a box, and collisions remain possible. When the central device is a switch, each port receives only the frames addressed to it, collisions disappear, and every link becomes dedicated bandwidth. Same physical shape, very different performance — the physical topology describes the wiring, and the center's intelligence describes the experience.

Figure 1.10: In a star topology, every device connects to a central device — a hub or a switch. Each cable serves exactly one station, radiating from the center like wheel spokes.

Star Topology Variations

Because every star is built around one central device serving many endpoints, the star's variations are really point-to-multipoint configurations — the difference is what sits at the center. In a star topology whose central device is a hub, the hub-attached links form the star: every port repeats the signal to every other port, and the total available bandwidth is shared among all attached devices. This was the common wiring closet of 10BASE-T Ethernet in the 1990s, and it survives in exam diagrams to this day.

In a star topology whose central device is a Wireless Access Point (WAP), the radio links-attached devices form the star: every laptop, phone, and tablet communicates with the WAP's radio directly, and the WAP bridges that wireless star onto the wired network behind it. Wi-Fi has made this the most common star variation on the planet — and it is also a perfect example of a point-to-multipoint topology implemented over radio instead of copper. Both variations share the same strengths and the same critical weakness: the central device is a single point of failure for everything attached to it.

Here is how the star scores against the three evaluation criteria — and why it became the default choice for modern office networks:

The Ring Topology

Third on the list is the shape that gave us token passing. In a ring topology, each device is connected to exactly two neighboring devices, forming a closed loop: data travels from device to device around the ring, passing through every intermediate station until it reaches its destination. Many rings pass data in one direction only — which is why a ring network diagram usually shows arrows pointing one way around the loop. If you trace the path of the data in a ring, you have also traced the path of a collision-free access method called token passing.

Token passing solves the "who talks next?" problem elegantly: a special frame called the token circulates around the ring, and a device can transmit only when it is holding the token. No two devices ever transmit at once, so collisions never occur — a completely different world from the contention-based bus. When the transmitting device finishes, it releases the token for the next station. The cost of that orderliness is latency: even an idle ring forces data to hop through every intermediate device between sender and receiver, and one broken connection can stop the entire ring unless the hardware supports a ring-recovery mechanism (like Token Ring's beaconing and wrapback, or FDDI's dual counter-rotating rings).

Figure 1.11: In a ring topology, each device connects to exactly two neighbors, forming a closed loop. A token circulates around the ring — a device transmits only while holding it.

Here is the ring's scorecard against the three evaluation criteria:

The Mesh Topology

Fourth is the shape that buys reliability at any price. In a mesh topology, every device is connected to every other device by its own dedicated cable — there is no shared backbone, no central hub, and no token to wait for. Any station can reach any other station directly, which makes the mesh the most fault-tolerant topology that exists: no single cable break can isolate a device, because alternate paths always exist.

That resilience comes with a mathematical price tag the exam expects you to calculate. The number of links a mesh requires grows quadratically with the number of devices:

Figure 1.12: In a mesh topology, every device connects to every other device with its own dedicated cable — maximum fault tolerance, maximum cabling cost.

Here is the mesh's scorecard — and you can already guess which column wins and which one loses:

Because full meshes are rarely affordable, real deployments use partial meshes — critical paths (say, between core routers in a data center or between branch offices) are fully redundant while edge devices keep single connections. On the exam, the word "mesh" with a diagram showing only some device-to-device links means a partial mesh; every-device-to-every-device means full mesh.

The Point-to-Point Topology

Fifth is the simplest possible connection — the shape every other topology is made of. A point-to-point topology consists of two devices connected directly by a single cable or wireless link: one transmitter, one receiver, nothing in between. No other device shares the link, so the entire bandwidth of that connection belongs to the two endpoints — the fastest, most private, and most predictable arrangement possible.

You already own several point-to-point links without thinking about it: the cable from your PC to the wall port, a serial link between two routers in a lab, a microwave dish aimed at another building, or the Bluetooth pairing between your phone and your earbuds. In larger networks, point-to-point links are the mortar between the bricks — WAN connections between branch offices and the uplinks between switches in a data center are all point-to-point at heart. Notice also that two devices cabled directly together form the smallest complete network possible: add a shared resource and Part 002's peer-to-peer definition is satisfied.

Figure 1.13: A point-to-point topology connects two devices directly with a single link — the entire bandwidth belongs to the two endpoints.

The Point-to-Multipoint Topology

Sixth is the one-device-to-many shape. In a point-to-multipoint topology, a single central device communicates with several other devices over shared links — one transmitter, many receivers. Every endpoint hears the central device, and the central device coordinates everything. The connection is the star's variation again, viewed from a different angle: a point-to-multipoint topology's central device is typically a hub, a WAP, or even just an antenna serving many stations at once.

The classic example you already know from the star section: a Wireless Access Point whose radio links serve every laptop, phone, and tablet in the building is point-to-multipoint in action. The same shape appears at WAN scale — a provider's single fiber or antenna serving a whole neighborhood, one satellite uplink serving thousands of receivers, or one cable-TV plant serving an entire street. Point-to-multipoint is also how many modern wireless ISPs and campus Wi-Fi deployments backhaul traffic: one strong central radio, many subscriber endpoints.

Figure 1.14: A point-to-multipoint topology links one central device — a hub, WAP, or antenna — to many endpoints. A wireless access point serving every laptop in a building is the classic example.

The Hybrid Topology

Seventh — and the one you'll actually meet in the real world — is the hybrid topology: a network that combines two or more of the topologies above into a single design. Almost no production network is a pure bus, ring, or mesh anymore. Instead, designers combine shapes to get the best trade-off of cost, scalability, and fault tolerance for each part of the network.

The most familiar example is the star-bus hybrid you've been using all along: individual workgroups are wired as stars to their own switches (the star part), and those switches are chained or trunked together into a shared backbone (the bus part). A star-ring hybrid is the Token Ring network we discussed in the logical-topology section — star wiring, ring data flow. Data centers push this further with a partial-mesh backbone feeding star-wired server racks. When the exam shows a diagram that isn't any single pure shape, don't panic: name the shapes it's built from and call it a hybrid.

Figure 1.15: A hybrid topology combines two or more base topologies — here, star-wired workgroups joined by a shared backbone. Most real-world enterprise networks are hybrids.

The beauty of the hybrid is that it lets you apply each shape where it pays off. Use stars where flexibility and cheap scaling matter (workstation access), a mesh where uptime is non-negotiable (core links), and a bus or ring where a simple shared path is good enough (legacy segments). The exam's favorite hybrid question is simply: "Which topology combines two or more different topologies?" — and the answer is always the one with the word hybrid in it.

Network Topology Comparison: Choosing by Cost, Scalability & Fault Tolerance

You've now seen all seven physical topologies. The N10-009 exam rarely asks you to merely name them — it asks you to evaluate them against cost, scalability, and fault tolerance, and to recommend one for a scenario. Here is the master comparison table to study the night before your exam:

* A star's fault tolerance is high for individual device links, but its central hub/switch is a single point of failure (SPOF) for the whole star.

Where You'll Actually Meet Each Topology

Textbook diagrams are clean; production networks are messy. Here's how the seven topologies show up in the real environments you'll be hired to run — context the exam's scenario questions love to test:

The Modern Office LAN: Stars Everywhere

Walk into any office built in the last twenty years and you're standing inside a star-bus hybrid. Each desk is a point-to-point run back to a wiring-closet switch (a star), and each switch uplinks to a core switch over a trunk (the bus or backbone). Wireless devices attach through a WAP — a point-to-multipoint star over radio. This is why "star" is the answer to most "design an office LAN" questions: cheap to scale, easy to troubleshoot, one cable per device.

The Data Center: Meshed Cores, Starred Edges

In a data center, uptime is money, so the core is where mesh earns its keep: spine-and-leaf and partial-mesh designs run multiple links between core switches so a single failed link or device is invisible to traffic. The server racks themselves remain star-wired to their top-of-rack switches. This partial-mesh-over-star pattern is the hybrid that keeps banks, hospitals, and cloud providers online.

The WAN: Point-to-Point and Point-to-Multipoint

Between buildings and cities, you'll find point-to-point links (a dedicated fiber or microwave shot between two routers) and point-to-multipoint links (one provider antenna or fiber serving many subscribers). When a provider offers "MPLS" or "metro Ethernet," you're buying logically point-to-point connections stitched together inside the carrier's own mesh — the label switching you met in Part 002.

CompTIA Network+ N10-009 PBQ Strategy: Topology Selection Under Exam Constraints

On the real CompTIA Network+ N10-009 exam, topology concepts are not tested solely through definitions. In the high-weight Performance-Based Questions (PBQs) at the very beginning of the exam, you are handed a visual network diagram alongside a list of operational requirements and strict budget limitations. You must drag-and-drop the correct cabling connections, determine necessary hardware interfaces, or select the best topology for each functional tier of an organization.

The PBQ Constraint Triad: How CompTIA Tests Topology Selection

Every exam scenario forces you to balance three competing design forces: Fault Tolerance (Redundancy), Deployment Cost (Cabling & Port Density), and Scalability. The exam writers design scenarios specifically to test whether you can recognize the engineering breaking point of each shape:

• Step 1: Check the Single Point of Failure (SPOF) Requirement: Does the prompt explicitly state that "the failure of any single cable or intermediary device must never interrupt service to critical nodes"? If yes, pure star, bus, and single-ring designs are instantly eliminated. Your shortlist is narrowed strictly to Full Mesh, Partial Mesh, or a dual-ring setup.
• Step 2: Run the Mesh Feasibility Math: If the scenario involves more than 6 to 8 devices, calculate the link requirement using n(n − 1) / 2. For 10 routers, you need 45 physical cable runs and 9 spare interfaces on every single router. If the prompt specifies a "moderate budget" or "limited interface modules," a full mesh is the classic distractor answer. The correct engineering choice is a Partial Mesh that provides redundant uplinks between critical core nodes while leaving edge connections direct.
• Step 3: Analyze Node Failure Isolation vs. Central Infrastructure: If the requirement specifies that "an issue with a user's local network cable must not cause packet disruption for adjacent coworkers," the answer is a Star Topology with a centralized switch. Never select a Bus or Ring when localized fault isolation is the primary objective.

Worked PBQ Scenario: Enterprise Hospital Imaging Infrastructure

To see how these principles translate under exam conditions, study this authentic scenario adapted from enterprise healthcare infrastructure:

Engineering Analysis & Solution:

• Radiology Server Farm (6 Nodes): Because uptime is absolute and node count is low (n = 6), apply the formula: 6 × 5 / 2 = 15 links. Dedicated dual-port 10GbE NICs in a Full Mesh topology provide direct, point-to-point data paths between all 6 servers with zero reliance on an external central switch. If any cable fails, traffic instantly switches to alternate direct routes.
• Nursing Workstations (40 Nodes): Running a mesh across 40 desks would require 40 × 39 / 2 = 780 cables—a logistical and budgetary impossibility. Deploying a Switched Star Topology (using Cat 6A drops to a distribution switch stack) fulfills the fault-isolation mandate: if Nurse Station 12's cable is damaged, only Station 12 goes down, while all remaining 39 stations operate uninterrupted.
• The Unified PBQ Solution: The complete architecture is a Star-Mesh Hybrid. Access switches gather the nursing stars and uplink via redundant trunk pairs directly into the meshed PACS core.

Exam-Day Scenario Decision Matrix: Rapid Topology Matching

When time is ticking down on your 90-minute exam timer, use this rapid constraint matrix to verify your topology selection before submitting your answers:

Quick Knowledge Check: Interactive N10-009 Topology Challenge

Before moving to Part 004, test your retention with these five exam-style multiple-choice questions. Select your answer for each question below; once all questions are answered, click View Result to check your score and detailed explanations.

Practical Challenge: Map Your Own Network's Topology Tonight

The fastest way to lock in the seven topologies is to find them hiding in the network you're already using. Pull out a sheet of paper and complete this three-step challenge:

1. Draw your home LAN as a star. Put your router in the center circle and sketch every device you can think of around it — laptops, phones, printer, smart TV, game console. Count how many use Wi-Fi (point-to-multipoint off the router's WAP) versus a cable (point-to-point runs).
2. Mark the single point of failure. Circle the device that, if it died, would take your whole network offline. That circle is the star's weakness, and it's the same weakness your ISP's point-to-multipoint antenna share has.
3. Find the hybrid. Below your home star, sketch your ISP's side: your router connects over one point-to-point link to the provider, whose network is a mesh of routers. Two shapes, one picture — that is the hybrid, and it's the exact diagram an exam scenario is built from.

Keep that drawing. When Part 004 introduces backbones and segments, you'll redraw the same network with the trunk — the backbone — called out separately, and you'll see the star-bus hybrid appear in your own handwriting.

Frequently Asked Questions About Network Topologies

These are the questions readers and new hires ask me most often about network topologies — answered the way I'd explain them to a junior admin on their first week:

You Can Now Read Any Network Diagram — Here's How to Lock It In

You can now answer the question this article set out to resolve — what is a network topology — with the precision the N10-009 exam demands: a physical topology maps where every device sits and how the media connects them, distinct from the logical topology of how data flows. You can name and grade all seven — bus, star, ring, mesh, point-to-point, point-to-multipoint, and hybrid — against cost, scalability, and fault tolerance, and you can spot the star-bus hybrid hiding in every office network you enter. Here's how to make it stick:

1. Recite the three fault-tolerance rules before you sleep: bus = terminators at both ends (one cable, no redundancy); star = one cable per device (center is the SPOF); mesh = n(n−1)/2 links (most tolerant, most costly). Those three sentences answer most topology questions.
2. Memorize the hybrid sentence: "A hybrid combines two or more topologies" — and practice naming the shapes in any mixed diagram you can find online.
3. Do the drawing challenge above — your home star, its single point of failure, and your ISP's mesh on one page. Drawing it once beats rereading it ten times.

Which physical topology does the network you're on right now actually use — and can you name the single point of failure in it? Tell me in the comments; identifying that weak link is exactly the skill Part 004's backbones and segments will formalize.