The Network Atlas
A Complete Guide to Network Types
Networks by Geographic Size
From your pocket to the planet
The Scale of Connection
Networks come in all sizes. Some connect devices within arm's reach. Others span continents. The geographic scope of a network determines everything: the technology used, the speed possible, who manages it, and what it costs.
Understanding network types by size is fundamental. When someone says "LAN" or "WAN," they're describing a category with specific characteristics, challenges, and solutions.
Let's explore each type, starting from the smallest.
PAN: Personal Area Network
A PAN is the smallest type of network — it connects devices within the reach of an individual person, typically within 10 meters. It's centered around you.
When you connect your wireless earbuds to your phone, pair your smartwatch, or link your laptop to a wireless mouse — you're using a PAN. These are your devices, communicating for your personal use.
Bluetooth: The dominant wireless PAN technology. Range ~10m, low power, designed for small data transfers and audio streaming.
USB: Wired PAN. When you connect a phone to a laptop via cable, that's a wired PAN.
Infrared (IrDA): Older technology, requires line-of-sight. Still used in some remote controls.
NFC: Near Field Communication. Range ~4cm. Used for contactless payments and quick device pairing.
LAN: Local Area Network
A LAN connects devices within a limited area — typically a single building, office, home, or floor. This is the most common type of network you interact with daily.
Your home Wi-Fi is a LAN. Your office network is a LAN. A school computer lab is a LAN. These networks are privately owned and managed by the organization or individual using them.
LANs are characterized by high speed (because distances are short) and low latency (because there are few hops between devices).
Key characteristics of LANs:
• Ownership: Owned and managed by a single organization or individual.
• Speed: High bandwidth (100 Mbps to 10 Gbps or more).
• Cost: Relatively inexpensive to set up and maintain.
• Security: Easier to secure since you control all the equipment.
CAN: Campus Area Network
A CAN connects multiple LANs within a limited geographic area — typically a university campus, corporate campus, military base, or hospital complex.
Think of a university: the library has its own LAN, each dormitory has its own LAN, the engineering building has its own LAN. The CAN ties all these LANs together into one unified network, usually owned and operated by a single organization.
CANs are larger than LANs but smaller than MANs. They're characterized by single ownership — one organization owns all the infrastructure. This distinguishes them from MANs, which typically involve public infrastructure or multiple organizations.
MAN: Metropolitan Area Network
A MAN spans a city or large town — larger than a campus, but smaller than a country. It connects multiple LANs and CANs across a metropolitan area.
When a city provides free public Wi-Fi across downtown, that's a MAN. When a company connects its headquarters to its branch offices across Mumbai, that's a MAN. Cable TV networks that serve a city are also MANs.
MANs often use fiber optic cables for high-speed connections and may involve ISPs or telecommunications companies to provide the infrastructure.
WAN: Wide Area Network
A WAN spans large geographic areas — countries, continents, or the entire world. The internet itself is the largest WAN ever built.
WANs connect multiple LANs and MANs over vast distances. A company connecting its offices in Delhi, London, and New York uses a WAN. A bank connecting ATMs across India uses a WAN.
Because of the distances involved, WANs typically rely on leased telecommunications lines, satellites, or the public internet (often with VPNs for security).
Key characteristics of WANs:
• Lower speeds: Compared to LANs. The speed of light still limits how fast data can travel long distances.
• Higher latency: More hops, more distance = more delay.
• Higher cost: Leasing lines across continents is expensive.
• Third-party infrastructure: You don't own the cables — you lease capacity from telecom providers.
Networks by Topology
How devices are physically and logically arranged
What is Network Topology?
Topology describes how devices in a network are arranged and connected. It's the network's shape — its architecture.
There are two types of topology:
• Physical topology: How devices are actually, physically wired together.
• Logical topology: How data flows through the network, regardless of physical layout.
These can differ. A network might be physically wired as a star (all devices connect to a central switch), but logically behave as a bus (data is broadcast to all devices).
Topology affects performance, reliability, cost, and scalability. Let's explore the main types.
Bus Topology
All devices connect to a single cable (the "bus" or "backbone"). Data sent by any device travels along the entire cable, and all devices receive it — but only the intended recipient processes it.
Think of it like a road with houses on both sides. Everyone can see the cars passing, but only the correct house opens the door.
Advantages: Simple, cheap, easy to set up for small networks. Requires less cable than other topologies.
Disadvantages: If the main cable fails, the entire network goes down. Performance degrades as more devices are added (everyone shares the same bandwidth). Difficult to troubleshoot.
Bus topology was common in early Ethernet networks (10BASE2, 10BASE5) but is rarely used today.
Star Topology
All devices connect to a central device (hub, switch, or router). All communication passes through this central point.
This is the most common topology today. Your home network is almost certainly a star — all devices connect to your Wi-Fi router.
Advantages: Easy to add or remove devices. If one cable or device fails, only that device is affected. Easy to troubleshoot and manage.
Disadvantages: If the central device fails, the entire network goes down. Requires more cable than bus topology.
Star topology dominates modern LANs. Ethernet switches create a star, and Wi-Fi access points create a wireless star.
Ring Topology
Each device connects to exactly two other devices, forming a closed loop. Data travels around the ring in one direction (or both, in dual-ring designs).
In traditional ring networks, a "token" circulates. Only the device holding the token can send data, preventing collisions.
Advantages: Equal access for all devices (no collisions). Performs well under heavy load. Data flows in one direction, making it predictable.
Disadvantages: If any device or cable fails, the ring breaks (unless dual-ring). Adding/removing devices disrupts the network. More difficult to reconfigure.
Token Ring (IBM) and FDDI used ring topology. Largely replaced by Ethernet star topologies, but ring concepts survive in technologies like SONET/SDH for telecommunications.
Mesh Topology
Devices connect to multiple other devices, creating redundant paths. In a full mesh, every device connects to every other device. In a partial mesh, only some devices have multiple connections.
Mesh provides maximum redundancy. If one path fails, data can take another route. This is how the internet's core works.
Advantages: Extremely reliable — no single point of failure. Fast communication (multiple paths). Self-healing: if one path fails, traffic reroutes.
Disadvantages: Expensive. Requires many cables and ports. Complex to set up and manage. Full mesh becomes impractical for large networks (n devices need n(n-1)/2 connections).
Used in internet backbone, data center interconnects, and wireless mesh networks (like mesh Wi-Fi systems for homes).
Tree Topology
Tree topology is a hierarchy of star networks. Multiple star networks connect to a central backbone or root node. Think of an organizational chart.
A building network might use this: each floor has its own star network (switch), and all floors connect to a central core switch.
Advantages: Highly scalable — add more branches easily. Easy to manage segments. Good for hierarchical organizations.
Disadvantages: If the root or backbone fails, large portions of the network go down. More complex cabling.
Very common in enterprise networks. The classic three-tier architecture (access, distribution, core) is a tree topology.
Hybrid Topology
A hybrid topology combines two or more topologies. Most real-world networks are hybrids — they use whatever topology makes sense for each part.
For example: star topology on each floor, mesh topology connecting floor switches, and tree topology connecting buildings. This is hybrid.
Why hybrid? Different parts of a network have different needs. User devices might use star (easy to manage). Critical infrastructure might use mesh (reliability). The network backbone might use ring (predictable performance). Hybrid lets you optimize each segment.
Topology Comparison
| Topology | Cost | Reliability | Scalability | Use Case |
|---|---|---|---|---|
| Bus | Low | Poor | Poor | Legacy, small temp networks |
| Star | Medium | Good | Good | Most LANs, home networks |
| Ring | Medium | Medium | Medium | Telecom backbone (SONET) |
| Mesh | High | Excellent | Medium | Internet backbone, data centers |
| Tree | Medium-High | Good | Excellent | Enterprise networks |
| Hybrid | Varies | Varies | Excellent | Most real-world networks |
Networks by Architecture
How devices relate to each other
Peer-to-Peer (P2P)
In a P2P network, all devices are equals (peers). There's no dedicated server. Each device can act as both client and server — sharing resources and consuming resources from others.
When you share a folder from your computer to another computer at home, that's P2P. BitTorrent is P2P — every downloader is also an uploader.
Advantages: No server to buy/maintain. Easy to set up. No single point of failure. Each peer is independent.
Disadvantages: No centralized security or management. Each device must be individually configured. Doesn't scale well. Finding resources can be difficult.
Client-Server
In client-server, there are two roles: servers that provide resources/services, and clients that request and consume them. The server is powerful and centrally managed.
When you browse the web, your browser (client) requests pages from web servers. When you log into a corporate network, your laptop (client) authenticates against a domain controller (server).
Advantages: Centralized management, security, and backups. Scales well. Easier to find and access resources. Dedicated server hardware performs better.
Disadvantages: Server is a single point of failure (unless redundant). Requires more investment. More complex to set up and maintain.
P2P: Small networks (≤10 devices), temporary setups, when no central control is needed, when budget is minimal.
Client-Server: Organizations, when security matters, when centralized backups are needed, when you need to scale, when you need centralized user management.
Special Network Types
Networks defined by purpose or technology
The Internet
The Internet is the global system of interconnected networks. It's a network of networks — millions of private, public, academic, business, and government networks linked together.
The Internet uses TCP/IP protocols and spans the entire world. It's the largest WAN ever built. When you "go online," you're connecting your device to this global network.
Key point: The Internet is public infrastructure. Anyone can connect. It's not owned by anyone — it's a cooperative system.
Intranet
An intranet is a private network that uses internet technologies (HTTP, TCP/IP, web browsers) but is only accessible to people inside an organization.
Think of it as your company's private internet. It might have internal websites, wikis, HR systems, file shares — all accessible only to employees. Outsiders can't reach it.
Extranet
An extranet extends an intranet to authorized external users — partners, vendors, suppliers, or customers. It's still private and controlled, but extends beyond the organization.
Example: A manufacturer might give suppliers access to inventory systems. A company might let partners access certain internal tools. This controlled external access is an extranet.
Internet: Public. Anyone can access. Global.
Intranet: Private. Only internal employees. Within organization.
Extranet: Private. Internal employees + authorized external partners.
VPN: Virtual Private Network
A VPN creates a secure, encrypted tunnel over the public internet, making it appear as if you're directly connected to a private network.
When you VPN into your company network from home, your traffic is encrypted and routed through the VPN server. It's like having a private cable from your home to the office — except it's virtual, built on top of the public internet.
Types of VPN:
• Remote Access VPN: Individual users connect to a corporate network from home/travel.
• Site-to-Site VPN: Connects entire networks (e.g., connecting branch offices).
VLAN: Virtual LAN
A VLAN creates logically separate networks on the same physical infrastructure. Devices on different VLANs can't communicate directly, even if they're connected to the same switch.
Example: In an office building, you might have VLAN 10 for HR, VLAN 20 for Engineering, VLAN 30 for Guest Wi-Fi. They all use the same switches, but they're isolated from each other.
Why VLANs? Security (isolate sensitive data), performance (reduce broadcast traffic), flexibility (reorganize without rewiring), cost savings (one physical network, multiple logical networks).
SAN: Storage Area Network
A SAN is a specialized, high-speed network dedicated to storage. It connects servers to shared storage devices (disk arrays, tape libraries) at the block level.
Unlike regular network file sharing (where you access files), SANs make remote storage appear as if it's a local disk. Servers can read and write blocks directly, enabling high-performance database hosting and virtual machine storage.
SAN technologies: Fibre Channel (FC), iSCSI (SCSI over IP), FCoE (Fibre Channel over Ethernet).
Complete Summary
LAN — Local Area Network (~1km, building, Ethernet/WiFi)
CAN — Campus Area Network (1-5km, campus, single ownership)
MAN — Metropolitan Area Network (5-50km, city-wide)
WAN — Wide Area Network (unlimited, countries/global)
Star — Central device, most common today
Ring — Closed loop, token passing
Mesh — Multiple paths, high redundancy
Tree — Hierarchical, scalable enterprises
Hybrid — Combination, most real networks
Client-Server — Dedicated servers, centralized, scalable
Intranet — Private internal network, employees only
Extranet — Intranet extended to partners
VPN — Encrypted tunnel over public internet
VLAN — Logical network segments on same physical hardware
SAN — High-speed dedicated storage network