When a CCNA instructor asks "what kind of network is this?", there are really two separate questions hiding inside it: how big is ...
When a CCNA instructor asks "what kind of network is this?", there are really two separate questions hiding inside it: how big is the area it covers, and how do the computers on it share work. In this post I will walk you through both, using networks you already touch every day in Nepal: your home Wi-Fi, a small office in Kathmandu, and the ISP that connects them to the rest of the world.
What is a computer network?
A computer network is two or more devices connected so they can exchange data and share resources such as files, printers, an internet connection or an application. The devices can be PCs, phones, servers, printers, IP cameras or network equipment like switches and routers. The links between them can be copper cable, fiber or radio (Wi-Fi, Bluetooth, mobile data).
The internet is the largest example: a network of networks, all exchanging traffic using TCP/IP.
We usually classify networks in two ways:
- By geographic size (PAN, LAN, CAN, MAN, WAN), which is what most exam questions mean by "types of network".
- By architecture or role model (peer-to-peer and client-server). Larger architectures such as centralized, cluster, cloud and virtualized networks get their own post: see Related tutorials at the end.
Networks by geographic size
PAN: Personal Area Network
A PAN covers the space around one person, usually a few meters. Your phone talking to wireless earbuds over Bluetooth, a smartwatch syncing with your phone, or a laptop connected to a wireless mouse are all PANs. A phone sharing its mobile data to a laptop by USB tethering or a hotspot is also often described as a PAN. IEEE standardized Bluetooth as 802.15.1, but the Bluetooth specification itself is maintained by the Bluetooth Special Interest Group (SIG).
LAN: Local Area Network
A LAN covers a single building or a floor: your home, a school computer lab, or an office. It is owned and managed by the same person or organization that uses it, which is the main thing that separates it from a WAN.
- Wired LANs use Ethernet (IEEE 802.3). The standard covers speeds from 10 Mbps all the way up to 400 Gbps and beyond. In a typical office today you will see 1 Gbps to the desk and 10 Gbps between switches.
- Wireless LANs (WLANs) use Wi-Fi (IEEE 802.11). The first 802.11 standard in 1997 ran at 2 Mbps and 802.11b at 11 Mbps; current Wi-Fi 6 (802.11ax) and Wi-Fi 7 (802.11be) access points can deliver several hundred Mbps to a single client in real conditions. The headline numbers on the box are theoretical maximums you will rarely see.
Nepal example: the fiber router your ISP installs at home (WorldLink, Vianet, Nepal Telecom FTTH and so on) creates a small LAN. Its LAN ports and Wi-Fi give your devices private addresses such as 192.168.1.x, and everything behind that router is your LAN.
CAN: Campus Area Network
A CAN connects several LANs across buildings on one site, for example a university campus or a hospital compound. The buildings are linked by fiber, and the whole thing is still owned by one organization. Cisco design guides use the word "campus" constantly.
MAN: Metropolitan Area Network
A MAN spans a city or a large town. It usually connects many sites and is typically built by a service provider or a city-wide organization rather than a single small company.
Nepal example: an ISP's fiber rings running around Kathmandu Valley, linking its points of presence in Kathmandu, Lalitpur and Bhaktapur, form a MAN. A bank that connects all its valley branches to its head office over leased fiber from an ISP is also using a MAN service. City cable TV networks are the classic textbook example. The old IEEE 802.16 (WiMAX) standard was designed for wireless MANs, but it has largely been replaced by 4G LTE and fiber.
WAN: Wide Area Network
A WAN connects networks over large distances: across a country, between countries, or across continents. Organizations almost never build their own long-distance links; instead they lease capacity from carriers and ISPs using technologies such as MPLS, leased lines, metro Ethernet, internet VPNs or SD-WAN.
Nepal example: a company with offices in Kathmandu, Pokhara and Biratnagar connects them with an MPLS or SD-WAN service from an ISP. That is the company's WAN. Nepal Telecom and the large ISPs operate WANs of their own that cover the whole country and connect to international upstream providers through India and China. The internet itself is the largest WAN.
Other names you will meet
- WLAN (Wireless LAN): simply a LAN built with Wi-Fi.
- SAN (Storage Area Network): a dedicated high-speed network that connects servers to shared disk storage, common in data centers.
- VPN (Virtual Private Network): not a size category but a secure tunnel over another network, often used to build a WAN over the internet.
You may also see "VWAN" or "LWAN" in some notes. These are not standard network categories, so do not worry about them for the exam.
Comparison table
The speeds below are typical figures you will see in practice, not maximums. Real numbers depend on the technology and on how much you pay.
| Type | Typical range | Typical speed | Usually owned by | Example in Nepal |
|---|---|---|---|---|
| PAN | Up to about 10 m | 1 to a few Mbps (Bluetooth), higher for USB | One person | Phone connected to earbuds and a smartwatch |
| LAN | One building or floor (Ethernet copper up to 100 m per cable run) | 100 Mbps to 10 Gbps wired; hundreds of Mbps on Wi-Fi | Home owner or organization | Home fiber router with Wi-Fi; office switch network |
| CAN | Several buildings on one site, up to a few km | 1 to 100 Gbps between buildings | One organization | University campus linking departments over fiber |
| MAN | A city, roughly 5 to 50 km | 1 to 100 Gbps on the provider backbone | ISP, telecom or city-wide organization | ISP fiber ring across Kathmandu Valley |
| WAN | Country, continent or global | From a few Mbps per branch to many Gbps on backbones | Carriers and ISPs; customers lease links | Company linking Kathmandu, Pokhara and Biratnagar offices via MPLS or SD-WAN |
Networks by role model: peer-to-peer vs client-server
Size tells you where a network is. The role model tells you who provides services and who uses them.
Peer-to-peer (P2P) network
In a peer-to-peer network every computer is equal. Each one can share its own files or printer and also use resources on the others, so each machine acts as both client and server. There is no dedicated server and no central list of users.
A home where one laptop shares a folder and another PC shares a USB printer is peer-to-peer. So is a small shop with three PCs in a Windows workgroup. Note that "peer" refers to equal roles, not to the internet; two PCs on the same switch can be peers. File-sharing applications such as BitTorrent use the same idea at internet scale.

Client-server network
In a client-server network, one or more dedicated servers provide services and the other computers (clients) request them. The server holds shared files, user accounts, email, a database or an application, and it decides who is allowed to access what. Every time you open a website you are using client-server: your browser is the client and the web server answers its requests.
A typical Nepali office with a Windows Server domain controller, a file server and an accounting application server (Tally or a local ERP, for example) is client-server.

Pros and cons side by side
Each point on one side has its opposite on the other, which makes it easier to remember for exams.
| Factor | Peer-to-peer | Client-server |
|---|---|---|
| Upfront cost | Low: no server hardware or server licenses | Higher: server hardware, OS licenses, possibly a rack and UPS |
| Setup | Simple; built into Windows, macOS and Linux | More planning: domain, DNS, permissions, backups |
| Skills needed | Basic user knowledge | A trained administrator |
| Administration | Done separately on every PC | Centralized: one place for users, permissions and policies |
| Security | Depends on each user; harder to enforce consistent rules | Easier to enforce consistent rules, but the server is a high-value target and must be protected |
| Backup | Data is scattered, so central backup is hard | Data lives on servers, so backup is simple to schedule |
| Scalability | Works well up to roughly 10 to 15 devices, then becomes messy | Scales to thousands of users by adding or upgrading servers |
| Failure impact | One PC failing only removes its own shared resources | If a server fails, its service stops for everyone unless you add redundancy |
| Best for | Homes and very small shops | Offices, schools, banks, any organization that needs control |
A useful way to think about cost: client-server costs more to start, but as the number of users grows it usually saves time and money because you manage everything once instead of on every PC. Those two statements are not a contradiction; one is about initial cost, the other about running cost at scale.
Putting it together: one day of packets in Kathmandu
Here is how all of these types appear in a normal day:
- Your phone connects to your Bluetooth earbuds (PAN).
- At home the phone joins your fiber router's Wi-Fi (LAN).
- Your traffic leaves the home router over the ISP's fiber and travels around the valley on the ISP's metro network (MAN).
- To reach a server outside Nepal, it crosses the ISP's national backbone and international links (WAN, and then the internet).
- At the office, you log in to a domain account checked by a server (client-server), while a colleague shares a folder from her laptop for a quick file transfer (peer-to-peer).
Prefer video? Watch more hands-on labs on the Technical Hakim YouTube channel.
Summary
- By size: PAN (one person), LAN (one building), CAN (one campus), MAN (one city), WAN (country or wider).
- The biggest practical difference between LAN and WAN is ownership: you own your LAN, but you usually lease WAN links from a provider.
- Ethernet is IEEE 802.3, Wi-Fi is IEEE 802.11, and modern speeds are far higher than old textbooks state.
- Peer-to-peer is cheap and simple for small groups; client-server costs more to start but is easier to manage, secure and back up as you grow.
Related tutorials
- Wi-Fi vs Ethernet: Comparing WLANs and Wired LANs in 2026
- Setting Up a Home Network in Nepal: ISP Router, Wi-Fi and a Separate Network for Smart Devices
- Network Architectures Compared: Centralized, Cluster, Cloud and Virtualized Networks
- OSI Model vs TCP/IP Model: Layers, Protocols and a Wireshark Walkthrough
- Network Infrastructure Explained: Devices, Cabling and a Small Office Design
Sources
- IEEE 802.3 Standard for Ethernet
- IEEE 802.11 Standard for Wireless LANs
- IEEE 802.15.1 (Bluetooth-based WPAN) and Bluetooth SIG Core Specification
- Andrew S. Tanenbaum, Nick Feamster and David Wetherall, Computer Networks, 6th edition
- Cisco Networking Academy, CCNA: Introduction to Networks
No comments