Computer Networks

What is a Computer Network?

A computer network is a collection of interconnected computing devices (such as servers, personal computers, smartphones, routers, switches, and smart IoT devices) that can communicate with each other, exchange data packets, and share resources (like internet connections, printers, and cloud storage). Devices on a network are connected using physical cables (copper Ethernet cables, fiber optics) or wireless media (Wi-Fi, cellular, satellite).

In our modern digital economy, computer networks form the infrastructure for the global Internet, corporate intranets, cloud computing platforms, and local smart-home ecosystems. Understanding how data travels across these networks is the starting point for all software engineering, system administration, and cyber security paths.

Types of Computer Networks

Networks are classified based on their geographic span, ownership, and performance characteristics:

Network Type	Definition	Geographic Scope	Performance Characteristics
LAN (Local Area Network)	Connects computers within a localized, private area.	A single room, home, office building, or school.	Extremely high speed (1 to 10+ Gbps) and very low latency.
WAN (Wide Area Network)	Connects multiple smaller networks across vast distances.	Cities, countries, or the entire planet (e.g., the Internet).	Slower speeds, higher latency, managed by telecom carriers.
PAN (Personal Area Network)	A short-range network centered around a single person.	Within ~10 meters (33 feet).	Low-power wireless (Bluetooth, Zigbee) or wired (USB).
MAN (Metropolitan Area Network)	Covers a larger geographic area than a LAN but smaller than a WAN.	An entire town, city, or large university campus.	Often uses high-speed fiber rings managed by municipalities.
WLAN (Wireless LAN)	A LAN where devices connect wirelessly using Wi-Fi standards.	Home, café, or office coverage area.	Highly convenient, but prone to signal interference and security risks.
SAN (Storage Area Network)	A dedicated high-speed network providing block-level data storage access.	Enterprise data centers.	Uses Fiber Channel or iSCSI to connect servers directly to storage arrays.

Introduction to Network Topologies

The layout or structure of a network is called its **Topology**. Topologies define how nodes (devices) are physically connected and how data logically flows between them:

• Star Topology — All nodes connect directly to a central device (usually a network switch or router). If one cable breaks, only that node goes offline, making it highly reliable and easy to troubleshoot. This is the standard topology for modern home and office networks.
• Mesh Topology — Every node connects directly to multiple other nodes. In a *full mesh*, every single node is connected to every other. This provides maximum redundancy and fault tolerance—if one link fails, data routes via another path. Mesh topologies are expensive to wire physically but are standard in critical infrastructure and wireless mesh networks.
• Bus Topology — All nodes connect to a single central cable (the "bus" or backbone). It is cheap and easy to install but has a single point of failure: if the central cable breaks, the entire network goes down. Data collisions are also common. It is considered a legacy topology.
• Ring Topology — Nodes are connected in a closed loop. Data packets travel in one direction around the ring, passing through each device. A single node failure breaks the entire loop. Like the bus, it is rarely used in modern office networks.

Key Performance Metrics in Networking

The performance and quality of a computer network are measured using several vital statistical metrics:

• Bandwidth — The theoretical maximum capacity of a network link to transmit data in a given time frame (commonly measured in Megabits per second, or Mbps). Think of bandwidth like the width of a water pipe: a wider pipe *can* carry more water at once.
• Throughput — The actual amount of data successfully transmitted over a link under real-world conditions. Throughput is almost always lower than bandwidth due to network congestion, protocol overhead, and hardware limits.
• Latency (Delay) — The time it takes for a packet of data to travel from the source to the destination (measured in milliseconds, or ms). High latency causes delay in video calls and online gaming.
• Jitter — The variation in packet arrival times. Consistent packet arrival is vital for real-time traffic like VoIP (Voice over IP) or live video streams. High jitter results in audio drops and stuttering video.
• Packet Loss — The percentage of data packets that fail to reach their destination. Caused by network congestion, corrupted signals, or faulty hardware, packet loss forces protocols like TCP to retransmit data, slowing down performance.

Network Reference Models

To design networks and ensure that devices from different manufacturers can communicate, the industry relies on standardized reference models:

• The OSI Model (7 Layers) — A conceptual framework that breaks network communication down into seven distinct layers, from physical electrical pulses to high-level software interfaces. Read our detailed OSI Model Guide.
• The TCP/IP Model (4 Layers) — The practical, real-world model that the Internet is built upon. Read our detailed TCP/IP Model Guide.

Frequently Asked Questions (FAQ)

❓ What is the difference between the Internet, an Intranet, and an Extranet?

• The Internet is the global, public network of interconnected computer networks accessible to anyone.
• An Intranet is a private network accessible only to an organization's internal employees (used for internal tools and documentation).
• An Extranet is a controlled private network that allows access to authorized external partners, clients, or vendors.

❓ What is the difference between a MAC address and an IP address?

A MAC (Media Access Control) address is a permanent, unique physical hardware identifier burned into every network interface card (NIC) at the factory (Layer 2). An IP (Internet Protocol) address is a logical address assigned by software to locate a device on a network (Layer 3) and can change when you move networks (e.g., home Wi-Fi vs. office Wi-Fi).

❓ How does data travel across a network?

Data is not sent as a single continuous block. Instead, software breaks files down into smaller, manageable chunks called **packets**. Each packet is labeled with the source and destination IP addresses. Routers and switches forward these packets across different network segments, and the receiving device reassembles them into the original file.

❓ What is latency and how can I reduce it?

Latency is the round-trip travel time for data packets. It can be reduced by using wired Ethernet connections instead of Wi-Fi, upgrading network routing hardware, closing background bandwidth-heavy applications, or using Content Delivery Networks (CDNs) to host website content closer to the users.

What's Next?

Advance your network engineering learning path:

• Learn how network layouts and hardware are deployed in Topologies & Hardware.
• Understand the standard reference framework in OSI Model.
• Explore the protocol suite of the Internet in TCP/IP.
• Study IP ranges and subnet calculations in IP Addressing & Subnetting.