Network Architecture

IB Syllabus: A2.2.1 – Network topologies, A2.2.2 – Function of servers (HL), A2.2.3 – Networking models, A2.2.4 – Network segmentation

Table of Contents

  1. Key Concepts
    1. Network Topologies (A2.2.1)
      1. Star Topology
      2. Mesh Topology
      3. Hybrid Topology
      4. Comparison of Topologies
    2. Networking Models (A2.2.3)
      1. Client-Server Model
      2. Peer-to-Peer (P2P) Model
      3. Comparison of Networking Models
      4. Real-World Applications
    3. Network Segmentation (A2.2.4)
      1. Subnetting
      2. VLANs (Virtual Local Area Networks)
    4. Server Types (A2.2.2)
  2. Worked Examples
    1. Example 1: Choosing a Topology
    2. Example 2: Client-Server vs Peer-to-Peer
  3. Quick Check
  4. Trace Exercise
  5. Spot the Error
  6. Fill in the Blanks
  7. Predict the Output
  8. Practice Exercises
    1. Core
    2. Extension
    3. Challenge
  9. Connections

Key Concepts

Network Topologies (A2.2.1)

A network topology describes the arrangement of devices and connections in a network. The topology you choose affects reliability, speed, scalability, cost, and how data collisions are handled. The 2027 IB syllabus covers three topologies: star, mesh, and hybrid.

Star Topology

        [Device 1]
            |
[Device 4]--[Central Switch/Hub]--[Device 2]
            |
        [Device 3]
  • All devices connect to a central node (switch or hub)
  • If one device fails, others are unaffected
  • If the central node fails, the entire network goes down (single point of failure)
  • Easy to add new devices (good scalability)
  • Most common in homes and small offices

Mesh Topology

    [A]--------[B]
    | \        / |
    |   \    /   |
    |    [E]     |
    |   /    \   |
    | /        \ |
    [C]--------[D]
  • Full mesh: every device connects to every other device
  • Partial mesh: some devices have redundant connections, but not all
  • Highly reliable: if one link fails, data can be rerouted through alternative paths
  • No single point of failure
  • Expensive to implement (many cables and connections)
  • Used in large organisations, data centres, and WANs where reliability is critical

Hybrid Topology

  • Combines two or more topology types (e.g., star-mesh)
  • Flexible – can be designed to meet specific needs
  • Example: a university with star topology in each building, connected by a mesh backbone between buildings
  • Benefits of multiple topologies; complexity of management

Comparison of Topologies

Factor Star Mesh Hybrid
Reliability Single point of failure (central node) Highly reliable (redundant paths) Depends on design
Speed Good (dedicated link to central node) Excellent (multiple paths reduce congestion) Varies
Scalability Easy (add to central node) Difficult (many new connections needed) Moderate
Data collisions Low (switch directs traffic) Very low (dedicated links) Varies
Cost Low to moderate High (many connections) Moderate to high
Example Home network, office LAN Data centre, military network University campus

IB exam note: The 2027 syllabus specifies star, mesh, and hybrid only. Bus and ring topologies from the old syllabus are NOT assessed.

Networking Models (A2.2.3)

Client-Server Model

[Client 1] ----\
[Client 2] -----→ [Central Server]
[Client 3] ----/
  • One or more servers provide services; clients request and consume services
  • Server manages resources centrally (files, databases, authentication)
  • Benefits: centralised management, security, backups, consistent data
  • Drawbacks: server is a single point of failure, expensive to set up and maintain, server overload with many clients

Peer-to-Peer (P2P) Model

[Peer A] ←→ [Peer B]
   ↕              ↕
[Peer C] ←→ [Peer D]
  • Every device is both client and server – all peers are equal
  • No central server needed
  • Benefits: no single point of failure, low setup cost, easy to set up
  • Drawbacks: harder to manage security, no centralised backup, performance depends on individual peers, harder to scale

Comparison of Networking Models

Feature Client-Server Peer-to-Peer
Management Centralised Decentralised
Security Strong (centralised policies) Weaker (each peer manages own security)
Cost High (server hardware + admin) Low (no dedicated server)
Reliability Server = single point of failure No single point of failure
Scalability Good with proper server infrastructure Poor for large networks
Backup Centralised and automated Each peer manages own backups

Real-World Applications

  • Client-server: web browsing (browser → web server), email (client → mail server), online banking, school networks
  • Peer-to-peer: file sharing (BitTorrent), blockchain/cryptocurrency, VoIP (Skype – historically), some gaming

Network Segmentation (A2.2.4)

What is network segmentation? Dividing a network into smaller, isolated sections to improve:

  • Performance: less traffic per segment = fewer collisions, faster speeds
  • Security: a breach in one segment does not automatically compromise others
  • Management: easier to monitor, troubleshoot, and apply policies to smaller segments

Subnetting

  • Dividing a network into smaller networks (subnets) using IP address ranges
  • Example: a school might have subnet 192.168.1.x for admin, 192.168.2.x for students, 192.168.3.x for servers
  • Each subnet has its own range of IP addresses
  • Traffic between subnets must pass through a router (which can enforce rules)

VLANs (Virtual Local Area Networks)

  • Logical segmentation – devices on the same physical switch can be in different VLANs
  • A VLAN groups devices by function, department, or security level, regardless of physical location
  • Example: all finance department computers across three floors can be on the same VLAN
  • VLANs improve security (VLAN traffic is isolated) and reduce broadcast traffic

Subnetting vs VLANs: Subnetting divides by IP address ranges (Layer 3 – Internet layer). VLANs divide by switch port assignment (Layer 2 – Network Interface layer). Both achieve segmentation, but VLANs offer more flexibility because devices do not need to be physically near each other.

Server Types (A2.2.2)

HL Only – Server types and their functions are assessed at HL only. SL students should understand the client-server model (A2.2.3) but are not required to describe specific server types.

Server Type Function Key Details
DNS Server Translates domain names (e.g., www.example.com) to IP addresses (e.g., 93.184.216.34) Acts as the internet’s “phone book”; without DNS, users would need to memorise IP addresses
DHCP Server Automatically assigns IP addresses to devices joining the network Manages a pool of available addresses; assigns leases with expiry times
File Server Stores, manages, and provides access to files on the network Centralised storage; access controlled by permissions; enables collaboration
Mail Server Sends, receives, and stores email messages Handles SMTP (sending), IMAP/POP3 (receiving); manages mailboxes
Proxy Server Acts as an intermediary between clients and the internet Caches web pages (faster access), filters content, hides client IP addresses (anonymity)
Web Server Hosts websites and serves web pages to clients via HTTP/HTTPS Processes requests from browsers; delivers HTML, CSS, JS, images

Factors to consider for each server type:

  • Scalability: can the server handle growth? (e.g., adding more users, more files, more traffic)
  • Reliability: what happens if it fails? (e.g., DNS failure = no website access; use redundancy)
  • Security: what data does it hold? (e.g., mail servers contain sensitive communications; need encryption)

Worked Examples

Example 1: Choosing a Topology

For each scenario, recommend a topology and explain why.

# Scenario Topology Explanation
1 A home network with 5 devices Star Simple, affordable, easy to add devices; central router handles traffic
2 A military communication network Mesh Maximum reliability; if one link is destroyed, data reroutes through alternative paths
3 A university campus with buildings connected by fibre, and star networks within each building Hybrid Star within buildings (simple, cost-effective), mesh between buildings (redundancy for critical backbone)

Example 2: Client-Server vs Peer-to-Peer

For each scenario, identify the most appropriate networking model and explain why.

# Scenario Model Explanation
1 Online banking application Client-server Requires centralised security, authentication, and data integrity
2 Friends sharing music files directly Peer-to-peer No central server needed; each person shares directly
3 School email system Client-server Mail server manages all accounts centrally with backups and security
4 Blockchain cryptocurrency network Peer-to-peer Decentralised by design; no central authority; all nodes verify transactions

Quick Check

Q1. Which topology has a single point of failure at its central node?

Q2. A large corporation needs maximum network reliability with no single point of failure. Which topology is most appropriate?

Q3. In the client-server model, what is the main drawback of the server?

Q4. Which network segmentation technique groups devices logically regardless of physical location?

Q5. (HL) Which type of server translates domain names into IP addresses?


Trace Exercise

A star network has 5 devices (A–E) connected to a central switch. Device A sends a message to Device D. Trace what happens in two scenarios: (a) the switch is working normally, and (b) the switch fails.

Trace: Star Network Communication

Device A sends a message to Device D through a central switch. For each step, describe what happens.

StepScenarioWhat Happens
1 Switch working
2 Switch working
3 Switch working
4 Switch fails
5 Switch fails

Spot the Error

A student wrote revision notes about networking models. One line contains an error in the definition. Click the line with the error, then pick the correct fix.

1Client-server: all devices are equal and share resources directly 2Peer-to-peer: one central server provides services to requesting clients 3VLANs: group devices by physical location only 4Subnetting: divides a network into smaller subnets using IP address ranges

Pick the correct fix for line 1:


Fill in the Blanks

Complete the summary of network architecture by filling in the correct term for each blank.

Fill in the blanks to complete this summary of network architecture:

NETWORK ARCHITECTURE
====================
In a  topology, all devices connect to a central switch or hub.

A  topology provides redundant paths, so there is no single point of failure.

The  model uses a central server to provide services to requesting devices.

 group devices logically regardless of their physical location on the network.

 divides a network into smaller networks using IP address ranges.

Predict the Output

A mesh network has 4 devices. Each device is connected to every other device (full mesh). How many direct connections are there in total?

Type a number:

A school network uses VLANs. The student VLAN and teacher VLAN are on the same physical switch. Can a student computer directly access a teacher's shared folder without going through a router?

Type Yes or No:


Practice Exercises

Core

  1. Topology Comparison – Create a table comparing star, mesh, and hybrid topologies across four criteria: reliability, cost, scalability, and one practical application for each.

  2. Client-Server vs P2P – Describe two advantages and two disadvantages of each networking model. Give one real-world example of each.

  3. Segmentation – Explain why a school might divide its network into separate segments for students, teachers, and administration. What security and performance benefits does this provide?

Extension

  1. Topology Selection – A hospital needs a network for: (a) patient monitoring devices in intensive care, (b) office computers for administrative staff, (c) connecting three buildings on its campus. Recommend a topology for each scenario and justify your choice based on reliability, cost, and scalability.

  2. Server Functions (HL) – Describe the function of DNS, DHCP, and proxy servers. Explain what would happen to a network if each of these servers went offline for one hour.

Challenge

  1. Network Redesign – A growing company currently uses a flat network (no segmentation) with a peer-to-peer model for 20 employees. They are expanding to 200 employees across 4 departments (Sales, Engineering, HR, Finance). Design an improved network architecture that includes: (a) an appropriate networking model with justification, (b) a suitable topology, (c) a network segmentation strategy (subnets and/or VLANs), and (d) the types of servers needed (HL). Evaluate the trade-offs of your design.

Connections

  • Prerequisites: Network Fundamentals – understanding network types and devices; topologies define how these devices are connected
  • Related: Protocols and Layers – the TCP/IP model layers map to different parts of the network architecture
  • Related: Cloud Computing – cloud services are delivered through client-server architecture; data centres use mesh topology
  • Forward: Data Transmission – packet routing depends on network topology; segmentation affects how packets are addressed
  • Forward: Network Security – segmentation is a key security strategy; server security is critical for protecting network resources

Back to top

© EduCS.me — A resource hub for IB Computer Science

This site uses Just the Docs, a documentation theme for Jekyll.