A-Level Computer Science / Unit 2: Networks, Connectivity and the Internet

2.1.2 Network Models and Client Types

🔒 Lesson slides are available to signed-in users. Sign in

2.1.2 Network Models and Client Types

A network model describes how connected computers divide responsibility for providing and using services. A device may request a resource, provide one, or perform both roles in different interactions.

This section compares client-server and peer-to-peer organisation, considers how roles can change across connected subnetworks, and examines how processing is divided between thin and thick clients.

By the end of this section, you should be able to:

  • describe the roles of clients, servers and peers;
  • explain how computers may take different roles across networks and subnetworks;
  • compare the benefits and drawbacks of client-server and peer-to-peer models;
  • justify a suitable network model for a given situation;
  • distinguish thin clients from thick clients by considering where processing takes place.

Roles within a network

The terms client, server and peer describe what a computer or program is doing in a particular communication. They do not always describe a permanent property of the device.

Client: hardware or software that requests or uses a network service.
Server: hardware or software that provides a network service.
Peer: a computer that can both request and provide resources directly.
System Role in the interaction What it does
Student laptop Client Requests a page from the school's learning platform.
Web server Server to the laptop Receives the request and returns the requested page.
Web server Client to a database server Requests the student's authorised course data.
Database server Server to the web server Checks the query and returns the required records.

Roles across subnetworks

A large organisation may divide its system into smaller subnetworks. For example, user devices may be on one subnetwork, application servers on another, and database servers on a more restricted subnetwork. Communication can pass between them while each system performs the role required for that stage of the task.

Subnetwork: a smaller network that forms part of a larger interconnected system.

Common mistake

Do not assume that one computer is always either a client or a server. A web server can provide pages to a browser while acting as a client when it requests data from another server.

The client-server model

In a client-server model, one or more servers provide managed services to clients. The service might involve authentication, file storage, printing, email, a web application or access to a database.

Client-server model: a network arrangement in which clients request services and dedicated servers provide and manage those services.

Typical request cycle

  1. The client sends a request.
  2. The server checks the request and may verify identity or permissions.
  3. The server processes the request or accesses stored information.
  4. The server returns a response to the client.

Benefits and drawbacks

Feature Potential benefit Possible drawback
Central administration Accounts, permissions and policies can be applied consistently. Specialist staff and server equipment increase cost.
Central data storage Backups, version control and malware scanning can be managed in one place. A server failure may make important data unavailable unless redundancy is provided.
Shared services Many clients can use the same application or dataset. A heavily used server or network link can become a bottleneck.
Controlled access Users can be given different permissions according to their responsibilities. A successful attack on a central server may affect many users.

Where it is suitable

  • a hospital system that stores confidential patient records;
  • a school platform that uses central accounts and permissions;
  • an online shop that manages stock, orders and payments;
  • a business file service that requires regular backups and audit records.

Exam tip

Explain the consequence of central control. For example, central accounts allow the organisation to remove a former employee's access from one managed system rather than changing settings separately on every workstation.

The peer-to-peer model

In a peer-to-peer (P2P) model, computers provide resources directly to one another. A peer acts as a client when requesting a resource and as a server when supplying one. A dedicated central server is not required for the shared resource.

Peer-to-peer model: a distributed arrangement in which peers request and provide resources directly to one another.

How peer-to-peer sharing can work

  1. A peer announces or makes a resource available.
  2. Another peer requests that resource.
  3. The supplying peer sends the data directly.
  4. In a larger distributed system, different parts may be supplied by several peers.

Benefits and drawbacks

Feature Potential benefit Possible drawback
No dedicated service server A small group can share resources with lower setup cost. There is no single point for managing permissions, updates or backups.
Distributed workload Requests can be spread across several peers rather than one central server. Performance depends on whether peers are online and how reliable their connections are.
Multiple resource sources Different file blocks may be obtained from different peers. It can be harder to confirm that a file is authentic, complete and current.
Direct sharing Users can exchange low-risk resources without central administration. Malware checking, access control and version management may be inconsistent.

Common mistake

Peer-to-peer does not mean that the computers stop acting as clients and servers. Each peer performs both roles at different times.

Comparing the models

The most suitable model depends on the required level of control, scale, security, availability and administration.

Decision factor Client-server Peer-to-peer
Control Services and permissions are centrally managed. Each peer manages its own shared resources.
Cost Requires server hardware, software and administration. Can use existing computers without a dedicated server.
Backup and security Policies can be applied consistently. Protection depends on individual peers.
Availability Can be dependable when redundant servers are used. A resource disappears when its supplying peers are unavailable.
Workload Requests are concentrated on managed servers. Requests can be distributed among peers.
Best fit Managed organisational services and sensitive data. Simple temporary sharing or distributed resource delivery.

Choosing and justifying a model

A strong justification links a requirement in the scenario to a relevant feature of the selected model and then explains the practical consequence.

Reasoning structure

Choice → model feature → consequence for the situation

Apply the reasoning

Veterinary practice: five clinics need central appointment records, controlled access and daily backups.

Suitable model: client-server. Central servers can hold the appointment database and apply permissions consistently, so authorised staff at every clinic can use the same current records. Managed backups also reduce the risk of each clinic maintaining incomplete copies.

Film club: six students need to exchange non-confidential draft clips during one weekend.

Possible model: peer-to-peer. The students can share files directly using their existing computers, avoiding the time and cost of setting up a dedicated server. They should still agree on filenames and keep a separate final backup.

Game update: a very large file must be delivered to many users at the same time.

A peer-to-peer distribution method may help. Different file blocks can be supplied by several peers, spreading traffic across multiple sources. A trusted central service may still publish the official file information and verification value.

Common mistake

Avoid answers such as “client-server is better” or “peer-to-peer is cheaper” without explaining why that point matters in the stated situation.

Thin clients and thick clients

Thin and thick clients are both used in client-server systems. The distinction concerns how application processing is divided between the client device and the server.

Thin client: a client that depends heavily on the server for application processing and mainly sends input and displays returned output.
Thick client: a client that performs a substantial amount of application processing locally.
Comparison point Thin client Thick client
Processing location Most processing takes place on the server. A substantial amount takes place on the client.
Client hardware May use simpler hardware and less local storage. Usually needs more capable hardware and storage.
Software maintenance Applications can often be updated centrally. Software may need installation and updates on individual clients.
Network dependence Strongly dependent on a reliable connection to the server. May continue some work when the network is unavailable.
Server workload Places more processing demand on central servers. Distributes more processing to client devices.

Examples

A booking desk may use a thin client because every workstation runs the same centrally managed application and local storage is unnecessary. A computer-aided design workstation is more likely to be a thick client because complex graphics require substantial local processing.

Real applications can lie between the two extremes. A browser application may carry out some processing locally while still relying on servers for authentication, data storage and larger tasks.

Common mistake

“Thin” and “thick” refer to the location and amount of processing, not to the physical dimensions of the computer.

Exam tip

When comparing client types, identify where processing occurs and then develop a consequence, such as hardware requirements, ease of maintenance or dependence on the network.

Interactive: client-server and peer-to-peer sharing

Switch between the models and start the transfer. Observe whether the file blocks come from one managed source or from several peers.

Practice

Try these questions

  1. Describe the role of a client and the role of a server in an online library system.
  2. Explain how one computer can act as both a server and a client in different interactions.
  3. Explain two benefits and two drawbacks of a client-server model.
  4. Explain why each peer can perform two roles in a peer-to-peer model.
  5. Explain one benefit and one drawback of distributing a file through several peers.
  6. A company stores confidential payroll records. Select a model and justify your choice.
  7. Five musicians need to exchange temporary rehearsal recordings at minimum cost. Select a model and justify your choice.
  8. Explain two differences between thin and thick clients.
  9. Explain why a thin client may be easier to maintain but more dependent on the network.
  10. Give a situation in which a thick client would be preferable and justify your answer.

Review

Key ideas

  • A client requests a service, while a server provides one.
  • A system can take different roles in different interactions or across subnetworks.
  • Client-server systems favour central management, controlled access and consistent backup.
  • Peer-to-peer systems allow peers to request and provide resources directly.
  • Each model has benefits and drawbacks; selection must depend on the scenario.
  • Thin and thick clients differ mainly in how processing is divided between client and server.

Quick self-check

  1. Can I distinguish a client, server and peer?
  2. Can I explain how roles may change across a larger system?
  3. Can I evaluate both network models?
  4. Can I justify a choice using evidence from a scenario?
  5. Can I compare thin and thick clients through processing location and its consequences?