Client-to-server network model

  • Clients are computers that request information

  • Servers are computers that provide information

  • A client depends on the server for information

  • For example, the client-side of a web application is a web browser. The web server provides the information.

Types of servers

  • Web servers

  • Application servers

  • Network Attached Storage (NAS)

  • Print servers

  • Mail servers

  • etc

Peer-to-peer model

  • In a P2P network, each and every computer in a network is responsible for providing data equally.

  • Each computer can provide data requested by other computers in the network.

  • At the same time, any computer can request data from any computer.

  • This type of network is not suitable for systems where there is a single source of truth (ie, databases, online ticketing systems etc)

  • Due to a lack of servers, this type of network may lead to difficulty in storing backups.

  • Known best for file sharing

Cloud Computing

  • The use of hardware and software to deliver a service over a network such as the internet is called cloud computing.
  • A webpage displays these services to a client

Traditional computing vs cloud computing

Traditional computing

  • Files are stored in the hard drive of a computer and processed using the same computer.

Cloud computing

  • Files are stored online and processed using services that are available online.

Pros and Cons


  • More cost effective
  • More reliable
  • Can be accessed from any computer with an Internet connection.
  • Web applications are updated more frequently than traditional software.


  • Requires an internet connection
  • Working in online application may lead to cyber attacks
  • Documents shared in web applications may be accessed by unauthorised users.


  • A webpage is a document that can be viewed in web browsers.


  • A server-side script is processed on the server when a user requests information
  • The processed script returns the result to the user’s request
  • Widely used in websites and apps that store a large amount of information
  • SQL, PHP, Python, Ruby, Java and many more are used as backend languages
  • The validation of data is also carried out on the server-side by comparing the data entered by the user with the existing data in the server.
  • Hackers may try to enter malicious JavaScript in the front-end.

Functions of server-side processing are:

  • Processing client’s queries
  • Provides data security
  • Encryption of data
  • Updating server database

Application Programming Interface (API)

  • API is a set of protocols that governs how two applications interact with each other

  • Websites often embed functions such as maps, spreadsheets etc

  • Websites typically do not create these functions independently

  • Instead, an API is used to include these functionalities within a website

  • For example, Google has an API, which various websites use to include their address in the “contact us” menu

  • A web application that uses API from various sources is called a mashup.

  • API also defines the protocols that will be used in the web service.

  • An example of such a protocol is a WebSocket which creates a connection between two computers

    • It provides a bidirectional communication channel between client and webserver over a single line.

WebSocket protocol

  • Firstly, the client sends a handshake request

  • Then, the server responds with a bidirectional channel on a single socket

  • Both client and server can send messages to each other simultaneously. In HTTP, the webpage need not be refreshed. The connection closes when one of them ends it.

  • The packets are also accepted without the usual security checks. As a result, this protocol helps us achieve fast and real-time communication.

  • This is widely used in online gaming and instant communication.

  • When this protocol is used, fewer webservers are required, saving transmission time, bandwidth, cost of hardware and cost of mobile communication.

Thick vs thin client computing

  • The term thickness refers to the independent processing and storage of a client in a client-server model.

Thick client

  • A thick client processes most of its data and requires little help from the server
  • Operates without server connection
  • Robust and reliable
  • Used for powerful software applications
  • Expensive to install as it requires highly specialised computers at the client side.
  • Software has to be installed at each client computer and maintained regularly.
  • Data is distributed, so data integrity issues may arise.

Thin client

  • A thin client is highly dependent on the server for data processing and storage
  • Without a server, the functionality of a thin client will be affected
  • The thinnest client that requires a server for most of its operations will only require minimum processing power and no storage.
  • This type of client system is often referred to as a dumb terminal.
  • Easy and inexpensive to set up
  • Easy to update and maintain the client systems
  • Data is secured at a centralised location
  • Client computers are affected when a server malfunctions
  • Setting up a highly efficient server is expensive
  • Highly reliable network connections are required
  • When the number of client computers increased, the performance of this model may be affected


  • CRUD stands for Create, Retrieve, Update, Delete
  • CRUD explains the main functions of a database. Without the four functions denoted by CRUD, it is not possible to have a complete database.
  • It also specifies how data is obtained and presented to the user.


  • Representational State Transfer
  • REST is a system design concept in which the HTTP request methods are used to interact with a database server through a webserver.
  • The client does not know how the data is received and processed by the database.
  • In this type of system, client and server systems can be developed independently.


  • JSON and XML are methods used to format data objects transferred across servers and web applications.
  • When a user places a GET request, the values in the table are returned


  • Based on objects and values, easy to read
  • Less number of lines of code generated
  • The syntax is simpler
  • Arrays are supported
  • Data is clearly defined and easier to parse
  • Due to limited number of data types, JSON is not suitable for all applications


  • Use markup tags to make it comparatively easier to read
  • Comparatively more lines of code
  • Complicated syntax, harder to make
  • Data is extracted from tags, harder to parse
  • Wide range of available data types