You are here
Computing Models
Computing models refer to different approaches to how data is processed.
- Mainframe
This is the oldest model, and uses a completely centralized approach. It uses single, large computer that everyone shares. This central computer is kept in a data centre, and users use it by means of remote thermals. - File server
Personal Computers (PCs) are connected by a Local Area Network (LAN) to a File Server. The file server is like the hub of the network. Data stored on the file server can be shared by other PCs on the network. All the the actual manipulation of the data is done by the PCs. The server merely acts as storage. As the number of PCs grows, this leads a problem called contention, when two PCs try to change the same piece of shared data simultaneously. - Client-Server
This builds on the file server model, but now the server shares in some the processing of the data. Think of it as a requester-responder model. The PC, called the Client, requests data from an application, usually a database, that is running on the server. The server send the data back to the client. The server places requests into a queue. Each request is handled in sequence, eliminating the contention problems of file server computing. - Cloud
Cloud computing is a form of outsourcing where all or part of the computing process is handled by a third-party service provider. The name comes from the use in diagrams of a cloud image to represent a complex network of components as an abstraction. The intent being that details of those components didn't matter.