Posted in Information Technology & Systems, Total Reads: 811
In information systems, clustering refers to the method of loosely or tightly connecting several servers (computers) so that they can virtually act as a single unified system.
Clustering is chiefly aimed at solving the problems of availability and scalability — quite a crucial requisite in an organisation. Often it happens that several users (need for scalability) want to run the same application at the same time (need for availability), for example, a budgeting system application on the closing day of a financial year. That is it for the user(s) the application should as if from a single server. With many users, this is not possible with a physically single server. So the technique of clustering solves this problem by creating such a unified server in the virtual space.
A schematic of a clustered server system is depicted in the adjoining figure. Bi- directional arrows represent flow of information to and from the server. Thus the benefits of clustering are:
1. Failure-proof:In this system, even if a single server fails, the others run unaffected.
2. Flexibility: Increased load requirements can be simply met by adding additional nodes / servers.
3. Distribution: Load is uniformly distributed among the multiple servers, improving server performance and durability.