Load balancing: Concepts

Load balancing distributes information requests to multiple servers. Busy sites typically employ two or more servers in a load balancing setup. Requests are distributed to multiple servers to make the workload for each server equal and to not overburden one particular server causing it to fail. One of the benefits of a load balanced cluster is that the redundancy of multiple Application Servers and Curators often reduces or eliminates downtime.

A load-balanced cluster provides redundancy and practically infinite scalability. A load-balanced cluster can consist of as little as two servers, or many servers. The term load balancing refers to incoming traffic distribution only; load balancing does not include back-end functionality such as data replication or mirroring. For information on replication and mirroring, refer to Replicating ADAM.

How can a load balancer be used in the IMPAX cluster?

A load balancer can be placed between the Clients and the Application Servers to manage the requests being sent to the Application Servers. For example, a user requests image data through the Client. The request is sent to the load balancer, and the load balancer sends the request to the appropriate Application Server. Because of the load balancer multiple Application Servers can be used to process requests from a larger number of Clients.

For additional information on including a load balancer in the IMPAX cluster, refer to Single cluster: Understanding how IMPAX components interact.