29-04-2017, 11:50 AM
Load balancing is a method of distributing tasks on multiple computers. For example, distributing incoming HTTP requests (tasks) for a web application on multiple web servers. There are some different ways to implement load balancing. I will explain some common load balancing schemes in this text. The following is a diagram illustrating the basic principle of load balancing:
![[Image: load-balancing-1.png]](http://tutorials.jenkov.com/images/software-architecture/load-balancing-1.png)
The main purpose of load balancing is to distribute the workload of one application on multiple computers so that the application can process a larger workload. Load balancing is a way of scaling an application.
A secondary goal of load balancing is often (but not always) to provide redundancy in your application. That is, if a server in a server cluster fails, the load balancer can temporarily remove that server from the cluster and divide the load on the running servers. Having multiple servers help each other in this way is typically called "redundancy". When an error occurs and the tasks are moved from the server that fails to a running server, it is often called "failover".
A set of servers running the same co-operative application is usually called a "group" of servers. The purpose of a cluster is typically both of the two goals mentioned above: Distribute load on different servers and provide redundancy / failover with each other.
The main purpose of load balancing is to distribute the workload of one application on multiple computers so that the application can process a larger workload. Load balancing is a way of scaling an application.
A secondary goal of load balancing is often (but not always) to provide redundancy in your application. That is, if a server in a server cluster fails, the load balancer can temporarily remove that server from the cluster and divide the load on the running servers. Having multiple servers help each other in this way is typically called "redundancy". When an error occurs and the tasks are moved from the server that fails to a running server, it is often called "failover".
A set of servers running the same co-operative application is usually called a "group" of servers. The purpose of a cluster is typically both of the two goals mentioned above: Distribute load on different servers and provide redundancy / failover with each other.
