WebLogic Server Load Balancing

Load Balancing in WebLogic Server 

Load balancing is about distributing work across multiple servers within a cluster. Ideally you want this work to be distributed evenly, to the capabilities and load on each server within the cluster. WebLogic is able to load-balance requests to members within a cluster when certain conditions are satisfied.

First, WebLogic requires that (most) components be deployed homogeneously to each server in the cluster. In the multi-tier application setup, the web application is targeted to all servers that belong to the presentation tier cluster. Because of this, incoming requests can be distributed evenly across the different members of the presentation tier. Similarly, the EJB components are available on all members of the object tier cluster. Because of this, incoming EJB calls from the presentation tier (or any external client) can be distributed across the different servers in the object tier.

When a component is deployed to a cluster, multiple copies of the component will be sitting on the different servers comprising the cluster. WebLogic uses the cluster-wide JNDI tree to record the availability of these replicas in the object tier, whereas a proxy plug-in typically holds the information about the availability of servers hosting the servlets and JSP pages.

WebLogic also must be aware of the status of the components that have been deployed to the cluster. This information forms the basis on which a load balancer can decide which server instances should be used when a request is made to the cluster. WebLogic uses a variety of methods to determine the health status of servers, including multicast heartbeats. By knowing when a server instance is under heavy load while the others aren't, a load balancer is able to direct new requests to the other members of the cluster. Hardware load balancers often have this capability