|
|
Many J2EE applications rely on Enterprise JavaBeans™ (EJBs) to implement key business logic. Therefore, providing a resilient and highly available EJB runtime system is a critical task for any EJB container provider. WAS V6 satisfies this requirement for EJB applications by providing an advanced high availability (HA) solution which guarantees that EJB requests can be serviced continuously even during various types of failures.
EJB clients can be servlets, JSPs, a J2EE client, stand-alone Java applications, or other EJBs. When an EJB client makes calls from within the WebSphere container, client container or outside of a container, the request is handled by the EJB container in one of the cluster members. If that cluster member fails, the client request is automatically redirected to another available server. In IBM WAS ND V6, the EJB HA is achieved by a combination of three WebSphere services: the HAManager, the EJB server cluster, and EJB workload management (WLM).
This section gives an overview and introduction into EJB WLM. For information about Enterprise Java Services workload management, see Chapter 7 "EJB workload management" of IBM WebSphere V6 Scalability and Performance Handbook, SG24-6392.
In WAS V6, the Deployment Manager is not a single point of failure for WLM routing. With the new High Availability Manager (HAManager), a failure of the Deployment Manager triggers the HA Coordinator, which carries the WLM routing information, to failover to any other server in the same HA core group, based on the defined HA policy. Thus, the WLM routing is a guaranteed service that is always available to the client even when a Deployment Manager failure occurs. See Chapter 6, WebSphere HAManager for more information.
The option to configure Backup clusters further enhances EJB application availability. See 2.7, Backup cluster support for more information.
