Application Service at the Edge enables you to optimize resource use and plan for high traffic scenarios by strategically dividing application processing execution between Edge Servers and back-end application servers. In Application Service at the Edge, some of the servlets and JSPs that make up the Web components of a J2EE (Java 2 Enterprise Edition) application are executed on a WebSphere Application Server that is colocated with the Caching Proxy. Other application components are directed to a WebSphere Application Server that is located at the back end.
To use Application Service at the Edge, you must determine which components of your Web applications can potentially be executed at the network edge, and which must executed at the origin application server. (Application developers and system architects usually must be consulted to make this determination because it requires understanding of what resources are available and how they are used.) The application then is reconfigured as two different application archives (.ear, or EAR files): one that will be used on the Edge Server and one that will be used on the back-end application server. Tools are provided to help you perform this division; refer to the Caching Proxy Administration Guide for details. Note that this reconfiguration must be repeated if any Web components in the application are modified.
Refer to the Caching Proxy Administration Guide for detailed information about system requirements and configuration steps for Application Service at the Edge. The required configuration steps include the following:
The configuration forms included with the Caching Proxy contain settings for the Application Router, which is a plug-in to the Caching Proxy. The Application Router serves as a proxy for application execution requests when Application Service at the Edge is enabled. It has the ability to route some processing requests to the back-end server and some to the Edge Components, then consolidate the results for return to the requesting client.