Chapter 1 Introduction to ICEfaces Enterprise Edition

Chapter 1 Introduction to ICEfaces Enterprise Edition

ICEfaces Enterprise Edition (ICEfaces EE) is based on the ICEfaces Community Edition (ICEfaces CE), augmenting that product with several key capabilities. The philosophy behind ICEfaces CE is to provide complete rich application development capabilities to the developer. ICEfaces Community and Enterprise Editions share the same core framework, component suite, and public APIs. Applications developed with ICEfaces CE are 100% compatible with ICEfaces EE. All ICEfaces CE documentation is relevant to ICEfaces EE, so you should be familiar with that documentation prior to investigating ICEfaces EE features described in this document.

The key differentiators of EE over CE relate to large-scale enterprise deployments of ICEfaces applications. In particular, the following features are described in this document:

Connection Management and Heartbeating - ICEfaces EE provides enhanced management of client/server connections. Key features like heartbeating, improved robustness of connections in enterprise deployments, and additional error handling facilities are also available. Refer to Chapter 2, Connection Management, for details describing heartbeat configuration and connection error handling.
Asynchronous HTTP Server Configuration and Deployment - The Asynchronous HTTP Server is required to support large-scale deployments of asynchronous mode ICEfaces applications. Refer to Chapter 3, Asynchronous HTTP Server, for details describing configuration of the server and front-end Apache web server.
Clustered Deployments - ICEfaces EE supports clustered deployment of large-scale, ICEfaces applications. There are two considerations for clustered deployments that use the asynchronous mode of ICEfaces:

The Asynchronous HTTP Server must be able to communicate with application servers on all nodes of the cluster, so a cluster-enabled JMS configuration is required.
The cluster-enabled Broadcast Render Manager must be used where group renderers include clients that are distributed across multiple nodes in the cluster.