basic administrative architecture

This article discusses basic concepts in the administrative architecture to help you understand system administration in a WebSphere Application Server environment. The fundamental concepts for WAS administration include software processes called servers, topological units referenced as nodes and cells, and the configuration repository used for storing configuration information.

Servers perform the actual running of the code. Several types of servers exist depending on the configuration. Each server runs in its own Java virtual machine (JVM). The application server is the primary run-time component in all WAS configurations. All WebSphere Application Server configurations can have one or more application servers. In some configurations, each application server functions as a separate entity. No workload distribution or common administration among application servers exists. In other configurations, workload can be distributed between servers and administration can be done from a central point.

A node is a logical group of WebSphere Application Server-managed server processes that share a common configuration repository. A node is associated with a single WebSphere Application Server profile. A WebSphere Application Server node does not necessarily have a one-to-one association with a system. One computer can host arbitrarily many nodes, but a node cannot span multiple computer systems. A node can contain zero or more application servers.

The configuration repository holds copies of the individual component configuration documents that define the configuration of a WebSphere Application Server environment. All configuration information is stored in .xml files.

A cell is a grouping of nodes into a single administrative domain. A cell can consist of multiple nodes, all administered from a deployment manager server. When a node becomes part of a cell (a federated node), a node agent server is installed on the node to work with the deployment manager server to manage the WAS environment on that node.

When a node is a standalone node, not part of a cell, the configuration repository is fully contained on the node. When a node is part of a cell, the configuration and application files for all nodes in the cell are centralized into a cell master configuration repository. This centralized repository is managed by the deployment manager server and synchronized to local copies that are held on each node. The local copy of the repository that is given to each node contains just the configuration information needed by that node, not the full configuration that is maintained by the deployment manager.

WebSphere Application Server types

This section discusses the three server types that interact to perform system administration.

Application Server: A WAS provides the functions that are required to support and host user applications. An application server runs on only one node, but one node can support many application servers.

Node agent: When a node is federated, a node agent is created and installed on that node. The node agent works with the deployment manager to perform administrative activities on the node.

Deployment manager: With the deployment manager, one can administer multiple application servers from one centralized manager. The deployment manager works with the node agent on each node to manage all the servers in a distributed topology.

The following diagram depicts the concepts that are discussed in this article.

The concepts that are discussed in this article form the basis of WAS administration. More detailed descriptions can be found in other sections.


Related Tasks

Setting up the administrative architecture




WebSphere is a trademark of the IBM Corporation in the United States, other countries, or both.
IBM is a trademark of the IBM Corporation in the United States, other countries, or both.