This section explains the overall procedure of configuring Cache object management, with references to other sections with more details. The ordering of tasks shown below is a suggestion only and is not a requirement.
● Work out a detailed plan for your project architecture. It’s good to begin with a simplified form of your final plan, for example one that does not use multiple instances of inference agents, and test before adding complexity.
For AIX only You must add the following property to all engine property files (be-engine.tra) files:If all defaults are appropriate for your environment, you need only specify the cluster name. Default multicast property values handle cluster member discovery. However, other options are available for use in different circumstances.You add the same set of cache cluster discovery settings to property files of all nodes in the cache cluster.In this step you configure certain cluster characteristics such as what caching scheme to use, whether a backing store is used, and the number of cache servers that must be running before the rest of startup actions occur, and cache logging properties. If Decision Manager is used, you also define some properties used for deploying virtual rule function implementations to BusinessEventsWith a few exceptions, you add the same set of cluster-wide properties with the same values to property files of all nodes in the cache cluster.Also see Chapter 17, Understanding Cache OM and Multi-Engine Features and Chapter 18, Understanding and Working With Cache Modes, especially Design Constraints With Cache Only Cache Mode.After you finish configuring and testing your caching solution, you may want to add a backing store for data persistence. See Chapter 24, Setting up a Backing Store Database.
Copyright © TIBCO Software Inc. All Rights Reserved.