Domain Objects Configuration

At system startup, domain object settings will determine how entity objects are stored and pre-loaded from the backing store to cache.

Warning: If a backing store is already set up, and you enable any Use Backing Store settings, you must update the backing store setup. The backing store will not operate correctly unless you do so. See Updating Existing Backing Store Schema .

Domain object settings let you configure various behaviors for objects generated by the inference engines and stored in a cache. Many options relate to the way objects move between cache and backing store, so that you can tune memory usage and performance as needed.

You can configure the various behaviors globally (at the default level) and you can set overrides at the object type level. (Not all object level settings, however, are overrides.)

The main options are as follows. (Other options pertain to more specific situations and all are documented in the reference tables.)

  • The mode: Cache plus Memory, Cache Only, or Memory Only. See Cache Modes and Project Design in TIBCO BusinessEvents Architect’s Guide to understand the effect of the different modes. Cache plus Memory (Cache + Memory in the UI) is a deprecated feature.
  • Whether the objects or handles or both are preloaded into the backing store at startup.
  • A preload fetch size (one setting for both objects and handles).
  • Whether the cache is limited or unlimited. If limited, you can specify the size of the cache. See Configuring a Limited (or Unlimited) Cache.

At the individual object type level only, you can also configure the following:

  • Whether the object is stored in the backing store or not.
  • A subscription preprocessor (used for Cache Plus Memory mode only).
  • A backing store table name (used when setting up a backing store. See JDBC Backing Store.

The settings are applied at the object level. For example, a contained concept can have a different limited cache setting from its container concept, and could be evicted from the cache at a different time.