CDD Agent Classes Tab Settings Reference
The following tables explain settings used with inference agents, and query agents.
Setting | Notes |
---|---|
Inference Agent and Query Agent Settings | |
Max Size (Local Cache) | Specifies the maximum number of objects (entities) in each agent’s L1Cache (inference agent) or local cache (query agent). The L1 cache is a local cache used by the inference agent for local access to recently used objects. It is used to optimize access to objects.
The query local cache is used in a way similar to the inference agent L1Cache. The query agent’s local cache stores cache data locally for efficient reuse. The local cache listens to and synchronizes the locally stored entity instances with those in the main cache, so that the local cache stays up-to-date. When the threshold is reached, oldest entities are removed first. Default is
|
Eviction Time (Local Cache) | Specifies an age limit on the cached entities in seconds. After this period, they are removed from the local cache.
Note
Age resets each time an entity is accessed by a query engine.
Default is
|
Queue Size (Shared Queue) | Used for destinations whose threading model is Shared Queue (see Threading Model in
CDD Collections Tab Input Destination Settings Reference ).
Specifies the queue size for the processing unit-wide shared queue. Note
In this release, set the same value for all agents configured to deploy in the same processing unit.
You can also use global variables as values for this setting.
If set to 0 (zero), the queue size is unlimited. Default is
|
Thread Count (Shared Queue) | Used for destinations whose threading model is Shared Queue (see Threading Model in
CDD Collections Tab Input Destination Settings Reference ).
Specifies the number of processing unit-wide shared threads. Note
In this release, set the same value for all agents configured to deploy in the same processing unit.
You can also use global variables as values for this setting.
As a guideline, set the value to the number of processors available to the JVM. Default value is
|
Max Active | Specifies the maximum number of active agents of this class. This value is used for fault tolerance. Deployed agents that are acting as standby can take over from active instances that fail. In many cases, there is no need to keep standby instances.
A value of
For example, the Max Active field is set to 1 and you start two BusinessEvents LiveView engine. As per the Max Active field, only one engine is active while the other one is passive. If the active engine fails, the passive engine takes over and become active. See Fault Tolerance of Agents in TIBCO BusinessEvents Architect’s Guide for more details. Default is
|
Inference Agent Settings | |
BusinessWorks Repo URL | If this project will integrate with a TIBCO ActiveMatrix BusinessWorks project, enter the Repo URL for the ActiveMatrix BusinessWorks project repo URL here.
Use forward slashes. |
Concurrent RTC | If checked, enables concurrent run to completion cycles, generally shortened to RTC cycles. (Also known in prior releases as concurrent Rete and
concurrentwm ).
Concurrent RTC does not require cache OM but does require local locking. The number of concurrent cycles is determined by the number of available threads. See Collections, Agent Classes, and Processing Units for details. Also see Concurrency and Project Design in TIBCO BusinessEvents Architect’s Guide for important information on using concurrency features. |
Check for Duplicates | By default,
TIBCO BusinessEvents checks if the external IDs (@extId ) of entities are unique within the current agent. To check for uniqueness of external IDs across the cluster, check this check box. Performing this check affects performance.
Default is unchecked. |