Reference data caching overview
Data caching occurs when the records of a transaction are cached in the JVM.
Starting in Yantra 5x 5.0 SP2, caching is enabled by default. The cached records
are stored in the JVM heap. Typically, around 350MB will be used by caching during steady-state. Of
course, the actual amount could depend on your operations. As a result, with caching enabled, you
should monitor the health of the JVM heap garbage collections.
Note: To monitor the health of the
JVM heap garbage collections, you can analyze VerboseGC logs. Whenever application servers are
started, the VerboseGC logs are automatically collected, if the Enable
VerboseGC property is set to true in the configuration for the
environment in the IBM UrbanCode Deploy Selfserv tool. For more information about exporting
application logs, see Exporting application logs. However, when you run the agent and
integration servers, you can select the Enable verboseGC output to collect
the verboseGC logs. The exported logs are copied to the
For memory constrained
environments, you may want to enable caching on specific tables./home/<tenant_code>/logs/<env_name>/oms_logs directory of the drop server.
You can analyze it to get additional details for GC and memory usage.