Repository objects that process data from DBMS, use various caching types to speed up the work. If an object is parametric, the samples, for which parameters set are saved. On opening the object the required data instance will be taken from cache. If any data was changed, it will be updated in cache and DBMS. On changing the object structure the use of saved cache samples is not available.
NOTE. It is recommended to delete saved samples using the Clear Cache button after the structure change. After that instances will be saved for a new object structure.
All cube types (except calculated), ADOMD cubes and time series databases use file cache. For virtual cubes the cache is not created, cube source cache is used instead.
The cache management is executed on the Cache tab in the object properties:
Various types of dictionaries and MDM dictionaries use their own caching mechanism. The tab to manage dictionary cache looks as follows:
NOTE. The database to store cached objects must be based on one of the following DBMS: Microsoft SQL Server, Oracle, Teradata, PostgreSQL/Postgre Pro. DBMS version must correspond to one of the supported versions.
See also:
Cache objects