...
NONE - No caching is used at the BI Server level. This may result in slower queries.
AUTO - The system will determine which query results should be cached when the execution time exceeds the QUERY_CACHE_TIME_THRESHOLD.
For materialized cubessemantic models, all query results are cached for SQL, MDX, and Kyvos visualization queries.
For a non-materialized cubesemantic model, only the SQL metadata queries are cached. No user queries will be cached.
For HOLAP cubessemantic models, the results of:
metadata queries will be cached.
user queries having only materialized dimensions and measures will be cached.
user queries having non-materialized dimensions or measures will only be cached when the partitioning strategy (Sliding Window) is defined on the cubesemantic model.
ALL - Enables query caching for all SQL, MDX, and Kyvos visualization queries when the execution time exceeds the QUERY_CACHE_TIME_THRESHOLD.
...
Recommendation:
Using AUTO is highly recommended. Choosing ALL may return old values on non-materialized cubessemantic models. For additional performance improvement with data validated for accuracy, set the property value as ALL. Set the value to NONE, only if performance is not an issue, and the BI Server has minimal resources.
...
| Info |
|---|
NOTE
|
...