Restrictions for caching in Data Virtualization for IBM Cloud Pak for Data

Restrictions for caching in Data Virtualization

When you create a cache in Data Virtualization, you must consider certain restrictions.

Caching in Data Virtualization is subject to the following restrictions:

You can't use the fetch first n rows clause.
You can't use the order by clause.
You can't use select statements that use the with clause.
Data types that are not supported:
- BINARY (BINARY is not supported; however, VARBINARY is supported)
- CHAR FOR BIT DATA, VARCHAR FOR BIT DATA
- DECFLOAT
- GRAPHIC, VARGRAPHIC
- LOB types (BLOB, CLOB, DBCLOB)
- LONG VARCHAR, LONG VARGRAPHIC (These types are deprecated in Db2®)
- TIME
- XML

Data Virtualization caching is built on Db2 Big SQL capabilities that involve Hadoop MQTs. The following restrictions apply.

You must be aware of restrictions in the CREATE TABLE (HADOOP) statement. For more information, see CREATE TABLE (HADOOP) statement in the Db2 Big SQL product documentation.

Data Virtualization caching is also built on Db2 MQT capabilities. The following Db2 MQT restrictions apply. For more information, see Materialized query table restrictions in the Db2 product documentation.

You must be aware of fullselect component restrictions. For more information, see fullselect in the Db2 product documentation.
If IBM Knowledge Catalog policy enforcement is enabled in Data Virtualization, direct access to cache tables (MQT tables with names that start with DV* in the CACHESYS schema) is denied to all users.