Validating OLAP Metadata

To test the validity of OLAP metadata, use the VALIDATE_CUBE and VALIDATE_DIMENSION procedures in the CWM2_OLAP_VALIDATE package. The validation process checks the structural integrity of the metadata and verifies that it is properly mapped to columns in tables or views.

You can determine whether or not a cube is valid by checking the INVALID column of the ALL_OLAP2_CUBES view. You can determine whether or not a dimension is valid by checking the INVALID column of the ALL_OLAP2_DIMENSIONS view.

Structural Validation

Structural validation ensures that cubes and dimensions have all their required components parts.

Cubes

To be structurally valid, a cube must meet the following criteria:

• It must have at least one valid dimension.
• It must have at least one measure.

Dimensions

To be structurally valid, a dimension must meet the following criteria:

• It must have at least one level.
• It may have one or more hierarchies. Each hierarchy must have at least one level.
• It may have one or more dimension attributes. Each dimension attribute must have at least one level attribute.

Mapping Validation

Mapping validation ensures that the metadata has been properly mapped to columns in tables or views.

Cubes

To be valid, a cube's mapping must meet the following criteria:

• It must be mapped to one or more fact tables.
• All of the cube's measures must be mapped to columns in a fact table. If there are multiple fact tables, all the measures must be in each one.
• Every dimension/hierarchy combination must be mapped to one of the fact tables.

Dimensions

To be valid, a dimension's mapping must meet the following criteria:

• All levels must be mapped to columns in a dimension table.
• Level attributes must be mapped to columns in the same table as the corresponding levels.