A dimension calculation view is mainly used to expose master data from source tables. You can combine multiple data sources, define filters, calculate additional attributes, and create hierarchies to provide a meaningful view of master data. A dimension calculation view is a highly-reusable object. You would typically define a dimension calculation view and share it with other colleagues who would consume it in their calculation views.
Dimension calculation views do not allow measures to be defined. The idea behind a dimension calculation view is to expose only attributes. Every column from the data source is treated as an attribute. This means any numerical column in the source record, such as price or salary, will appear in the result set as an attribute. By default, a full list of every individual value is produced in the output, and not a sum of the values. If you want to sum values then you need a cube calculation view.
If you would like to produce only a distinct list of attribute values, you can specify aggregation on any column. Then each unique value appears only once in the result.
A dimension calculation view does not need to be based only on a single table. You can combine master data tables in a dimension. For example, you could join the customer table to the customer contacts table to generate a comprehensive view of customer information.
Although dimension calculation views can be directly consumed by most reporting tools, it is more likely that your dimension calculation views will be used as objects within cube calculation views, and the cube calculation view will be consumed directly.
Sharing Dimension Calculation Views
Dimension calculation views are reusable objects and can be shared between several cube calculation views.
For example, the product dimension can be used in a purchase order cube calculation view and also in a sales order cube calculation view. Both cube calculation views require information about products and the dimension calculation view provides that.