You design a SQL Server 2008 Analysis Services (SSAS) solution. The data source view has tables as shown in the exhibit. (Click the Exhibit button.)
The FactInternetSales measure will be queried frequently based on the city and country of the customer.
You need to design a cube that will provide optimal performance for queries.
Which design should you choose?
A.
Create two dimensions named Customer and Geography from the DimCustomer table and the DimGeography table, respectively.
Create a materialized reference relationship between the Geography dimension and the FactInternetSales measure by using the Customer dimension as an intermediate dimension.
B.
Create two dimensions named Customer and Geography from the DimCustomer table and the DimGeography table, respectively.
Create an unmaterialized reference relationship between the Geography dimension and the FactInternetSales measure by using the Customer dimension as an intermediate dimension.
C.
Create a dimension named Customer by joining the DimGeography and DimCustomer tables.
Add an attribute relationship from CustomerKey to City and from City to Country.
Create a regular relationship in the cube between the Customer dimension and the FactInternetSales measure.
D.
Create a dimension named Customer by joining the DimGeography and DimCustomer tables.
Add an attribute relationship from CustomerKey to City and from CustomerKey to Country.
Create a regular relationship in the cube between the Customer dimension and the FactInternetSales measure.
Explanation:
Tip: “design a cube … optimal performance” = “a dimension … City to Country”