Hello
I have a SQL Server 2008 R2 Cube with the following objects (only listing the relevant stuff):
1. FactDailyBalances
2. DimAccount
3. DimDate
The objects are linked as a star schema and our users make extensive use of the Date dimension when performing ad-hoc analysis of daily balances.
The Account dimension has two attributes (OpenedDate and CloseDate) which are normal SQL Server Date fields. Because Account and Date are not linked (i.e. no Snowflake), the users aren't able to make use of the many useful features of the Date dimension. In fact, they have now formally requested that they can use the Date dimension in the same way that they do with the fact data.
What is the best design for this requirements? Should I snowflake and link Account to Date? If so, what is the best implementation for this in SQL Server 2008 R2 Analysis Services?
Thanks in advance.
I have a SQL Server 2008 R2 Cube with the following objects (only listing the relevant stuff):
1. FactDailyBalances
2. DimAccount
3. DimDate
The objects are linked as a star schema and our users make extensive use of the Date dimension when performing ad-hoc analysis of daily balances.
The Account dimension has two attributes (OpenedDate and CloseDate) which are normal SQL Server Date fields. Because Account and Date are not linked (i.e. no Snowflake), the users aren't able to make use of the many useful features of the Date dimension. In fact, they have now formally requested that they can use the Date dimension in the same way that they do with the fact data.
What is the best design for this requirements? Should I snowflake and link Account to Date? If so, what is the best implementation for this in SQL Server 2008 R2 Analysis Services?
Thanks in advance.
Amir