Model Transaction Detail For SSAS

  • Dimension_Customer

    CustomerId

    Name

    Dimension_Date

    DateId

    Date

    Month

    Fact_Sale

    CustomerId

    DateId

    TotalSale

    Let’s assume the application that saved the sale accepts a comment from the user (free-form and nothing we want to aggregate by). I would like to save this in the data warehouse so we can use Excel to show the comment for each sale.

    Here is an example of how I would like to see the data in Excel. If this isn’t possible without a dimension then what about “right click” show details? What are the best options to model this?

    January

    TestUser1 $100.00

    *Drill down to comments*

    TestUser2 $200.00

    TestUser3 $300.00

  • If you need to access granular data that is mostly likely at the same level as your fact, there is no simple way to do it with SSAS.

    I've actually had to do something similar to this and the method I used was to add a custom action that would run a SQL query against the SQL DB to return the rowset based on the dimension attributes.

    If this is a must and you have the time to spend on developing it, you need to install ASStoredProcedures[/url] which will allow you to execute stored procedures.

    Have the detail data stored on your fact table.

    Create a stored procedure on your SQL database with the data that accepts dimension attributes (Customer.Name, Date.Month, Date.Date). Based on these attributed, the stored proc should filter out the Fact data and return the limited data set.

    Create an Action in SSAS and use the ExecuteSQL to run the stored proc and pass it the dimension attributes. Setup the action as a Rowset.

    If everything is configured correctly, you should be able to run the custom action in excel and get the data set from the stored proc.

    The concept works, but I built this at my previous company over a year ago so I no longer have access to any of the code I used so I apologize for any inaccuracies.

Viewing 2 posts - 1 through 1 (of 1 total)

You must be logged in to reply to this topic. Login to reply