none
Mask Dimension Attributes based on user

    Question

  • We are implementing a new sales cube using SSAS 2008 R2. One of the business requirements is that users can only view identifiable customer information that belongs to them. In other words, the requirement is to mask customer data that doesn’t belong to the user but not actually hide it. For instance, a sample dataset might look as follows:

    Sales Group

    Customer

    Revenue

    ABC

    $8,900

    Customer 1

    $1,500

    Customer 2

    $5,000

    Customer 3

    $2,400

    XYZ

    $2,000

    Customer 4

    $500

    Customer 5

    $1,500

    If the user only has access to “Customer 1” the other customer attribute data should be masked and aggregated as follows:

    Sales Group

    Customer

    Revenue

    ABC

    $8,900

    Customer 1

    $1,500

    HIDDEN

    $7,400

    XYZ

    $2,000

    HIDDEN

    $2,000

    So, in sales group “ABC” “Customer 1” is still visible but “Customer 2” and “Customer 3” have been merged into a masked Value (i.e. “HIDDEN”), the same is done in sales group “XYZ”. This way their revenue is still visible but the customer name has been masked. The legacy, non-OLAP, reporting environment that we are replacing currently provides this functionality. However, I am currently as a loss as to how to accomplish this is SSAS. How can this be done? TIA.

    samedi 5 mai 2012 07:31

Réponses

Toutes les réponses