This document demonstrates multiple Data Access Control (DAC) scenarios using different combinations of State and Category filters. Its purpose is to explain how DAC works across multiple scenarios and how access rules can be designed according to specific business requirements. Based on the requirement, permission tables can be created and DACs can then be defined using those permission tables. The cases show how single and multiple DACs behave, including OR and AND logic, unrestricted values, exclusions, overlapping access groups, and conflicting rules that return no data. Screenshots are included to illustrate each configuration and its result.
Implementation steps:
Step 1: Create the permission table.
Step 2: Create the DAC based on the permission table.
Step 3: Assign the DAC to the relevant views or analytical models.
I have not included detailed instructions for Steps 1 to 3 because they are already covered in many blogs on the SAP Community. This document assumes that readers working with SAP Datasphere are familiar with creating permission tables and DACs and assigning DACs to views or analytical models.
Refer this link for best practices : https://help.sap.com/docs/SAP_DATASPHERE/be5967d099974c69b77f4549425ca4c0/a032e51c730147c7a1fcac125b4cfe14.html?locale=en-US#loioa032e51c730147c7a1fcac125b4cfe14__section_best_practices
Note: As a best practice, create a dedicated permission space to store permission tables. Govern access by sharing the space only with the appropriate developers who are responsible for maintaining row-level security.
Dataset: The examples use the open Sales Data Set available on Kaggle.
Testing and privacy note: Each case was tested using a single User ID. User ID is hidden in the screenshots because they are personal information.
In Sales Dataset , we have state and Category dimensions holding data for multiple US states and multiple Categories.
Case 1: State = California or Alabama and Category = Furniture
To show data where State = California and Category = Furniture OR State = Alabama and Category = Furniture
Condition
(State = California AND Category = Furniture)
OR
(State = Alabama AND Category = Furniture)
SAC output
Case 2: All California Categories or Alabama Furniture
Condition
(State = California AND Category = *)
OR
(State = Alabama AND Category = Furniture)
SAC
Case 3: All Categories for California or Alabama
Condition
(State = California AND Category = *)
OR
(State = Alabama AND Category = *)
Case 4: Combining Two DACs on the Same View
DAC 1 ( Type Operator and Values )
Condition :
(State EQ Alabama AND Category = *)
OR (State EQ California AND Category = *)
DAC 2 (Type Single Only)
Condition:
(State EQ Alabama AND Category = Furniture)
OR (State EQ California AND Category = Furniture)
Explanation: When two DACs are assigned to the same view, they are combined using AND logic. A row is shown only if both DACs allow it; therefore, the more restrictive DAC takes precedence.
Result: SAC will show California Furniture or Alabama Furniture.
Case 5: States from Alabama to California, Excluding Arizona
Requirement: User A should see states from Alabama to California, excluding Arizona
Explanation: The configuration below excludes Arizona and all categories associated with it.
Case 6: All States and Categories Except California Furniture
Requirement: Show all states and categories except the Furniture category for California.
Case 7: California with All Categories
Explanation: This is an alternative configuration in which the Category does not need to be specified explicitly.
SAC
Case 8: Overlapping Groups Without Duplicate Rows
Requirement: Show California with all categories, together with all states that have the Furniture category.
Explanation: Group 1 and Group 2 overlap, but the overlap does not create duplicate rows.
Case 9: Conflicting DACs Assigned to One View
DAC 1 ( Type : Single value ) : Show State = Alabama and Category = Furniture.
DAC 2 Type ( Operator and values ) : Show State = California and Category = All.
Result: If DAC 1 and DAC 2 are both assigned to a single view, no data will be shown.
Explanation: Each row must satisfy both DACs. DAC 1 allows only Alabama, while DAC 2 allows only California. Because a row cannot belong to both states at the same time, no rows pass the combined filter.
Comparison with SAP BW Analysis Authorizations
>Aggregation Authorisation in SAP BW
In SAP BW, aggregation authorisation using the colon (:) can allow a user to view an aggregated result for a characteristic without granting access to its individual values. SAP Datasphere DACs work differently: they apply row-level filters, removing unauthorised records before the result is aggregated.
Example
|
Cost Center |
Employee |
Salary |
|
CC100 |
Mark |
100 |
|
CC100 |
Bob |
100 |
|
CC100 |
John |
100 |
SAP BW result: With “:” authorisation for Employee, the user can run a query by Cost Center and view the complete aggregated value for CC100: 300.
If the user drills down by Employee, SAP BW blocks access to the individual employee values. The user can therefore see the full total without viewing the underlying salary details.
SAP Datasphere DAC result: A DAC filters out rows that the user is not authorised to access. If the user is authorised to view only Mark’s record, the rows for Bob and John are excluded before aggregation. The value shown for CC100 is therefore 100.
Key takeaway: SAP BW Analysis Authorizations can support aggregated reporting while restricting access to lower-level characteristic values. A SAP Datasphere DAC controls which rows are available to the user, so excluded rows do not contribute to the aggregate. A standard DAC therefore does not provide the same behaviour as BW aggregation authorisation.
>Key-figure (column-level) security:
SAP BW can restrict key figures through 1KYFNM, allowing sensitive measures to be hidden from selected users. SAP Datasphere DACs provide row-level security only. To achieve a similar separation in SAP Datasphere, separate analytical models may be required for different user groups.
>Hierarchy authorisations:
SAP BW supports authorisation at hierarchy-node level, including level restrictions, aggregated-only display, and hierarchy versions and validity dates. SAP Datasphere offers more limited hierarchy-based access control through DACs.
>Define once and enforce broadly:
In SAP BW, an authorisation-relevant characteristic is checked across the InfoProviders and queries that use it. In SAP Datasphere, a DAC must be assigned to each relevant view or analytical model. If a required assignment is missed, the data may remain unprotected.
Summary
SAP Datasphere DACs are well suited to straightforward row-level filtering at scale. SAP BW Analysis Authorizations are stronger where the requirement includes partial visibility, such as totals without underlying details, hidden key figures, or more advanced hierarchy authorisations, as well as strict audit and compliance controls. In a migration from SAP BW to SAP Datasphere, aggregation authorisations and key-figure authorisations require deliberate redesign rather than a one-to-one mapping.
Read More Technology Blog Posts by Members articles
#abap