Datasphere is SAP's latest data warehouse product. As a software-as-a-service solution, it is based on the new HANA Cloud Services. In the short or long term, Datasphere should be able to map a full-fledged enterprise data warehouse in addition to a self-service data preparation use case.
SAP is investing heavily in the integration of both solutions. The soon-to-be-available "Model Transfer Connection" will also enable the transfer of BW semantics from queries. This will allow semantics defined in SAP BW to be reused. We will closely monitor the new developments and provide regular updates here. The following recorded SAP WebSession provides initial insights.
Our scenario
In our very simple scenario, we have two tables with transaction data containing sales and costs per retail area, in this case simply city.
The retail team would like to have a clear overview of how much profit is generated per time period in a specific city.
Data model in SAP BW/4HANA
In SAP BW/4 HANA, the following simple data model exists for transaction data:
The aDSO cost has the following simple structure:
Structure of aDSO's turnover:
Our integrative SAP BW/4HANA scenario
In our fictional, very simple case, let's imagine that SAP BW does not contain any master data for the cities. However, a department has a CSV file with the translation of IDs into descriptions of the cities. In addition, the "Profit" column is to be calculated in Datasphere based on the BW/4 and CSV data. We have already created a table for the CSV file in Datasphere (see also: Datasphere – Introduction to Data Modeling).
Join the two composite providers
To integrate the data from BW/4 into Datasphere, you must first set up the connection to BW/4, as described in the post "Datasphere connection to local systems." Once the connection is established, you can start building a data model in Data Builder, regardless of the source of the Datasphere. In our case, we create a graphical view:
Navigate to remote sources → Select connection BW/4 or HANA
Add both CompositeProviders to the Graphical View.
Hide the "InfoProv" and "Recordmode" columns for CP_EXP and CP_REV respectively.
Check the join condition of the 2 CompositeProviders.
Hide the "City" and "Date" columns as they are duplicates.
Et voilà:We have connected two SAP BW/4 HANA composite providers and joined them together. Uniform reporting is now possible.
Join the BW/4HANA data with the mapping table
We have stored the names of the cities in a Datasphere table, which is populated from a CSV file (for details, see also here). We can easily add this "City_Names" table to our graphical view:
Hide the last "City" column, which only contains the abbreviations for the cities.
Creation of a calculated key figure based on the consolidated data
Our model is still missing the profit metric. To implement this, we create a new "Calculated Column":
Please note: To use the Graphical View in stories, it must be deployed as an Analytical Data Set.
We also hide the "Cost_Amount" and "Revenue" columns, as only the profit is to be taken into account in the subsequent evaluation.
Visualization of data directly in the Datasphere
Based on our very simple mixed data model, we can quickly visualize the data using the integrated Storybuilder in SAP Analytics Cloud (SAC):
Summary and outlook
Christopher Kampmann
Head of Business Unit
Data & Analytics
christopher.kampmann@isr.de
+49 (0) 151 422 05 448
