Overview:
Consider a migration scenario of analytical reports from SAP BW to S4H Embedded analytics. There will be lot of customizations and it’s difficult to replace all of them in short time. It will be good to use CDS views marked with data extraction as a source to embedded BW objects and use Bex features on top of these to develop analytical reports referring to existing BW system. After successful conversion, analytical reports which can be replaced by CDS consumption views can be identified.
I am considering sales scenario, to identify Quotation, Order & Contract amounts by Customer and Sales value by Material.
I will develop CDS views (Extraction capable) & BW objects as one flow. I will develop CDS Views (Consumption) as another flow.
Then validating the results of both approaches will help in deciding the right source of analytical reports.
Architecture:
Step by Step Process:
I. Create Basic views for all the required dimensions with required attributes.
II. Create View for Sales order Header and associate with Customer. Mark this view for Data extraction
Below annotations are required to mark the view for data extraction
@Analytics.dataCategory: #FACT
@Analytics.dataExtraction.enabled: true
III. Create Sales order Item and associate with Material. Mark this view for Data extraction
IV. Create Data sources based on CDS views enabled for data extraction
Select ODP - ABAP CDS Views as Source System type
Select CDS views created for Data extraction
Data source based on Sales Header CDS view was created
Repeat the same steps to create the data source on Sales Item CDS view
Data sources queues will be identified in ODQMON
V. Develop Open ODS views referring data sources created on CDS views earlier
BW Modeling perspective is required in Eclipse to carry out these steps.
Select source Type 'Data source' and select the data sources created in previous step to get the fields into Open ODS
Create Open ODS views for Sales Header & Item data.
Net Price will set with aggregation 'Maximum' to check the behavior.
No need to generate the data flow as we will use these objects as logical layer
VI. Include these open ODS views into Composite provider
Aggregation to be set again at composite provider if it is different from 'Summation'.
VII. Develop Bex queries on top of the composite provider
Bex queries were also developed in Eclipse editor.
RKFs were created to get Order, Quotation & Contract amount separately.
VIII. Develop CDS views on top of basic views created and mark them for analytics
Composite views were built to pass parameters for currency conversion & Union of different document types.
Attached document has details of the CDS views Z_PY_SALES & Z_PY_Sales_Itm
To mark the views for consumption below annotations were required.
@VDM.viewType: #CONSUMPTION
@Analytics.query: true
Output of these queries can be verified in Fiori launch pad.
We can validate the same using RSRT by prefixing with '2C'. Output for the consumption views was shown in next step.
CDS view - Sales value by Customer CDS view - Sales value by Material
Source - Sales Header Data Source - Sales Item Data
IX. Validate the data from CDS & Bex queries output
Customer Sales Validation between CDS & BEx
Sales Value & Price validation between CDS & BEx using Sales Item data flow
As both are matching these reports can be replaced by CDS views easily. The reports referred in example are quite simple but analytical reports in BW
will have lot of logic in transfomrations & Bex Query Designer. Hence this approach will help in identifying the analytical reports which are good for converting to S4H Embedded Analytics using CDS consumption views.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
7 | |
4 | |
3 | |
2 | |
1 | |
1 | |
1 | |
1 | |
1 |