Skip to main content

19.1 BDC Mental Model

1. The old mental model​

A traditional landscape often looks like:

SAP systems -> BW -> BI
                 |
              extracts
                 |
          cloud platforms

The result can be duplicated data, duplicated semantics and many integration contracts.

2. The BDC mental model​

SAP + third-party data
          |
     integration
          |
  business context
          |
  governed data products
          |
 analytics / AI / applications

SAP positions BDC around unified and governed SAP and third-party data, while Datasphere provides core integration, modeling and semantic capabilities.

3. Five architectural layers​

Source layer​

Systems of record and external platforms.

Data foundation​

Integration, persistence and harmonization.

Knowledge layer​

Business semantics, metadata, relationships and governed products.

Consumption layer​

Analytics, planning, applications and AI.

Operating layer​

Security, lifecycle, monitoring, governance and FinOps.

4. Why context matters​

A table called CUSTOMER is not enough.

An enterprise needs to know:

  • what customer means,
  • which system owns it,
  • how it relates to sales,
  • what attributes are trusted,
  • which transformations were applied,
  • who may consume it.

5. Architect exercise​

Take a BW query and identify:

  1. source systems,
  2. physical data,
  3. business semantics,
  4. reusable data product candidates,
  5. consumers,
  6. governance controls.

Then redesign it using the BDC mental model.