Utilizing business terms and definitions already stored in a business reporting model it is possible to automatically derive a formal business glossary. The first of such tools based on semantic technologies and metadata stored in MMX Metadata Repository now supports SAP Business Objects Universes.

Bottom Up Approach

Many top-down business terminology design initiatives suffer with lack of interest and details or missing touch of realty. The existing reporting environment or the BI system might be good idea to look for business terms, definitions and database object relations. The bottom-up approach might be the fast and right way to document real-life terminology or a solid base for mixed approach and further manual management of the business terminology. We need a methodology and a technology to take the challenges like integration and consolidation of similar terms or BI system inner life and dependencies with database objects.

We introduce the working solution for the SAP Business Objects Universes to collect and consolidate the BO definitions, generate business reporting glossary and create relations between concepts, implementations, reports and database objects. The used methodology based on semantic integration technologies and it's tested many times in real-life environments. The main components of the solution are following:

  • Previously scanned BO Universe and reports metadata.
  • Terms and definition aggregation based on names re-coding rules and similarity estimation techniques.
  • Concept extraction and creation with aggregated definitions.
  • Concepts linking with BO objects to maintain implementation and documentation relations.
  • Concepts linking with reports and database objects to create conditional semantic annotations to technical world.
  • Business Glossary tools for business terminology usage and maintenance.

The following illustration (Figure 1) contains the main idea and the components of the solution and DW environment:

Figure 1. Reporting Glossary Automation and Usage.

Use Case

We point out some figures from the case studies which should illustrate the effect of the automated solution:

  • Business Objects Source:
  • Number of BO Universes: 30+
  • Number of Reports: 40 000+
  • Number of Business Terms/Objects: 8 000+
  • Number of DW Views and Tables used in BO: 2000+

  • Business Glossary Results:
  • Number of consolidated Business Concepts: 2000+
  • Number of Concept and BO links: 8 000+
  • Number of Concept and Report links: 10 000+
  • Number of Concept and DB links: 10 000+

To conclude the use case we can see the 80:20 ratio between the real system definitions and consolidated concepts. The four times smaller amount of the managed business concepts (compared to all BO definitions) is the good start for business glossary management where small amount of meaningful definitions is connected to large amount of technical objects and implementations. Another magic effect comes form the BO formal query definitions that allows us to connect concepts with db views and tables and it gives the meaning and business semantics to the db structures.

Comments

comments powered by Disqus

Features

  • Automated Reporting Glossary derivation
  • BO Universe object definitions consolidation
  • Concept linking to BO terms and reports
  • Concept linking to DB objects
  • Conditional annotations to DB objects

Technical

MMX Framework Business Objects metadata scanner is designed for and works with:

  • BOXI 3.1 and 4.0
  • .UNW and .WID files
  • Derived Tables
  • Linked Universes
  • Multiple Contexts
  • Report query structure fully decoded

Contacts