Mastering Data Management

More in Business Process Management

• Improving Business Processes with MDM
• Mobility: Inventing the New Division of Labor
• Take Charge of your Data Environment: IBM Performance Management Solutions
• Top 5 InfoSphere Demos at IOD

More in MDM

• Top Books for your MDM Library
• Recent Reads: MDM & Data Governance
• Big Data Governance: A Framework
• Predictive Data Governance for Business Transformation

More from Lawrence Dubov

• MDM Vs Data Governance: History and 8 Predictions
• What’s Harder: Starting or Sustaining MDM?
• Conclusion: Quantifying Assets
• 3 Essential Questions for Data Quality: Benchmarks, Penalties & Latency

Share this Post

Recent Comments

• Lorraine Fernandes on Response: A Universal Identifier Isn’t a Silver Bullet
• Henrik Liliendahl Sørensen on Response: A Universal Identifier Isn’t a Silver Bullet
• Response: A Universal Identifier Isn’t a Silver Bullet | Mastering Data Management on 3 Best Practices in Identity Management

MDM takes the science and art of data warehousing dimensions and ODS development to the next architectural level.

On many engagements we are asked how Master Data Management relates to business intelligence, data warehousing (DW) and operational data store (ODS) concepts developed in the 80s and 90s. Many enterprises have developed their enterprise information programs based on these key concepts.

A DW is typically thought of as the key construct optimized for reporting. On the other hand, ODS, with a number of ODS types known, is a database managing consolidated transactional information.

DW data models are oftentimes de-normalized and organized in star schemas for higher data retrieval performance while ODS models are built with the third normal form (3NF) in mind to support transactional processing.

So, where does MDM fit? How does it coexist with DW and ODS? What happens to pre-existing or under-development DWs and ODSs if we implement MDM?

From the data warehousing perspective over 80% of the data warehousing complexity and effort reside in building high quality data warehousing dimensions. The value of a data warehouse may turn marginal if the data warehouse has duplicate suspects, products etc.

The same applies to relationships and hierarchies. Errors in relationships and hierarchies cause incorrect aggregated values in data warehousing reports, which may cause significant business issues. Many of our MDM implementations (analytical MDM) enable data warehouses by building complex dimensions and their internal structures (relationships and hierarchies within dimensions).

MDM takes the science and art of data warehousing dimensions and ODS development to the next architectural level. In particular this includes:

• Master data modeling – Master data modeling styles can be optimized for entity resolution as opposed to dimensional browsing. Typically both modeling styles should be considered
• Advanced statistical methods - Applying advanced statistical methods and adaptive learning to resolution of entities and relationships achieves greater accuracy and efficiency than typically attained with a traditional ETL.
• Data hubs with ODS - The notion of a data hub significantly enhances the notion of its legacy parent ODS. That said, it should be considered on a case by case basis if a data hub totally eliminates the need for ODS or takes only part of its functions and both constructs should coexist.
• Data hubs with SOA - Data hubs have evolved as services oriented constructs and demonstrate a great agility and synergy with SOA. It is a good idea to think of a data hub as a service actively resolving entities and relationships other than just a data repository filled with data cleansed by ETL outside of the data hub

Overall data hubs have evolved as constructs proven to be able to considerably enrich the big DW/ODS picture developed in the 80s and 90s. MDM and data hubs specifically do not diminish the well known parents like DW, ODS or ETL but rather enhance them bringing the science and art of information development to the next level.