Data Vault: A Question of Early or Late Integration with M. Müller (English)

We’re happy to finally welcome meetup regular Michael Müller as a speaker. He’s a data vault expert based in Germany and the chairman of the German-speaking Data Vault User Group (DDVUG).

Abstract:

In his book ‘The Elephant in the Fridge’, John Giles laid out a very good and workable business driven strategy for data vault: Model the business objects and create your data vault from it. Wherever the source doesn’t match, correct it through business rules.

Many data warehouse automation tools ingest the sources, create the data vault from there and them you apply the business rules in order to get a business-oriented view. Often there is a lot of integration work to be done as well. The reality of the sources often leads either to a lot of logic in the business vault with interdependent steps and quite a bit of complexity – or to a total disregard of the business viewpoint. And in result to very badly integrated data with a lot of maintenance problems. An earlier integration will help with these problems.

Let’s take a closer look on the issue of early and late integration:
• What can be done in an earlier stage (or even in the stage) within the rules of data vault without changing data itself?
• What are possible scenarios?
• Which kind of patterns can we identify?
• What are the dangers?
• What are the rules for this kind of early integration?

Speaker:

Michael Müller has been working in the BI field for 20 years. His focus is on data modeling and data architecture as well as on Data Vault, DWH automation and embedding BI into the company. He is currently on the board of the German-speaking Data Vault User Group (DDVUG) e.V.

His motto: “Methods are more important than tools – knowing ‘why does it makes sense doing it’ is more valuable than just knowing ‘how to do it’.”

Agenda:

19:00 Welcome & Introduction
19:10 Data Vault: A Question of Early or Late Integration (Michael Müller)
20:00 Discussion & Conclusion