Metadata-driven Data Warehouse Automation with Lars Rönnbäck (English)

Data modeling guru Lars Rönnbäck will join us again to share his experiences with building data warehouse automation frameworks.

Abstract:

Whenever a task is to some degree repeatable, automation is beneficial, such that given the right tools it is possible to work more efficiently and be less prone to error.
In this presentation we will look at the development of three data warehouse automation frameworks built over the last 20 years. All of these have been or are still used in production, even if the glory days are over for the first two. The lessons learnt from these endeavors will be viewed with both positive and negative conclusions drawn.
The session will finish with a demo of the current framework, and some experiences from six data warehouses running it, all maintained by a single person. There is still labor involved, but it is the right kind of labor – the one where effort yields appreciable results.
There will also be a word of warning, given the direction some alternative frameworks are headed in.

Speaker:

Lars Rönnbäck has been working within the areas of Data Warehousing and Information Modeling for more than 20 years and with some of the largest companies in Sweden.
He is also an associated researcher at Stockholm University in the field of divergent information, with several published papers on the topic. He was the first to show that the sixth normal form could be used efficiently in practice, in the form of Anchor modeling. Ideas from that work were taken into temporal dimensional modeling, in an attempt to solve many issues concerning slowly changing dimensions.
During his work, Lars has built several frameworks for automating tasks related to data management, and developed new algorithms that increase efficiency when working with complex data.

Agenda:

19:00 Welcome & Round of Introductions
19:15 Metadata-driven Data Warehouse Automation (Lars Rönnbäck)
20:00 Discussion & Conclusion