I'm trying to find out how a data warehouses architecture i.e., central data warehouse with distributed OLTP, affects the type of methodology used to create the data warehouse e.g., top-down, bottom-up, hybrid.


Larissa Moss' Answer: OLTP is not part of a data warehouse architecture and has nothing to do with the type of methodology to use or even the development approach to take (top down, bottom up, hybrid). OLTP files and databases (centralized or distributed) are used as data sources by the ETL (extract/transform/load) process to populate any of the following target databases in the data warehouse environment (a.k.a. Corporate Information Factory): ODS (operational data store), EDW (enterprise data warehouse) and DM (data mart). As for the type of methodology, it must be data driven, iterative, agile and cross-organizational. The best (and cheapest) resource for a data warehouse methodology (with a complete DW/BI work breakdown structure on the CD in the back of the book) is available through www.amazon.com: Moss, Larissa T. and Shaku Atre. Business Intelligence Roadmap: The Complete Project Lifecycle for Decision-Support Applications. Boston, MA: Addison-Wesley, 2003.

As for the type of development approach, very few companies apply the top-down approach, which starts by creating a high-level but relatively complete entity-relationship diagram (conceptual business data model) up front. Most companies apply the bottom-up or a hybrid approach. The bottom up approach places emphasis on quickly building multi-dimensional point solutions for known business requirements. The danger with the bottom-up approach is that it can quickly deteriorate into stovepipe data mart development, because it pays little attention to cross-organizational integration activities. The hybrid approach usually combines some cross-organizational architectural activities from the top-down approach with the requirements-driven, project-specific activities from the bottom up approach. The methodology mentioned above (Business Intelligence Roadmap: The Complete Project Lifecycle for Decision-Support Applications) supports the hybrid approach.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access