What is the purpose of designing a logical data model for a data mart or warehouse?


Larissa Moss’s Answer: The purpose of a logical data model (LDM) is not database design. The purpose of an LDM is to facilitate the integration of data from the disparate and redundant source systems and to enable the data analyst to normalize data elements found in those source files. An LDM is completely process independent and only captures the business objects and business actions as they exist in the real world for the enterprise. It is the only place where you can have "one fact in one place". The word "fact" here does not refer to its most recent usage in multidimensional databases, but to a business attribute which only has one – and only one – true "home" in a "business entity." True business integration (one fact in one place) can only be accomplished with an LDM, not in physical databases because (for performance reasons) we still store data redundantly in our target databases. Dirty data can only be found during the LDM process and not during database design. The reason is that the research to be done on the content (domain) of every source data element and its resolution (finding the one and only one correct "home" for it – where it "lives" in the real business world) requires domain analysis and normalization skills as opposed to the analysis required for designing tables and columns, which is based on access path analysis and denormalization. For more information, I recommend Chapter 10, Data Models in the book "Data Warehouse Project Management" by Sid Adelman and myself, published by Addison- Wesley, ISBN 0-201-61635-1, as well as the book "The Data Warehouse Challenge: Taming Data Chaos by Michael Brackett," published by John Wiley & Sons, ISBN 0- 471-12744-2

Clay Rehm’s Answer: My first thought is how you could design and implement a successful data warehouse without one?! There are many benefits of a properly designed logical data model. Most importantly, the logical data model is the communication device between the business staff providing the requirements and the data warehouse development team providing the solution. The logical data model allows the data warehouse team to capture the business requirements which is easy for the users driving the requirements to understand and easy for the DBA to convert the logical data model into a physical database. The logical data model can be built in a tool like Microsoft PowerPoint or even on a whiteboard. Whatever tool you use is not the issue. The point is that the business requirements are captured in a way that is easily communicated back to the users.

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