Software // Information Management
News
3/26/2007
12:04 PM
Connect Directly
RSS
E-Mail
50%
50%

Kimball University: Think Critically When Applying Best Practices

Best practices are precision tools that should be wielded precisely and skillfully. This article describes five best practices drawn from the Kimball Method that often are described incorrectly.

Dimensional modeling is a design discipline focused on optimizing the business intelligence platform for business users' ease of use and query performance. To achieve the goals of being simple and fast we describe a set of very specific design recommendations:

  • Conformed master dimensions form the bedrock of the enterprise BI/DW system and by themselves address the central issues of integration
  • Fact tables are derived directly from measurement processes found in familiar transaction-based applications. A fact table should never be departmentally or functionally bound but rather depends only on the "physics" of the original measurement process.
  • Fact tables should always be populated at the most atomic level possible for maximum flexibility. Atomic data lets business users ask constantly changing, far-ranging and very precise questions. It also assures the extensibility of additional attributes, metrics or dimensions without disrupting existing reports and queries.
  • Exposing snowflaked or normalized dimension tables directly to end users is strongly discouraged. We have shown repeatedly that properly designed denormalized (flat) dimension tables contain precisely the same information content as normalized schemas. The only difference is complexity, as experienced by the end users. We embrace (and teach) normalized designs in the extract/transformation/load (ETL) phases; however, we avoid normalization in the user-accessible presentation area.

Our Differences of Opinion and Facts and Fables articles describe these concepts in greater detail.

4. Use Conformed Dimensions for Integration

Data integration and consistency are key goals of any enterprise business intelligence effort. Data integration requires organizational consensus to establish and administer common labels and measures enterprisewide. In the Kimball Method, these labels and measures reside in conformed dimensions and conformed facts, respectively. Conformed dimensions are typically built and maintained as centralized, persistent master data during ETL, then reused across dimensional models to enable data integration and ensure consistency.

We enthusiastically support the recent Master Data Management (MDM) and Customer Data Integration (CDI) trends, as they are very consistent with the "conformed approach." For more insight, read The Matrix: Revisited and Integration for Real People articles at Intelligence Enterprise.

Previous
2 of 3
Next
Comment  | 
Print  | 
More Insights
The Agile Archive
The Agile Archive
When it comes to managing data, donít look at backup and archiving systems as burdens and cost centers. A well-designed archive can enhance data protection and restores, ease search and e-discovery efforts, and save money by intelligently moving data from expensive primary storage systems.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - August 27, 2014
Who wins in cloud price wars? Short answer: not IT. Enterprises don't want bare-bones IaaS. Providers must focus on support, not undercutting rivals.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Howard Marks talks about steps to take in choosing the right cloud storage solutions for your IT problems
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.