Implementing a Microsoft SQL Server Parallel Data Warehouse Using the Kimball Approach
SQL Server Technical Paper
Writer: Warren Thornthwaite
Technical Reviewers: Jesse Fountain, Barbara Kess, Stuart Ozer
Published: June 2011
Applies to: SQL Server 2008 R2
This white paper explores how the Kimball approach to architecting and building a data warehouse/business intelligence (DW/BI) system works with Microsoft’s Parallel Data Warehouse, and how you would incorporate this new product as the cornerstone of your DW/BI system. For readers who are not familiar with the Kimball approach, we begin with a brief overview of the approach and its key principles. We then explore the Parallel Data Warehouse (PDW) system architecture and discuss its alignment with the Kimball approach. In the last section, we identify key best practices and pitfalls to avoid when building or migrating a large data warehouse to a Microsoft SQL Server PDW system. This includes a look at how Parallel Data Warehouse can work with other Microsoft offerings such as SQL Server 2008 R2, the Microsoft Fast Track Reference Architecture for Data Warehouse, and the new Business Data Warehouse to provide a complete solution for enterprise-class data warehousing.