Dw design 4_bus_architecture
•
0 likes•386 views
The document discusses the data warehouse bus architecture approach for building an enterprise data warehouse incrementally. It involves defining a standard bus interface that separate data marts can implement independently while still plugging together cohesively. During the architecture phase, a master set of standardized dimensions and facts is designed to provide uniform interpretation across the enterprise and establish the data architecture framework. The rows of the bus matrix correspond to individual data marts, which should be kept separate if their sources, processes, or implementation sizes differ significantly. Common dimensions must be consistent in meaning and attributes across different fact tables.
Dw design 4_bus_architecture
- 1. DATA WAREHOUSING Multi Dimensional Data Modeling. DW Bus architecture and matrix
- 2. 2
- 3.  Obviously, building the enterprise’s data warehouse in one step is too daunting, yet building it as isolated pieces defeats the overriding goal of consistency.  For long-term data warehouse success, we need to use an architected, incremental approach to build the enterprise’s warehouse.  The approach we advocate is the data warehouse bus architecture. 3
- 4. ï‚¡ By defining a standard bus interface for the data warehouse environment, separate data marts can be implemented by different groups at different times. ï‚¡ The separate data marts can be plugged together and usefully coexist if they adhere to the standard 4
- 5. ï‚¡ During the limited duration architecture phase, the team designs a master suite of standardized dimensions and facts that have uniform interpretation across the enterprise. This establishes the data architecture framework 5
- 6. ï‚¡ The rows of the bus matrix correspond to data marts. ï‚¡ You should create separate matrix rows if the sources are different, the processes are different, or if the matrix row represents more than what can reasonably be tackled in a single implementation iteration. 6
- 7.  Common dimensions for different processes should be the same.  A dimension that has exactly the same meaning and content when being referred from different fact tables.  Where attributes apply, they should mean the same thing.  Roll-up dimensions conform to the base-level atomic dimension if they are a strict subset of that atomic dimension. (granularity) 7
- 8. 8
- 9. ï‚¡ Revenue, profit, standard prices, standard costs, measures of quality, measures of customer satisfaction, and other key performance indicators (KPIs) are facts that must be conformed. 9
- 10. ï‚¡ The Data Warehouse Toolkit.Second Edition.The Complete Guide to Dimensional Modeling.Ralph Kimball.Margy Ross 10