Kimball DW
Terms from Kimball's Dimensional Data Warehousing concepts
Terms
undefined, object
copy deck
- Aggregated Facts
- A table of precalculated summary facts to improve performance
- Business Process
- Operational activity the organization engages in to accomplish its primary goals
- Bus Matrix
- Maps business processes to common entities/objects
- Analytic Theme
- A major challenge or opportunity faced by the organization
- Degenerate Dimesion
- A dimension without any attributes of its own
- Snowflaking
- Splitting out a dimension into further tables, closer to normalization
- Slowly Changing Dimensions
- Approach to handling changes in dimension attributes over time
- Drill Across
- Using a dimension to look across more than one business process.
- Aggregate Dimension
- A choice to split out a higher level dimension, to support when some of the fact data is at one level of detail, while other is at higher level of detail (grain).
- Dimension
- Give CONTEXT to facts. NOUNs of the business.
- Hierarchy
- Natural way to group data within a dimension to support DRILL DOWN.
- Type 3 SCD
- A slowly changing dimension to be tracked over time, but with copies of columns for each history version. NOT recommended approach
- Type 2 SCD
- A slowly changing dimension that the business desires to see the effects of change over time, through start and end dates.
- Fact
- Measurement associated with a specific business process. VERBs of the business.
- Variable Depth Hierarchy
- A dimension that may have varying degrees of hierarchies, i.e. not a fixed number of hierarchies across the dimension
- Type 1 SCD
- A slowly changing dimension that the business has decided, does NOT need to have history tracked over time.
- Bridge Table
- A compromise table between dimensions to support a many to many relationship
- Surrogate Key
- A key created just for the data warehouse, that is not the natural key from the source system
- Business Process Prioritization
- Uses grid to rank processes based on business value and implementation feasibility
- Grain
- Level of detail of the data
- 3 Fact Table Types
- Transaction, Periodic Snapshot, Accumulating Snapshot