PASS Community Summit 2008: Program Sessions (Business Intelligence)

Dates are arguably the single most important dimension in an SSAS cube. It is hard to imagine a cube without one. Date is often the most useful dimension to tie measure groups together, allowing integration of multiple sources. Dates are often one of the most problematic dimensions in an SSAS cube. Dates for different fact tables are often at different grain (e.g., orders by day, plan by month). Date dimensions often have multiple hierarchies (e.g., Calendar Month, Fiscal Week), and often need attributes split across multiple axes (e.g. Years on rows, Months on Columns). Dates dimensions often appear multiple times in different roles (e.g., Order Date, Ship Date), which multiplies the impact of less-optimal design. As if that weren't enough, dates are very often used in calculations (e.g., growth from previous month, YTD), and the formulas are highly affected by attribute relationships, which can be particularly complex with dates. This session will teach you effective design practices for dates, and demonstrate how good design can make it much easier to create date-related formulas that work as expected.

Register Now for PASS Community Summit 2008!