Database Reference
In-Depth Information
Chapter 1. Introduction
to
Multidimensional Data Model Design
In this chapter we will discuss the differences between Multidimensional databases
(cubes) and Tabular Models in order to help you decide which is best for meeting your
particular needs. More details on how to implement each of these will be covered in
the rest of this cookbook. In this chapter we will explore the following concepts:
• The business value of Business Intelligence
• The challenges and barriers faced when implementing Business Intelligence
• Strategies for overcoming these challenges and barriers
• Choosing multidimensional or Tabular Models
• Star- and Snowflake-relational schema
• A sample scenario for choosing the Snowflake schema
Introduction
Business Intelligence
(
BI
) used to be a competitive advantage for businesses that
could afford it. Today, BI is increasingly becoming a fundamental and critical function
of every business, which means it can no longer operate on an exclusively strategic
basis. Also, it must be responsive to changing business needs in a time frame that
allows the business to address those needs. The demand for more responsiveness
(less time to implement) of BI continues to increase in parallel with the demand for
more functionality.
In response to this increasing pressure on BI to perform at "Business speed", a new
capability has emerged in the form of Self-service BI. These Self-service tools allow
business users to acquire and analyze data from a variety of sources according to
their specific needs at that moment. There are challenges and concerns that come
with giving this capability to business users. We'll discuss these in detail in this
chapter, but Self-Service BI is a good and growing solution to an important business
need.
Microsoft's offerings for Self-Service BI include tools such as Power Pivot, Power
View, and Tabular Models in Analysis Services. These tools continue to gain accept-
ance and are an increasingly presumed capability in Microsoft BI environments. It's
