Creating Analysis Services Cubes - SQL Server Analysis Services 2012 Cube Development - page 66

Chapter 3. Creating Analysis Services
Cubes
In this chapter, we will cover the following recipes:
• Defining measure groups and measures
• Setting measure properties
• Browsing the cube data
• Dimension usage with measure group
• Examining cube file structures
• Partitioning strategies
• Defining partition slice
• Merging partitions
• Defining aggregation designs
• Distinct count measure groups
• Enabling write-back feature
• Deployment options
Introduction
This chapter contains recipes for building SQL Server Analysis Services ( SSAS )
cubes. Much like we did in Chapter 2 , Defining Analysis Services Dimensions , let's
start with building straightforward cubes and then work our way up to more complex
requirements.
Before you delve into building cubes, you need to become familiar with terms describ-
ing the architecture of an Analysis Services cube. Each Analysis Services database
may contain one or more cubes. You can build each cube based on one or multiple
fact tables found in the relational data warehouse. Each measure group represents a
collection of numeric metrics (or measures ) found in a single fact table. However, de-
pending on the requirements, you may have to build several measure groups based
on a single fact table. Each measure group within a cube may use multiple dimen-
sions available within the database. Not all database dimensions need to be exposed
with each measure group. Each measure group can be split up into multiple partitions
to speed up processing as well as querying operations.