If the customer already has Visual Studio 2005 installed, the BI project
templates are added to their existing Visual Studio installation. A link will appear on
the start menu to BIDS, but there is really only one copy of Visual Studio installed
on the computer.
Before venturing into the actual tools and techniques of building a cube, it??™s
important to point out that technically, a cube can be built from a full-normalized,
OLTP database. Nothing in the tools prevent cubes from being built from normalized
schemas, yet this chapter just spent time discussing how to create the dimension and
fact tables for a star schema. There??™s actually a very good reason for this.
While a cube can be built from a normalized schema, there are many reasons why
creating a star schema is a best practice. These reasons include, but are not limited to
the following:
Data for a warehouse is often pulled from multiple sources, so performing an
ETL process into a single location allows for the transformations that make
data consistent.
Relational data warehouses can be queried using relational reporting tools and
performance is often faster than querying a normalized schema.
Building cubes off a product OLTP structure can dramatically slow the
performance of the OLTP system.
OLTP systems are real-time and cubes are usually a snapshot in time, meaning
that the cube might not match the underlying data soon after it is processed.
Star schemas are designed using the same concepts as cubes (measures and
dimensions), so their structure easily translates to cubes, as well as being more
readable by humans.
Pages:
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94