Database Reference
In-Depth Information
Class
Description
TotalDollars
TotalUnits
AP
Dishwasher
$1,190.00
2
AP
Dryer
$1,319.80
4
AP
Gas Range
$990.00
2
AP
Microwave Oven
$600.00
4
AP
Washer
$399.99
1
HW
Iron
$241.45
11
SG
Home Gym
$1,589.90
2
SG
Treadmill
$2,580.00
2
FIGURE 9-20
Query results for total sales by class and part
300
Roll up. When you view specific aggregate data, you roll up the data to view and analyze higher
levels of aggregation. Rolling up the data is the exact opposite of drilling down the data. For
example, the sales manager might start with the query results for total sales by class and part
(see Figure 9-20), click the appropriate button to roll up the data for the query results for the
total sales by class (see Figure 9-19), and then click another button to roll up the data for the
query results for total sales (see Figure 9-15).
Data mining consists of uncovering new knowledge, patterns, trends, and rules from the data stored in a
data warehouse. You use data mining software to answer questions such as the following:
Which products best attract new customers?
What factors best predict which customers default in making payments?
What are the optimal seasonal inventory levels based on predicted economic factors?
What is the optimal number of customers to assign to each sales rep?
t sift through the data in
them to find answers to those questions. Instead, with minimal user interaction, data mining software
attempts to answer the questions by using sophisticated analytical, mathematical, and statistical techniques.
Because data warehouses often contain enormous amounts of data, users can
'
Rules for OLAP Systems
E. F. Codd (Codd, E. F., S. B. Codd, and C. T. Salley.
“
Providing OLAP (On-line Analytical Processing) to User-
Analysts: An IT Mandate.
Arbor Software, August, 1993) formulated 12 rules that OLAP systems should follow.
The 12 rules serve as a benchmark against which you can measure OLAP systems. The 12 rules are as follows:
”
1. Multidimensional conceptual view. Users must be able to view data in a multidimensional way,
matching the way data appears naturally in an organization. For example, users can view data
about the relationships between data using the dimensions of parts, customer locations, sales
reps, and time.
2. Transparency. Users should not have to know they are using a multidimensional database nor
need to use special software tools to access data. For example, if users usually access data using
a spreadsheet, they should still be able to use a spreadsheet to access a multidimensional
database.
3. Accessibility. Users should perceive data as a single user view even though the data may be
physically located in several heterogeneous locations and in different forms, such as relational
databases and standard files.
4. Consistent reporting performance. Retrieval performance should not degrade as the number of
dimensions and the size of the warehouse grow.
5. Client/server architecture. The server component of OLAP software must be intelligent enough
that a variety of clients can be connected with minimal effort.
6. Generic dimensionality. Every dimension table must be equivalent in both its structural and
operational capabilities. For example, you should be able to obtain information about parts as
easily as you obtain information about sales reps.
