1 Chapter 7 Enterprise Databases, Data Warehouses, and Business Intelligence.
-
Upload
derrick-morrison -
Category
Documents
-
view
215 -
download
0
Transcript of 1 Chapter 7 Enterprise Databases, Data Warehouses, and Business Intelligence.
1
Chapter 7
Enterprise Databases, Data Warehouses, and Business
Intelligence
2
Objectives Advantages of shared databases. Compare relational vs. object oriented
databases. Describe the differences between schemas,
views, and indexes. Shared vs. distributed databases. Data warehouses and Business Intelligence.
3
Enterprise Data – Scaling Up Database: A collection of data and information
describing items of interest to an organization.
Enterprise Database: A collection of data designed to be shared by many users within an organization.
4
Both Actual Data and Schema are Shared
5
Database Mangement The Functions of Database Management:
Integrating Databases Reducing Redundancy Sharing Information Maintaining Integrity Enabling Database Evolution
6
DBMS in Systems
7
Enterprise Data Model Enterprise Data Model/Entity Relationship: A
graphical representation of the items (the entities) of interest about which data is captured and stored in the database.
8
Schema Schema: The structure of a database.
Schema for Relational Database Relational Database: A database in which
the data are structured in a table format consisting of rows and columns.
9
Relational Schema
10
Object Orientation Schema for Object-Oriented Database
Object-oriented Database: A database that stores data and information about objects.
Object: A component that contains data about itself and how it is to be processed.
Action/Method: An instruction that tells a database how to process an object to produce specific information.
11
Object Oriented Schema
12
User views
View: A subset of one or more databases, created either by extracting copies of records from a database or by merging copies of records from multiple databases.
13
Enterprise Database StructuresViews (Continued)
14
Indexing Index: A data file that contains identifying
information about each record and its location in storage.
Record Key: In a database, a designated field used to distinguish one record from another.
15
Enterprise Database StructuresIndexes (Continued)
16
Integration Web-based Integration: Makes data from
enterprise databases available to users connecting through the Internet (including enterprise intranets and extranets).
17
Databases and the Internet
18
Distributed Databases Shared Database: A database shared among
many users and applications.
Distributed Database: A database that resides in more than one system in a distributed network. Each component of the database can be retrieved from any node in the network.
19
Partitioning and Replication Partitioning: A method of database distribution in
which different portions of the database reside at different nodes in the network. Vertical Horizontal
Replication: A method of database distribution in which one database contains data that are included in another database. Real time Cascade Batch
20
Distribution Strategies Geographic Distribution Strategy: A database
distribution strategy in which the database is located in a region where the data and information are used most frequently.
Functional Distribution Strategy: A database distribution strategy in which the database is distributed according to business functions.
21
Designing a Distributed Database Database Directory: The component of a shared
database that keeps track of data and information.
Other Design Factors Storage Costs Processing Costs Communication Costs Retrieval and Processing Reliability Frequency of Updates and Queries
22
Data Warehouses and OLAP Data Warehouse: A large data store, designed
from inquiries, that combines details of both current and historical operations, usually drawn from a number of sources.
Online Analytical Processing (OLAP): Database processing that selectively extracts data from different points of view.
23
Comparison of Enterprise Databases and Data Warehouses
24
Data Warehouse
25
Data Warehouses and OLAPDefinition
Data Mining: Uses software designed to detect information hidden in the data.
Data Marts: Processed to focus on a specific area of activities or isolated scientific or commercial processes.
Business Intelligence: Supporting Managerial Decision Making
Issues MIS: Reporting Data-Driven DSS: Business Intelligence Model- -Driven DSS: Models and Modeling GDSS and ESS Case Study: MasterCard
28
Decision Levels and Application Systems
Business Operations
TacticalManagement
Strategic
Mgt.
DSS
MIS
Tran
sact
ion
Proc
essi
ng
From R.N. Anthony, Planning and Control Systems: A Framework for
Analysis. Harvard University (1965)
29
MIS vs. DSS (Data Driven and Model Driven)
MIS: Provides reports based on routine flow
of data. Assists in general control of the organization. Exception reports used to reduce volume and
focus on items that require management attention.
30
MIS Reports Paper or online Can includes text, graphs, or both. Batch vs. Real-time Fixed vs. Ad Hoc (a continuum) Summary vs. Detail Types include:
Exception Trend Validation (such as Trial Balance)
31
Data-Driven DSS(a.k.a. Business Intelligence) Also known as. Query/inquiry, Data Mining, and OLAP
(Online Analytical Processing).
Goal is to determine where we are or where we’ve been.
“Business Intelligence” has emerged as common term.
Sometimes also called Datamining, though this generally implies using statistical techniques such as correlation analysis and clustering to find patterns and relationships in large databases.
32
Goals of BI Enables users to identify and understand the
key trends and events driving their businesses. Allows employees to sift through and analyze
large amounts of data that the company makes available for them.
Helps business managers at all levels make better decisions quicker.
33
What is BI Used For? To perform trend analyses on product, sales,
event (i.e. promotions and advertising campaigns) and financial information. Sales per office or region and then drill down to lower
level details to uncover what is driving the trends. It is also used for exception-reporting and for
budgeting, planning, and forecasting.
34
BI Tool Capabilities Support large volumes of data and an unlimited
number of dimensions Can aggregate data
Sums, averages, maximums, minimums, percentage of total, and user-defined functions or rules.
Can contain analytical engines that perform computations. Rankings, ratios, or variances (i.e., This-year-to-last-
year or actual-versus-budget comparisons), Revenue or expense allocations, Currency conversions, etc.
Raises
0500
1000150020002500300035004000
Caulkins Jihong Louganis Naber Spitz Weissmuller
dolla
rs
0.0%
20.0%
40.0%
60.0%
80.0%
100.0%
Raise Raise pct Performance
Most BI Tools also include graphics capabilities
36
Data Sources for BI Include Relational Data Bases (including Data
Warehouses) Data Marts
Star Schemas Facts and Dimensions
Cubes (Facts and Dimensions)
37
Data Warehouse
OLTP Database3NF tables
Operationsdata
Predefinedreports
Data warehouseStar configuration
Daily datatransfer
Interactivedata analysis
Flat files
38
Data Warehouses Contain Data from Many Sources (a.k.a. Domains)
39
Cube Example: Sales Information
Sales information can be represented in the cube below. You will be able to derive many measures based on the dimensions below
Region
Dep
artm
ent
Time
40
Some Leading BI Vendors Enterprise Query/Reporting (RDBMS Based):
Actuate Crystal Reports Information Builders / WebFocus
OLAP (Data Mart and Cube Based): MicroStrategy Hyperion Oracle Business Objects (also includes reporting tools) Cognos (also includes reporting tools)
41
Demo Sites Cognos PowerPlay:
http://naade02.msfc.nasa.gov/workforce/index.html
http://www.cognosdemo.com/temple/
Information Builders Web FOCUS: www.informationbuilders.com/test_drive/inde
x.html
www.nyc.gov/html/doh/html/rii/index.html
42
For more information . . . Bill Inmon:
http://www.billinmon.com/
Ralph Kimball: http://www.rkimball.com/
Data Management Review: http://www.dmreview.com/
Data Warehouse: http://www.datawarehouse.com
DSS: Decision Support System Models
sales revenueprofit prior154 204.5 45.32 35.72163 217.8 53.24 37.23161 220.4 57.17 32.78173 268.3 61.93 47.68143 195.2 32.38 41.25181 294.7 83.19 67.52
Sales and Revenue 1994
Jan Feb Mar Apr May Jun0
50
100
150
200
250
300
LegendSalesRevenueProfitPrior
Database
Model
Output
data
to a
nalyz
e
results
Optimization
1 2 3 4 5 6 7 8 9 101
3
5
0
5
10
15
20
25
Ou
tpu
t
Input Levels
Maximum
Model: definedby the data pointsor equation
Control variables
Goal or outputvariables
Why Build Models? Understanding
the Process Optimization Prediction Simulation or
"What If"
Scenarios
Prediction
0
5
10
15
20
25
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2
Time/quarters
Ou
tpu
t
Moving AverageTrend/Forecast
Economic/regressionForecast
46
Marketing Sales ForecastGDP and Sales
1000
1200
1400
1600
1800
2000
2200
2400
2600
2800
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
Quarter
GD
P
30
40
50
60
70
80
90
100
Sal
es
GDP
Sales
Forecast
forecast
Note the fourth quarter sales jump.
The forecast should pick up this cycle.
47
Time Series Components
time
sales
Dec Dec Dec Dec1. Trend2. Seasonal3. Cycle4. Random
Trend
Seasonal
A cycle is similar to the seasonal pattern,but covers a time period longer than a year.
Collect data over timeIdentify trendsIdentify seasonal effectsForecast based on patterns
48
Forecasting Uses Marketing
Future sales Consumer
preferences/trends Sales strategies
Finance Interest rates Cash flows Financial market
conditions
HRM Labor costs Absenteeism Turnover
Strategy Rivals’ actions Technological change Market conditions
Simulation
0
5
10
15
20
25
1 2 3 4 5 6 7 8 9 10
Input Levels
Ou
tpu
t
Goal or outputvariables
Results from alteringinternal rules
50
Group Decision Support Systems (GDSS)
Interactive computer-based system. Facilitates solution to unstructured problems. Set of decision makers working together as a
group.
EIS: Enterprise Information System (aka Executive Information System and Executive Support System)
Easy access to data
Graphical interface Non-intrusive Drill-down
capabilities
EIS Software from Lightship highlights ease-of-use GUI for data look-up.
52
Digital Dashboard
http://www.microsoft.com/business/casestudies/dd/honeywell.asp
Stock market
Exceptions
Plant or management variables
Equipment details
Products
Quality control
Plant schedule