Transcript Meeting a Business Need

Meeting a Business Need
Chapter 2
Overview
Defining
DW Concepts
& Terminology
Planning
For a
Successful
Warehouse
Choosing a
Computing
Architecture
Meeting a
Business
Need
Planning
Warehouse
Storage
Modeling
The Data
Warehouse
ETT
(Building
The
Warehouse)
Analyzing
User Query
Needs
Supporting
End User
Access
Managing
The Data
Warehouse
Project Management
(Methodology, Maintaining Metadata)
Characteristics of OLTP
Systems
Characteristic
OLAP
Typical operation
Update
Level of analytical requirement
Low
Screens
Amount of data per transaction
Unchanging
Small
Data level
Detailed
Age of data
Current
Orientation
Records
Why OLTP Is Not Suitable
for Complex Analysis
OLAP
Complex Analysis
Information to support
day-to-day service
Historical information
to analyze
Data stored at transaction
level
Data needs to be integrated
Database design: Normalized
Database design:
Denormalized, star schema
Management Information Systems and
Decision Support
Ad hoc access
Production
platforms
Operational reports
Decision makers
MIS systems provided business data
Reports were developed on request
Reports provided little analysis
capability
Decision support tools gave personal ad
hoc access to data
Analyzing Data from
Operational Systems
Data structures are complex
Systems are designed for high performance
and throughput
Data is not meaningfully represented
Data is dispersed
OLTP systems may be unsuitable for intensive
queries
Production
platforms
Operational reports
Data Extract Processing
Operational systems
Extracts
Decision makers
End user computing offloaded from the
operational environment
User’s own data
Management Issuess
Operational systems
Extracts
Extract explosion
Decision makers
Productivity Issues
 Duplicated effort
 Multiple technologies
 Obsolete reports
 No metadata
Data Quality Issues
No common time basis
Different calculation algorithms
Different levels of extraction
Different levels of granularity
Different data field names
Different data field meanings
Missing information
No data correction rules
No drill-down capability
From Extract to
Warehouse DSS
Data warehouse
Internal and
external systems Controlled
Reliable
Quality information
Single source of
Decision makers
Advantages of Warehouse
Processing Environment
No duplication of effort
No need for tools to support many
technologies
No disparity in data, meaning, or
representation
No time period conflict
No algorithm confusion
No drill-down restrictions
Business Motivators
Know the business
Reinvent to face new challenges
Invest in products
Invest in customers
Retain customers
Invest in technology
Improve access to business information
Be profitable
Provide superior services and products
Business Motivators
Provides supporting information systems
Get quality information
- Reduce costs
- Streamline the business
- Improve margins
Technological Advances
Parallelism
- Hardware
- Operating
system
- Query
- Index
- Applications
Large database
64-bit architecture
Indexing techniques
Affordable, costeffective
Open systems
Robust warehouse tools
Sophisticated end user
tools
Growth Motivators and
Inhibitors
Successful implementations
Decreased risk
Robust extraction software
Improving price to performance ratios
Improved staff training
Year 2000 compliance
Skills shortage
Lack of integrated metadata
Data cleaning cost
Typical Uses of Data
Warehouse
Airline
Banking
Health Care
Investment
Insurance
Retail
Telecommunications
Manufacturing
Credit card suppliers
Clothing distributors
Summary
This lesson covered the following topics:
Describing why an online transaction
processing(OLTP) systems is not suitable for
complex analysis
Describing how extracting processing for
decision support querying led to data
warehouse solutions employed today
Explaining why businesses are driven to
employ data warehouse technology
Identifying some of the industries that
employ data warehouses