Tài liệu Module 1: Introduction to Data Warehousing and OLAP pptx

The module introduces data warehouses and OLAP systems and describes the differences between relational data marts and OLAP cubes.. Explain the differences between relational data marts

Contents

Overview 1

Introducing Data Warehousing 2

Understanding Data Warehouse Design 18

Review 40

Module 1: Introduction

to Data Warehousing and OLAP

to represent any real individual, company, product, or event, unless otherwise noted Complying with all applicable copyright laws is the responsibility of the user No part of this document may

be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without the express written permission of Microsoft Corporation If, however, your only means of access is electronic, permission to print one copy is hereby granted

Microsoft may have patents, patent applications, trademarks, copyrights, or other intellectual property rights covering subject matter in this document Except as expressly provided in any written license agreement from Microsoft, the furnishing of this document does not give you any license to these patents, trademarks, copyrights, or other intellectual property

 2000 Microsoft Corporation All rights reserved

Microsoft, BackOffice, MS-DOS, Windows, Windows NT, <plus other appropriate product

names or titles Replace this example list with list of trademarks provided by copy editor

either registered trademarks or trademarks of Microsoft Corporation in the U.S.A and/or other countries

<This is where mention of specific, contractually obligated to, third party trademarks, which are added by the Copy Editor>

The names of companies, products, people, characters, and/or data mentioned herein are fictitious and are in no way intended to represent any real individual, company, product, or event, unless otherwise noted

Other product and company names mentioned herein may be the trademarks of their respective owners

Instructor Notes

This module introduces students to data warehousing and online analytical processing (OLAP)—their uses, essential concepts, terminology, and architecture

The module describes the value of deriving business information from raw operational data, and the process of using defined types of business analysis to drive decision support systems The module introduces data warehouses and OLAP systems and describes the differences between relational data marts and OLAP cubes

Finally, the module introduces OLAP technology Students will learn the fundamentals of dimensions, members, and cubes The materials also explore methods for visualizing multidimensional databases

After completing this module, students will be able to:

! Describe characteristics, goals, and applications of a data warehouse

! Understand the need of and use for OLAP solutions

! Describe data warehouse design

! Understand the reasons for implementing OLAP models and describe their components

! Visualize a multidimensional database

Materials and Preparation

This section lists the required materials and preparation tasks that you need to teach this module

Required Materials

To teach this module, you need the following materials:

! Microsoft® PowerPoint® file 2074A_01.ppt

! Microsoft Excelfile DEMO_01.xls

! Local cube file DEMO_01.cub

Preparation Tasks

To prepare for this module, you should:

! Read all the student materials

! Read the instructor notes and margin notes

! Practice the lecture presentation and demonstration

! Review the Trainer Preparation presentation for this module on the Trainer Materials compact disc

! Review any relevant white papers that are on the Trainer Materials compact disc

Presentation:

60 Minutes

Lab:

00 Minutes

Other Activities

Difficult Questions

Below are difficult questions that students may ask you during the delivery of this module and answers to the questions These materials delve into subjects that are within the scope of the module but are not specifically addressed in the content of the student notes

1 Is a data mart synonymous with a star schema?

Not necessarily The data mart is a subset of a data warehouse with data specific to a particular subject or business activity It can be relational or multidimensional

A relational data mart may have one or many star schemas that belong

to the data mart and contain data particular to a subject

Multidimensional data marts use star schemas behind the scenes to support multidimensional data structures called cubes

2 Are data marts only composed of summary data?

No Data marts can contain detailed data in addition to summarized data Using summarized data marts is a way to enhance query performance

3 Do you need to purchase Microsoft SQL Server™ 2000 in order to use Microsoft SQL Server 2000 Analysis Services?

Yes Analysis Services is bundled with SQL Server However, you can install Analysis Services without using—or installing—SQL Server

4 What are reasons to use OLAP technology instead of relational database technology?

OLAP technology provides fast, intuitive access to numeric data It gives users the ability to browse the database themselves, without needing intermediate parties to develop queries OLAP technology provides a central calculation engine to model complex business models and processes

5 Is Measures a dimension?

When administering a cube, Measures are treated differently from dimensions When browsing a cube and when using MDX, Measures is simply a dimension with only one level—and no All level

6 Is a cell that is empty—that is, it has no value—still a cell?

Yes The intersection of a member from each dimension forms a cell, whether that cell is populated or not The cell does not take any physical storage space, but a cube is a logical construct and does not reflect the physical storage

Displaying the Animated PowerPoint Slides

All the animated build slides are identified with an icon of links on the lower left corner of the slide

! To display the Data Warehouse System Components slide

This slide shows the components of a data warehouse system In the slide, data flows from sources systems to users Integrate this information with material from the student notes

1 Advance to the first animation that displays, at the bottom of the slide, the user data access, the data sources, and a data access line

Explain that the purpose of a data warehouse is to expose business information to users The data that users are interested in is that which resides in source systems

2 Advance to the second animation to display a data access line that connects the user data access to the data sources

Explain that although users require the data in the source system, directly accessing a source system can lead to several problems Because source systems are optimized for the inserts and updates associated with essential business processes, user queries often burden these systems and interfere with these essential processes In addition, because these systems are constantly changing, you will find that user data retrieval can produce differing results and lead to inconsistent reports

Given the limitations of source system reporting, explain that the best way

to meet the business analysis needs of an organization is by using a data warehouse Note that the transfer of data from the source system to users becomes the primary function of the data warehouse

3 Advance to the third animation to dissolve the data access line between the users and data sources and to display the staging area

Describe the characteristics of a staging area and note how data is extracted from source systems for staging

4 Advance to the fourth animation to display the data marts

Describe a data mart Mention that data marts can reside in relational databases or in OLAP cubes

5 Advance to the fifth animation to display the data warehouse

Explain that the data warehouse is a virtual union of the subject-specific data marts and cubes

6 Advance to the sixth animation to display the user data access lines to the data warehouse

Reiterate that the business analysis needs of an organization define the need for a data warehouse Given this need, the transfer of data from the source system to users becomes the primary function of the data warehouse

Module Strategy

Use the following strategy to present this module:

! Introducing Data Warehousing Present the differences between raw data and information Describe the characteristics of online transaction processing (OLTP) source systems and give some examples of OLTP systems Present the characteristics of a data warehouse and describe the components of a data warehouse system

! Defining OLAP Solutions Begin by introducing the basic characteristics of OLAP databases Give examples of common OLAP applications Explain the differences between relational data marts and OLAP cubes in terms of data storage, data content, data sources, and data retrieval Finally, introduce OLAP in

SQL Server 2000 and discuss its two main OLAP components—the SQL Server database and Analysis Services

! Understanding Data Warehouse Design Introduce the concept of a star schema and describe its characteristics Next, present the components of a fact table—foreign keys and measures—and explain the concept of the fact table grain Describe the characteristics of dimension tables and give examples from a data warehouse Finally, define

a snowflake schema as a variation of a star schema in which hierarchies are stored in dimension tables

! Understanding OLAP Models Define the key components of the OLAP database—measures, dimensions, and cubes Compare OLAP dimensions and relational dimensions Next, define the components of a dimension—levels and members—giving examples of each Discuss the family terms that describe the relationships between levels and members in a dimension Describe the characteristics of measures Finally, to summarize the requirements for building OLAP cubes

by using relational data sources, discuss how the relational source relates to the OLAP cube

! Applying OLAP Cubes Define a cube as the logical storage structure for an OLAP database Explain that each cell of a cube holds one value Describe how users isolate data with a cube Introduce the concepts of slicing and dicing data in a cube, and drilling up and drilling down through the levels in a hierarchy Discuss the visualization of multidimensional data, using spreadsheets to illustrate the concept Finally, connect to an OLAP cube by using a Microsoft Excel PivotChart® to demonstrate the power of OLAP

Overview

! Introducing Data Warehousing

! Defining OLAP Solutions

! Understanding Data Warehouse Design

! Understanding OLAP Models

! Applying OLAP Cubes

This module introduces you to data warehousing and online analytical processing (OLAP)—their uses, essential concepts, terminology, and architecture

You will learn about the value of deriving business information from raw operational data, and the process of using defined types of business analysis to drive decision support systems

You are introduced to data warehouses and OLAP systems and will learn the differences between relational data marts and OLAP cubes

Finally, you are introduced to OLAP technology You will learn the fundamentals of dimensions, members, and cubes The materials also explore methods for visualizing multidimensional databases

After completing this module, you will be able to:

! Describe characteristics, goals, and applications of a data warehouse

! Understand the need of and use for OLAP solutions

! Describe data warehouse design

! Understand the reasons for implementing OLAP models and describe their components

! Visualize a multidimensional database

In this module, you will learn

about data warehousing,

OLAP systems, and OLAP

cube fundamentals

# Introducing Data Warehousing

! Raw Data vs Business Information

! OLTP Source Systems

! Data Warehouse Characteristics

! Data Warehouse System Components

This section defines the differences between raw data and derived information, describes online transaction processing (OLTP) systems, and introduces data warehouse systems An understanding of data warehouse system components is important when you begin to design and implement decision support systems The following topics are discussed:

! Raw data versus business information

! OLTP source systems

! Data warehouse characteristics

! Data warehouse system components

Topic Objective

Introduce the concept of

data warehousing

Lead-in

This section defines the

differences between raw

data and derived

information, describes OLTP

systems, and introduces

data warehouse systems

Raw Data vs Business Information

! Capturing Raw Data

! Deriving Business Information

! Turning Data into Information

Turning raw data into valuable information is a core analysis process that drives the operations and business decisions of a company

Capturing Raw Data

A company typically captures large amounts of data daily This data often consists of raw facts that reflect the current state of the business

Examples of raw data include:

! An international retail music store chain captures sales data for every product purchase, return, and exchange around the world A raw fact may describe the Chicago branch of this music store selling $10,000 worth of merchandise in June of 2000

! A financial institution captures data for each customer’s checking and savings account A raw data fact may describe Stefan Knorr withdrawing

$50 from his checking account this morning in Amsterdam

On the surface, this data provides an indication of what happens in the business However, the captured data can perform many more functions The captured data can help a company understand how it currently operates and help a company plan its operations in the future

Deriving Business Information

The process by which you can derive business information from raw data involves:

! Examining the raw data in several different contexts and from several different points of view

Topic Objective

To describe the differences

and relationships between

raw data and business

information

Lead-in

Turning raw data into

valuable information is a

core analysis process that

drives the operations and

business decisions of a

company

Delivery Tip

Ask students about the

types of systems that they

work with that capture raw

data, derive business

information, and turn data

into information

By using this process, consider how the raw data from the previous examples is converted to valuable business information

The Chicago Music Store Raw Data: The Chicago branch of this music store sold $10,000 worth of

merchandise in June 2000 However, the Chicago branch sold $15,000 in June

1999 The Chicago branch sales goal for June 2000 is $20,000

Derived Information: It appears as if the Chicago branch did not meet its sales

goal for June 2000 and did not perform as well as the previous year Business analysis is now required to determine the cause of the decline in sales

Typical business questions arising from this analysis include:

! What products are selling in the Chicago store?

! What products are not selling?

! What is the effect of product promotions?

The Financial Institution Raw Data: Stefan Knorr withdrew $50 from his checking account this morning

in Amsterdam Stefan’s primary residence is located in Los Angeles, California

In the past month, Stefan has withdrawn money from London, England; Oslo, Norway; and Stockholm, Sweden

Derived Information: Stefan apparently travels extensively throughout Europe

Perhaps he would be interested in a special ATM card that allows unlimited access to his checking account in 16 different countries for an additional yearly fee However, additional analysis is required to verify that he meets other requirements for the new ATM card

Typical business questions arising from this analysis include:

! What is the average daily balance of his account?

! How many times has this customer been overdrawn in the last 2 weeks? In the last 2 months? In the last 2 years?

! For what other promotions does he qualify?

Turning Data into Information

After the value of meaningful business analysis is recognized in an organization, data and information requests become numerous and frequent Satisfying these requests can be a complex task as you navigate through the large amounts of captured source data and attempt to consolidate, analyze, and distribute information to other members of the organization

To meet these requests, a company typically implements a decision support system dedicated to providing data and information that can be used to perform meaningful business analysis

A company’s investment in these decision support systems is usually very large

OLTP Source Systems

! OLTP System Characteristics

$ Processes real-time transactions of a business

$ Contains data structures optimized for entries and edits

$ Provides limited decision support capabilities

OLTP System Characteristics

OLTP operational systems:

! Process real-time transactions of a business

OLTP systems conduct essential business processes by tracking real-time transactions OLTP systems continually change to represent the current state

of the business As the OLTP system processes new transactions, data is updated or inserted into the OLTP system immediately

! Contain data structures optimized for entries and edits

Because the performance of these systems is critical to keeping track of essential business processes, data structures are optimized for data entry and edits

! Provide limited decision support capabilities

Decision support goals are not a priority of OLTP systems Reporting from operational systems may supply the most current data However, directly accessing a source system can have a negative impact on source system performance and produce inconsistent reports due to the volatility of the OLTP system

Topic Objective

To define an OLTP source

system

Lead-in

Here are the characteristics

of a database designed for

an OLTP environment

Key Point

Point out that OLTP

systems are optimized for

inserts and updates, not

user queries

OLTP System Examples

OLTP operational system examples include:

! Order-tracking applications, such as catalog sales

! Customer-service applications, such as setting up customer accounts

! Point-of-sales applications, such as paying for items at a grocery store

! Service-based sales applications, such as cellular telephone billing

! Banking functions, such as deposits and withdrawals

Ask students to list

operational system

examples in their own

organizations

Data Warehouse Characteristics

! Provides Data for Business Analysis Processes

! Integrates Data from Heterogeneous Source Systems

! Combines Validated Source Data

! Organizes Data into Non-Volatile, Subject-Specific Groups

! Stores Data in Structures that Are Optimized for Extraction and Querying

A data warehouse system has components that move data from a source system

to users who want to perform data analysis The primary function of a data warehouse system is to support an organization’s business analysis processes

A data warehouse:

! Provides data for business analysis processes

A data warehouse is a data store that supports an organization’s business analysis processes Often, it is implemented as an enterprise-wide decision support system, installed to provide a reporting environment that facilitates data analysis by providing extensive decision support capabilities

! Integrates data from heterogeneous source systems

Operational systems and, sometimes, external systems are the sources for data warehouses These heterogeneous source systems can contain transformed and integrated source data from OLTP systems, previous-version systems, text files, and spreadsheets

! Combines validated source data

A data warehouse combines heterogeneous source data that has been authenticated according to previously defined business rules It is important that the integrity of data in a data warehouse meet the standards of the business rules and processes

The primary function of a

data warehouse system is to

support an organization’s

business analysis

processes

! Organizes data into non-volatile, subject-specific groups

A data warehouse stores data as non-volatile, subject-oriented data sets A data warehouse is a static environment Data is updated and inserted into the data warehouse periodically The frequency of data updates and inserts depends on business analysis requirements

! Stores data in physical structures that are optimized for data distribution and querying

A data warehouse facilitates data retrieval and analysis, and therefore query performance is important Thus, the design of a data warehouse is important for optimal data distribution and querying

Data Warehouse System Components

Data Warehouse

Data Access

User Data Access

Data Sources

Data Input

Staging Area

Data Marts

The data warehouse system contains several components that transfer data from

a source system to users who want to perform data analysis It is important to understand the role of a data warehouse system and where it persists in the data flow of an organization

User Data Access

The purpose of a data warehouse in an organization is to expose business information to users Users analyze data to derive business information and thereby make decisions The data that users are interested in is the data from operational source systems

Even though users require the data in these source systems, directly accessing a source system can lead to several problems Because source systems are optimized for the inserts and updates associated with essential business operations, user data access queries often burden and interfere with essential business processes In addition, because these systems are constantly changing, you will find that user data retrieval can produce differing results and lead to inconsistent reports

Given the limitations of source system reporting, the best way to meet the business analysis needs of an organization is to use a data warehouse The transfer of data from the source system to users becomes the primary function

of the data warehouse system

The transfer of data from source system to user is the critical path of

a data warehouse system

A data warehouse system

contains many components

that move data from its

source system to users who

perform data analysis

Delivery Tips

Use this slide to introduce

OLAP solutions and data

marts and to transition into

the next section that

describes OLAP solutions

Use the slide to explain

each of the data warehouse

system components and the

relationships of the

components

Before explaining the above

slide, review Displaying the

Animated PowerPoint Slides

in the Other Activities

section of the Instructor

Notes

Important

Data Sources

Source systems are known as OLTP systems or legacy systems in a mainframe environment Source systems are the operational systems that capture the transactions of a business and supply data to the data warehouse or data mart

A source system can be relational or non-relational Source systems do not generally contain large amounts of historical information, as they are continually updated to reflect the current state of the business

Staging Area

The staging area, or data preparation area, is a collection of processes that

cleans, transforms, combines, and prepares source data for use in the data warehouse or data mart In a staging area, source system data is transformed into common formats, checked for consistency and referential integrity, and prepared to load into the data warehouse database A staging area:

! Is on one or several computers

! May not be based on relational technologies

! Does not support user reporting

! Can be built in relational or OLAP databases

! Can contain detailed or summarized data, which may or may not be shared across data marts

The definition of a data mart can vary In this course, the data mart is a subset of a data warehouse with data specific to a particular subject or business activity The data marts you will create in this course will be OLAP databases

Data Warehouse

In this course, the data warehouse is defined as a virtual union of data marts with integrated information that is shared across data marts In other circumstances, a data warehouse may be defined as a centralized, integrated data store providing data to the data marts Either definition is correct

The definition of a data warehouse can vary from organization to organization In this course, the data warehouse is defined as a virtual union of data marts with integrated information shared across data marts

Note

Note

# Defining OLAP Solutions

! OLAP Databases

! Common OLAP Applications

! Relational Data Marts and OLAP Cubes

! OLAP in SQL Server 2000

In the previous section, you learned about data warehousing and the flow of data from source systems to users This section focuses on one area of the data warehouse—the OLAP database The section introduces OLAP databases, describes common applications implemented by using OLAP technology, differentiates relational data marts and OLAP cubes, and describes the OLAP database solution available in Microsoft® SQL Server™ 2000

Topic Objective

To define OLAP solutions

Lead-in

This section introduces

OLAP solutions and defines

how they are used to

provide users with fast,

flexible data access

OLAP Databases

! Optimized Schema for Fast User Queries

! Robust Calculation Engine for Numeric Analysis

! Conceptual, Intuitive Data Model

! Multidimensional View of Data

$ Drill down and drill up

$ Pivot views of data

OLAP technology provides an alternative to relational database technology, offering fast, flexible data viewing, analysis, and navigation The following are characteristics of OLAP technologies:

! OLAP databases have an optimized schema for fast user queries

OLAP queries are very fast, and allow for more interactive use from users than typical relational database management system (RDBMS) reporting applications OLAP cubes store various levels of summarized data in data structures highly optimized for user queries

! OLAP databases have a robust calculation engine for numeric analysis You use OLAP cubes for numeric analysis, from producing simple sales reports

to performing complex allocation algorithms Many advanced calculations performed by OLAP calculation engines cannot be performed by relational databases because of analytical limitations in the RDBMS database engines

! OLAP is a conceptual, intuitive data model

More than a particular database technology, OLAP is a conceptual, intuitive data model that users can easily understand without the development of custom reporting applications

technology, offering fast,

flexible data viewing,

analysis, and navigation

! OLAP provides a multidimensional view of data

Cubes provide a multidimensional view of data that extends beyond standard two-dimensional analysis OLAP allows flexible data viewing, analysis, and navigation

• Users can drill down and drill up through various levels of summarized

data In OLAP cubes, data is stored in both detailed and summarized levels OLAP cubes give users the opportunity to easily drill down—that

is, to double-click top-to-bottom through the summarized levels to more detailed levels of data—or drill up from lower levels to more

summarized levels of data

• Users can pivot views of data Users can easily switch the rows,

columns, and pages in OLAP reports The term pivoting defines the intuitive mouse action by users that changes the orientation of their reports

Common OLAP Applications

! Executive Information Systems

multidimensional data in graphical formats

! Financial applications Many different types of financial applications use OLAP databases for reporting, planning, and analysis Examples of financial applications include financial reporting, month-close analysis, product profitability analysis, budgets and forecasting, and financial modeling Financial analysts use OLAP extensively for ad hoc analysis of financial and operational data to answer questions from senior management

! Sales and marketing applications Many types of sales and marketing applications frequently use OLAP where slice and dice capabilities and timeliness of information are important Examples include booking and billing applications, product analysis, customer analysis, and regional sales analysis

! Operations applications OLAP databases are adapted to a wide range of operational analyses, including manufacturing throughput and efficiency, customer service effectiveness, and product cost analysis

Topic Objective

To introduce common OLAP

applications

Lead-in

OLAP databases are

adapted to a wide range of

business applications Let

us talk about some

Discuss the types of OLAP

data marts with which you

may have worked

Relational Data Marts and OLAP Cubes

Relational Data Mart

Relational

Non-relational Sources

Relational and Non-relational Sources Non-relational SourcesRelational and

Relational and Non-relational Sources

Data Extract Queries

Fast Performance for Data Extract Queries Faster Performance for Data Extract Queries

Faster Performance for Data Extract Queries

Most organizations use a combination of relational data marts and OLAP cubes

to meet their decision support needs

Given their common decision support goals, relational data marts and OLAP cubes differ greatly in data storage, data content, data sources, data retrieval, and business analysis capabilities

Data Storage

Relational data marts and OLAP cubes differ in how they store data:

! Relational data marts store data in structures supported by relational database technologies

! OLAP cubes store data in multidimensional structures These structures can use both relational and multidimensional database technologies

Data Content

Relational data marts and OLAP cubes differ in their data content:

! Relational data marts store detailed and summarized data in relational structures

! OLAP cubes store summarized data in n-dimensional structures

Data Sources

Relational data marts and OLAP cubes differ in how they are sourced:

! Relational data marts can centralize source data from one or many heterogeneous source systems that may or may not be relational

! OLAP cubes can be sourced from relational or non-relational sources, and

Topic Objective

To explain the relationships

between relational data

marts and OLAP cubes

Lead-in

It is important to understand

the differences between

relational data marts and

OLAP cubes

Point out that the table

presents a a general

comparison The

comparision points may vary

depending on the particular

relational or OLAP

technology

Ask students to participate

in a comparision discussion

Data Retrieval

Relational data marts and OLAP cubes differ in how they retrieve data:

! Relational data mart structures are optimized for data retrieval

! OLAP cube structures are also optimized for data retrieval Because aggregated data is stored in these n-dimensional structures, query performance exceeds that of relational data marts

OLAP in SQL Server 2000

! Microsoft Is One of Several OLAP Vendors

! Analysis Services Is Bundled with Microsoft SQL Server 2000

! Analysis Services Include

OLAP technology is not unique to Microsoft Several companies distribute OLAP database engines Many companies also sell applications that provide user interfaces that interact with OLAP database engines

SQL Server 2000 includes two main OLAP components: the SQL Server database and Analysis Services Both products are included on the same installation media, but you can install either component without installing the other

Two distinct but overlapping tools are included in Analysis Services:

! The OLAP engine and its related components

! A data mining tool Data mining tools search for patterns in large quantities

of data

Analysis Services includes multiple types of OLAP storage schemes, such as:

cube structures that are separate from the relational database source of information

database

of multidimensional cube structures and relational database tables

# Understanding Data Warehouse Design

! The Star Schema

! Fact Table Components

! Dimension Table Characteristics

! The Snowflake Schema

Before you can create an OLAP database and understand its components, you must first understand the data warehouse components that you use to build the OLAP databases

This section describes data warehouse design concepts including the star schema, fact tables, and dimension tables It is important to understand how all the elements interact, because you define OLAP cubes from these data

warehouse components

This section discusses the following components:

! The star schema

! Fact table components

This section describes data

warehouse design concepts

including the star schema,

fact tables, and dimension

tables