Module 1: Introduction to Data Warehousing and OLAP Contents Overview Introducing Data Warehousing Defining OLAP Solutions 11 Understanding Data Warehouse Design 18 Understanding OLAP Models 24 Applying OLAP Cubes 32 Review 40 Information in this document is subject to change without notice 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 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, are either registered trademarks or trademarks of Microsoft Corporation in the U.S.A and/or other countries 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP Instructor Notes Presentation: 60 Minutes Lab: 00 Minutes 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 Excel file 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY i ii Module 1: Introduction to Data Warehousing and OLAP 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 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 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 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 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 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 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP iii 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 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 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 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 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 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 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY iv Module 1: Introduction to Data Warehousing and OLAP 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP Overview Topic Objective To provide an overview of the module topics and objectives ! Introducing Data Warehousing Lead-in ! Defining OLAP Solutions ! Understanding Data Warehouse Design ! Understanding OLAP Models ! Applying OLAP Cubes In this module, you will learn about data warehousing, OLAP systems, and OLAP cube fundamentals 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP # Introducing Data Warehousing Topic Objective Introduce the concept of data warehousing ! OLTP Source Systems ! Data Warehouse Characteristics ! This section defines the differences between raw data and derived information, describes OLTP systems, and introduces data warehouse systems Raw Data vs Business Information ! Lead-in 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 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP Raw Data vs Business Information 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 Capturing Raw Data Deriving Business Information $ ! Gathering data recorded in everyday operations Deriving meaningful information from raw data Turning Data into Information $ Implementing a decision support system 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 ! Determining how these facts relate to other data ! Understanding how this data reflects overall business goals and objectives BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 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 weeks? In the last months? In the last 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 in terms of expense, time, and effort The return on this investment is reflected in how well the decision support system can satisfy the business needs of the organization BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 28 Module 1: Introduction to Data Warehousing and OLAP Dimension Family Relationships Topic Objective # To review key family relationships in a dimension USA North West Oregon Washington South West California Lead-in OLAP uses family relationship terms to describe how members relate to each other # # # # # # Delivery Tip Use the build slide to step through each of the dimension terms For each bullet of the list to the right, the colors of the members change—the primary members change to red, and the secondary members change to blue USA is the parent of North West and South West North West and South West are children of USA North West and California are descendants of USA North West and USA are ancestors of Washington North West and South West are siblings Oregon and California are cousins All are dimension members OLAP uses family terms to describe relationships between members and between levels of a dimension In the preceding illustration, the following are family relationships: ! Parent USA is the parent of North West and South West ! Child North West and South West are children of USA ! Descendant All members below USA are its descendants For example, North West and California are two descendants of USA ! Ancestor California has two ancestors, South West and USA USA is the ancestor of South West in addition to being its parent ! Sibling North West and South West are siblings to each other Oregon and Washington are also siblings Siblings share a common parent ! Cousin Oregon and California are cousins Cousins have parents that are siblings ! Member Regardless of specific familial relationships, everyone in the greater family is a member BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 29 Cube Measures Topic Objective To describe the characteristics of cube measures ! Are the Numeric Values of Principle Interest Lead-in ! Correspond to Fact Table Facts ! Intersect All Dimensions at All Levels ! Are Aggregated at All Levels of Detail ! Form a Dimension Measures are the numeric values of principle interest to users Measures are the numeric values that users analyze Every cube must contain at least one measure, but cannot have more than 1,024 measures total The following are true of measures: ! Measures must be numeric ! Measures correspond to fact table facts Only one fact table can be used to design a cube Therefore, measures can only come from one table in the cube’s data source ! Measures intersect all dimensions at all levels ! Measures are aggregated at all levels of detail across all dimensions For example, a Sales measure can be accessed at the monthly, quarterly, and year total levels of the Time dimension, in addition to all levels in all other dimensions of the cube ! All cubes by definition contain a Measures dimension You can think of the measures in a cube as members of the Measures dimension Note The Measure dimension contains only one level In other words, the measure dimension does not contain a hierarchy BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 30 Module 1: Introduction to Data Warehousing and OLAP Relational Data Sources Topic Objective To finish the section with a discussion of how the relational data source relates to the OLAP cube ! Star and Snowflake Schemas $ ! Lead-in Now that you understand the components of a data warehouse and an OLAP cube, let us summarize how the two are related Are required to build a cube with Analysis Services Fact Table $ $ ! Contains measures Contains keys that join to dimension tables Dimension Tables $ $ Use this slide to summarize the requirements for building OLAP cubes by using relational data sources Must exist in same database as fact table Contain primary keys that identify each member Analysis Services creates a new layer on top of an existing relational warehouse The purpose of that new layer is to make access to the data very fast and very flexible Star and Snowflake Schema Before you build a cube with Analysis Server, your source data must be staged in a relational database, in a star or snowflake schema To use Analysis Services successfully, you must know what Analysis Services expects the data warehouse to look like If students are confused at this point, let them know that the process of creating cubes reinforces all these data source requirements Fact Table To be usable by Analysis Services, the fact table in the cube data source must contain a column for each measure In addition, the fact table must contain dimension keys that join to dimension tables The fact table must contain rows at the lowest level of detail you might want to retrieve for a measure In other words, the fact table contains rows only for the lowest level of members of each dimension Analysis Services cannot use a fact table that stores aggregates, such as quarter and year totals Dimension Tables From an OLAP perspective, dimension tables serve three purposes A dimension table: ! Contains member names ! Contains the hierarchy definition ! Contains other attributes BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 31 Dimension tables must exist in the same database as the fact table Some organizations store dimensional information in separate, isolated data sources In such cases, all information must be consolidated into one data source before developing Analysis Services OLAP cubes In the data warehouse, the primary key column for a dimension table must contain a unique value for each member of the dimension The primary key column of each dimension table must match one of the key columns in the related fact table Each key value that appears once in the dimension table will appear multiple times in the fact table BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 32 Module 1: Introduction to Data Warehousing and OLAP # Applying OLAP Cubes Topic Objective To present the topics covered in this section ! Querying a Cube ! Defining a Cube Slice ! Working with Dimensions and Hierarchies ! Visualizing Cube Dimensions ! This section describes OLAP cube fundamentals by demonstrating the methods for visualizing multidimensional databases Defining a Cube ! Lead-in Connecting to an OLAP Cube This section describes OLAP cube fundamentals by demonstrating the methods for visualizing multidimensional databases Contents include: ! Defining a cube ! Querying a cube ! Defining a cube slice ! Working with dimensions and hierarchies ! Visualizing cube dimensions ! Connecting to an OLAP cube BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 33 Defining a Cube Topic Objective To introduce the concept of a cube Lead-in Cubes are the logical storage structures for OLAP databases Chicago Denver me ns ion Grapes Cherries Melons Apples Q1 The next few slides highlight query patterns of users by using the example cube introduced here Use this slide to define the cube and to describe the three dimensions in this example Q2 Q3 Time Dimension Q4 od uc ts Di Detroit Pr Market Dimension Atlanta Cubes are the logical storage structures for OLAP databases They combine dimensions and measures into intuitive and flexible models that users manipulate to create queries A cube defines a set of related dimensions that form an n-dimensional grid: ! Each cell of the cube holds one value, exactly like a spreadsheet ! The value of each cell is an intersection of the dimensions The cube in the preceding illustration stores sales for all products, all markets, and for all time periods To retrieve an annual total, users choose a product and a market, and sum up the four quarterly cells to retrieve the annual sales BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 34 Module 1: Introduction to Data Warehousing and OLAP Querying a Cube Topic Objective To demonstrate how users isolate data with cubes Lead-in Sales Fact Atlanta Markets Dimension Grapes Cherries Melons Apples Dallas Q2 Q3 Time Dimension Q4 Pr Q1 Di me n Denver sio n Chicago od uc ts Users must be able to isolate the data in which they are interested, in an intuitive and timely manner Delivery Tip This slide builds a query in the following order: Apples Q4 Atlanta Sales Fact When you press the ENTER key, the first three steps of the query highlight in the above order The Sales Fact intersection appears automatically if you not press ENTER The Sales cube in the illustration contains three dimensions: ! Time ! Products ! Markets Sales facts are stored at the intersections of all dimensions in the cube Users typically want to see only the data applicable to their market, to the products they manage, and in time periods relevant to their business questions A user who manages Apples in the Atlanta region wants to query the cube for Q4 Sales values The user can easily query this information without having to sift through all the other cube data Note Many times users cannot see other members and data because they not have security access to those members and data values The highlighted Sales Fact is considered an intersection, or a cell, in the cube One member from each dimension describes an intersection In this example, the following members describe the Sales intersection: ! Apples ! Atlanta ! Q4 BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 35 Defining a Cube Slice Topic Objective To define the concept of slicing OLAP cubes Atlanta Chicago Grapes Cherries Melons Apples Detroit Q1 This slide is not a build slide When delivering this slide, define the term slice in the context of the Cherry distribution manager Q2 Q3 Time Dimension Q4 Di me ns ion Denver Pr od uc ts The manager of Cherry distribution wants to view Cherry data across all time periods, for all markets Markets Dimension Lead-in The term slice is used in cubes to define a member or group of members that are isolated and then evaluated across other dimensions You can think of a slice as being a subset of a cube The manager of Cherry distribution wants to view Cherry data across all time periods, for all markets The manager queries a slice of the cube to analyze Cherries across all members of all other dimensions The term slice is also used as a verb, indicating the act of querying a subset of a cube For example, the Cherry manager slices the cube, analyzing the Cherry sales for all markets across all time periods BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 36 Module 1: Introduction to Data Warehousing and OLAP Working with Dimensions and Hierarchies Topic Objective To present the purposes of OLAP dimensions and hierarchies ! Lead-in Dimensions Allow You to $ OLAP dimensions and hierarchies provide the basis for navigating in a cube $ ! Hierarchies Allow You to $ $ Delivery Tip Deliver this slide by navigating the build slide For each second-level bullet, an animation illustrates the purpose of that concept Slice Dice Drill Down Drill Up One of the major purposes of an OLAP database is to provide a flexible, intuitive model for browsing data Dimensions and hierarchies provide that flexibility Dimensions Allow You to Slice and Dice A dimension consists of multiple members that can be compared to members of other dimensions A dimension is what enables you to slice and dice data in a cube ! When you slice a dimension, you select a single member from that dimension For example, you may want to focus on only a single product, such as Colony Muffins A slice allows you to ignore everything except that one product ! When you dice a cube, you put multiple members from a dimension on an axis and then put multiple members from a different dimension on another axis This allows you to view the interrelationship of members from different dimensions Hierarchies Allow You to Drill Down and Drill Up All dimensions contain a hierarchy, and for most dimensions, the hierarchy consists of multiple levels The multiple levels of a hierarchy allow you to drill up and drill down ! When you drill down on a member of a hierarchy, you see all the children of that member You can drill down one member at a time, or you can drill through all the members of a level at the same time Drilling down allows you to focus on specific data or a possible problem area ! When you drill up on a member or group of members in a hierarchy, you collapse the detail so that you see only the summary information for the parent of those members Drilling up allows you to see a bigger picture with fewer details BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 37 Visualizing Cube Dimensions Topic Objective To discuss the visualization of multidimensional data Lead-in Here we see how multiple dimensions can be illustrated by using spreadsheets Delivery Tip To deliver this content, open the DEMO_01.xls workbook that contains these worksheets It is not difficult to conceptualize a three-dimensional cube from a twodimensional relational table That is because a three-dimensional cube can be illustrated and visualized However, visualizing four or more dimensions can be difficult to accomplish Spreadsheets can be helpful for visualizing dimensions Key Points The word cube implies three dimensions In reality, an OLAP cube is ndimensional Think of the previous spreadsheet as a two-dimensional cube The cube contains two dimensions—Region and Time The cube stores sales across all regions and across all time periods The spreadsheet illustrates that every member intersects with every other member of the other dimensions This is true regardless of the level of the member’s hierarchy For instance, Portland intersects with Jan, Q1, and Year Total In the example, all intersections contain sales data However, most cubes not store data for all intersections Cubes usually contain missing data points in a majority of the cube intersections, or cells BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 38 Module 1: Introduction to Data Warehousing and OLAP Suppose a third dimension—Product—is added to the model In this example, the company sells three products: Gadgets, Gizmos, and Widgets One way to model the cube is by using multiple worksheets in the same workbook: Even though the model contains three dimensions, it still is true that every member intersects with every member of the other dimensions Suppose you add a fourth dimension—Employee—consisting of four members: Jones, Phelps, Smith, and Williams The cube can be visualized by using multiple, multi-sheet workbooks: BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 39 Connecting to an OLAP Cube Topic Objective State USA To connect to an OLAP cube by using a PivotChart Sales Units Lead-in 6000 We just saw how to model four dimensions in multiple spreadsheets Now, let us connect to a multidimensional cube by using a single spreadsheet Level 02 5000 Year Quarter 4000 Sheri Now mer - 2001 - Quarter Sheri Now mer - 2001 - Quarter 3000 Sheri Now mer - 2001 - Quarter Sheri Now mer - 2001 - Quarter 2000 Sheri Now mer - 2000 - Quarter Sheri Now mer - 2000 - Quarter 1000 Sheri Now mer - 2000 - Quarter Sheri Now mer - 2000 - Quarter Bagels Muffins Sliced Bread Bread Dairy Category Delivery Tips Click the chart in the slide during your presentation to automatically connect to a local cube file located in C:\Moc\2074A\Demo\ DEMO_01.cub The PivotChart automatically opens and you can navigate the cube in the chart When you drill down, select other members, and so on, the chart automatically connects to the cube file If the cube file does not exist in the defined location, the chart prompts you to connect to the proper cube source Cheese Deli Meats Frozen Chicken Meat Subcategory To fully understand the power of OLAP technology, you must observe a client application connecting to an OLAP cube Microsoft Excel 2000 PivotTables® and PivotCharts® connect to Analysis Services OLAP cubes and provide users with a flexible and intuitive interface to use when accessing OLAP data When you connect to a cube, you can: ! Drill up and drill down dimension hierarchies by using the mouse ! Easily change the axes in which dimensions reside ! Select individual members to slice dimensions ! View several members in a single dimension Without OLAP technology, you create an endless number of separate reports to perform the same analysis that simple mouse-clicks achieve in OLAP OLAP solutions reduce the need for standard reports, report writers, SQL query syntax, long query times, and report printouts Perform the steps outlined in the student notes to demonstrate to users the power of OLAP BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY 40 Module 1: Introduction to Data Warehousing and OLAP Review Topic Objective To reinforce module objectives by reviewing key points Lead-in The review questions cover some of the key concepts taught in the module ! Introducing Data Warehousing ! Defining OLAP Solutions ! Understanding Data Warehouse Design ! Understanding OLAP Models ! Applying OLAP Cubes What is the purpose of a staging area for a data warehouse? A staging area is a collection of processes that cleans, transforms, combines, and prepares source data for use in the data warehouse or data mart What is the purpose of OLAP? To provide fast, flexible access to multidimensional data for reporting and analysis Describe the differences between relational data marts and OLAP cubes Relational data marts store detailed and summarized data in twodimensional structures, optimized for data retrieval Relational data marts tend to store more detailed data than multidimensional data marts Multidimensional data marts contain data in n-multidimensional structures Because aggregated data is stored in these n-dimensional structures, query performance exceeds that of relational data marts BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY Module 1: Introduction to Data Warehousing and OLAP 41 Define the main relational components from which you build an OLAP cube Fact table—A central table in a data warehouse or data mart presenting numeric data in a context that describes a specific event in a business Measure—A quantitative, numerical column in a fact table Measures typically represent the values that users analyze Dimension table—A table in a data warehouse or data mart representing a business entity What must a fact table in an OLAP data source contain? The fact table in an OLAP cube data source must contain all cube measures In addition, the fact table must contain dimension information, usually in the form of dimension keys that join with the cube dimension tables BETA MATERIALS FOR MICROSOFT CERTIFIED TRAINER PREPARATION PURPOSES ONLY THIS PAGE INTENTIONALLY LEFT BLANK ... 24 Module 1: Introduction to Data Warehousing and OLAP # Understanding OLAP Models Topic Objective To introduce OLAP models Lead-in Now we will discuss OLAP models and their components ! OLAP Database... ONLY Module 1: Introduction to Data Warehousing and OLAP Overview Topic Objective To provide an overview of the module topics and objectives ! Introducing Data Warehousing Lead-in ! Defining OLAP. .. Module 1: Introduction to Data Warehousing and OLAP Data Retrieval Relational data marts and OLAP cubes differ in how they retrieve data: ! Relational data mart structures are optimized for data