WIXLIB

02 919216 ftoc.qxp11/21/0610:28 PMPage xiv

03 919216 flast.qxp11/21/0610:01 PMPage xvAcknowledgmentsWe begin by acknowledging the support of our families, especially ourwives who realize that authors sometimes get a bit cranky and difficult asdeadlines approach. Although they probably hope there is not another bookcoming from any one of us soon, we realize such an undertaking would notbe possible without the support of Jodie Stackowiak, Donna Rayman, andLuAnn Greenwald.Special thanks to the folks at Wiley Publishing who worked their magicto turn the documents and screen-captured images from our laptops intothe book you have in front of you. We would especially like to thank TomDinse, our Senior Development Editor, and Bob Elliott, Wiley’s ExecutiveEditor who understood the need for such a book.Within Oracle, we have had the great fortune to work with many peopleskilled in this area. From Oracle Development, we would like to acknowledge the following who have provided us with guidance over the yearsthat was especially relevant for this book: George Lumpkin, Robert Ash,Hermann Baer, Andrew Holdsworth, Paul Narth, Jean-Pierre Dijcks, PaulRodwick, Chris Leone, and Ray Roccaforte. We would also like to acknowledge the contributions of business intelligence specialists in the TechnologyBusiness Unit, especially Louis Nagode, Gayl Czaplicki, Derrick Cameron,Jim Bienski, Alan Manewitz, Joan Maiorana, and the Enterprise Technology Center under Jim Olsen where we were able to illustrate some of thelarge-scale management capabilities.xv

03 919216 flast.qxpxvi11/21/0610:01 PMPage xviAcknowledgmentsOracle also has a great many business intelligence specialists within sales,consulting, and other organizations around the world who remind us ofthe day-to-day challenges that their customers face when building thesesolutions. Some of the key individuals who influenced the content in thisbook include David Pryor, Susan Cook, Steve Illingworth, Nick Whitehead,Jon Ainsworth, Kevin Lancaster, Craig Terry, Joe Thomas, Rob Reynolds,Rich Solari, Nuge Ajouz, Ken McMullen, Brian MacDonald, and PatrickViau. There are many more, of course.Lastly, much of the content in this book is based on the experience of theauthors. Some of the descriptions of what to avoid are based on observations we made of less successful techniques used by Oracle’s customers.But many of Oracle’s customers and partners provide innovation and techniques that take product features and turn them into useful solutions. Wehave had the fortune of dealing with both types of customers, and thisbook is much stronger and realistic because of what these customers andthe Oracle partners have shared with us. So, thank you to all of you whomwe have worked with over the years. We especially hope you find thisbook to be of value as you build and develop your own solutions.

03 919216 flast.qxp11/21/0610:01 PMPage xviiIntroductionWe are now decades into deploying decision support systems, data warehouses, and business intelligence solutions. Today, there are many booksthat describe data warehousing and design approaches. There are manybooks that describe business intelligence. There are many books thatdescribe the Oracle database. So you may be asking, why did the authorsdecide to write this book?The fact is, the authors of this book still hear comments from many ofyou that business intelligence and data warehousing projects are problematic. This seems to be true regardless of database technologies or businessintelligence tools selected and deployed. While the wealth of Oracle skillsand resources that exist might make this less true where Oracle technologyis part of the solution, the number of implementations that face significantissues and the repetition of mistakes convinced the authors that too fewprojects are approached holistically. Not many of the books that are available as resources look at Oracle business intelligence and data warehousing in such a manner.This book attempts to give you a single reference that covers a diverserange of relevant topics in providing a holistic approach. It covers the database and platform technology, of course. But it also covers business intelligence tools, emerging business intelligence applications, architecture choices,schema selection, management and performance tuning, requirementsgathering, and justifying the project. Tips are included throughout the bookbased on real experience and implementations.xvii

03 919216 flast.qxp11/21/0610:01 PMPage xviiixviii IntroductionYour tendency might be to jump to sections you know something aboutor suspect as being a possible solution to a problem in order to furtherdevelop your knowledge of a specific topic. Although you should findvalue in using this book that way, keep in mind that the reason many implementations struggle is due to ignoring areas that should be understood andare outside the core competencies of those engaged in the project.To sum it up, the goal of this book is more than about gaining academicknowledge. If this book attains its objective, you will gain knowledge thatyou can apply to your own project such that your deployed solution will beviewed as successful technically within Information Technology (IT), butalso successful because it delivers the business value that your businesscommunity recognizes.Who This Book Is ForThis book should appeal to a wide audience. Although those in IT will findit particularly useful, more technically inclined business analysts and managers should also find value in topics such as justifying projects and evaluating deployment choices.Within IT, the day-to-day management and modification of such aninfrastructure often falls on database administrators, programmers, andsystems managers. Certainly, we cover topics of interest to this group.Unfortunately, the value of architects and project managers in deployingand updating such solutions is often overlooked. There is plenty in thisbook that should also appeal to that audience.How This Book Is OrganizedThis book is divided into three parts:Oracle Business Intelligence DefinedCustom Built Data Warehousing SolutionsBest PracticesPart I: Oracle Business Intelligence DefinedPart I provides a broad background as to possible Oracle-based solutionsand how you might deploy them. The database and business intelligencetools are introduced here, but other related topics are covered as well in thefollowing four chapters.

03 919216 flast.qxp11/21/0610:01 PMPage xixIntroductionChapter 1: Oracle Business IntelligenceA broad introduction of Oracle business intelligence is provided. Topicsintroduced include Oracle’s transactional business intelligence, integrationcomponents, and components in custom-built data warehouses and business intelligence solutions.Chapter 2: Oracle’s Transactional Business IntelligenceSometimes called operational business intelligence, this chapter describesOracle’s Daily Business Intelligence modules, Balanced Scorecard solution,and Data Hubs (used in master data management). You are provided withguidance as to where such solutions might be particularly useful and whya data warehouse might also be deployed to augment such a solution.Chapter 3: Introduction to Oracle Data WarehousingThe introduction of this topic covers the wide array of features in the Oracle database that are relevant in data warehousing. More detailed explanations are provided in Part II of this book. In addition, the chapter covers thedata models that Oracle provides for its applications as pre-built datawarehousing solutions.Chapter 4: Choosing a PlatformThe basics of choosing a hardware platform are covered including scalingup versus scaling out and how to size your choice. Specifics addressedunder these broad topics include high availability considerations, manageability considerations, and approaches to benchmarking.Part II: Custom-Built Data Warehousing SolutionsMost business intelligence solutions today are custom built. Part IIdescribes design approaches and deploying and managing business intelligence tools and an Oracle data warehousing database. These areexplained in the following five chapters.Chapter 5: Designing for UsabilityCovering approaches to design, topics in this chapter include how to leverageOracle features and an illustration of how these features can be used to provide solutions to needs driven by a business scenario. Schema approachesare described including third normal form, star schema, hybrids, andOnline Analytical Processing (OLAP).xix

03 919216 flast.qxpxx11/21/0610:01 PMPage xxIntroductionChapter 6: Business Intelligence ToolsThis chapter introduces using and deploying Oracle’s wide array of business intelligence tools, including portals, reporting, and ad hoc query andanalysis tools. The Oracle Business Intelligence Suites (Standard Editionand Enterprise Edition) are covered. In addition, the Oracle database support provided by business intelligence tools available from other vendorsis described.Chapter 7: Data LoadingEmbedded extraction, transformation, and loading (ETL) features providedby the Oracle database are described in this chapter. Oracle WarehouseBuilder’s role in ETL, target data warehouse design, data quality analyses,and metadata management is also described.Chapter 8: Managing the Oracle Data WarehouseOracle Enterprise Manager provides a useful interface often used in managing Oracle data warehouses as described in this chapter. The Grid Controlinterface for managing clusters is illustrated, as are interfaces for basic performance monitoring, administration, and maintenance.Chapter 9: Data Warehouse Performance Tuning and MonitoringTypical performance challenges are described and proven approaches tosolving such challenges are presented. We then illustrate using suchapproaches to tune the data warehouse first described in the business scenario presented in Chapter 5.Part III: Best PracticesUnderstanding the technology is great, but is no guarantee of success. PartIII will help you identify potential risk and best approaches for mitigatingrisk as you develop and deploy your solution. These best practices aredescribed in the following three chapters:Chapter 10: Scoping the Effort and an Approach for SuccessThis chapter describes how to uncover initiatives by your business community, securing business sponsorship, endorsing a methodology, projectstaffing, and managing risk.

03 919216 flast.qxp11/21/0610:01 PMPage xxiIntroductionChapter 11: Understanding Business NeedsBusiness needs for better business intelligence might be driven by a poorlydesigned solution or by a new business requirement. Examples of less optimal solutions and how they can impact the business are first described inthis chapter. Typical project drivers driven by business requirements are thenpresented followed by suggestions on how to build support for a project.Chapter 12: Justifying Projects and Claiming SuccessGetting the go-ahead to build a solution often requires financial justification. This chapter identifies the potential costs you should consider andwhere business benefits might come from. Financial benefits are computedfor a variety of scenarios and computing return on investment (ROI) isdescribed.Illustrations in the TextOracle product illustrations in the text are captured from recent Oraclesoftware versions. You should be able to leverage similar capabilitiesregardless of your Oracle software version provided you have Oracle Database 10g, Oracle Business Intelligence 10g, or newer releases of these products. We frequently indicate when key features were introduced in theseproducts so that if you have older releases deployed, you can understandlimitations you might face.From HereTo become an expert on this topic usually requires years of practice andlearning, implementations for a variety of companies and organizations,and the uncommon ability to feel equally at home discussing needs anddetails among both IT and business co-workers. For those that grow insuch expertise, there is great opportunity and potential reward.This book is intended to help lay that foundation. Of course, your success will depend not only on what you read in the following chapters, butalso on how you put what you learn here into practice in the solutions youwork on. As you now start reading this book, it is our hope that it will helpyou avoid many common pitfalls and that you will gain a better perspective on how to attain professional success in building and deploying suchprojects.xxi

03 919216 flast.qxp11/21/0610:01 PMPage xxii

04 919216 pt01.qxp11/21/0610:02 PMPage 1PA R TIOracle BusinessIntelligence DefinedIn This PartChapter 1: Oracle Business IntelligenceChapter 2: Oracle’s Transactional Business IntelligenceChapter 3: Introduction to Oracle Data WarehousingChapter 4: Choosing a Platform

04 919216 pt01.qxp11/21/0610:02 PMPage 2

05 919216 ch01.qxp11/21/069:59 PMPage 3CHAPTER1Oracle Business IntelligenceBusiness intelligence can be defined as having the right access to the rightdata or information needed to make the right business decisions at theright time. The data might be raw or might have been analyzed in someway. Having access to such information enables management of the business by fact instead of by primarily relying on intuition.This is a broad definition of business intelligence and is not limited todata warehousing alone. Although a data warehouse is often used to provide such a solution and is the primary focus of most of this book, we’llbroaden the discussion to also include business intelligence gained fromon-line transaction processing solutions. Business analysts and users ofbusiness intelligence don’t really care about — or want to understandwhere their information comes from. They simply want access to suchsources. So the solution you choose to deploy will depend on the kind ofinformation that is needed.This chapter provides a broad discussion of Oracle’s business intelligence offerings and should help you better understand all of the solutiontypes available for deployment. We conclude this chapter by discussingsome of the emerging business needs that will lead to a further blending ofdata warehousing and transactional systems. In subsequent chapters inthis section of the book, we provide more details as to how and why you’d3

05 919216 ch01.qxp411/21/06Part I 9:59 PMPage 4Oracle Business Intelligence Defineddeploy transactional business intelligence and data warehousing solutions. We also discuss some of the platform strategies for deployment.After the introductory first section of this book, we describe in muchgreater detail the area of business intelligence that you are probably mostinterested in: custom-built data warehousing solutions using Oracledatabases. We provide examples of how you can design, use, and managevarious capabilities of the Oracle database and Oracle business intelligencetools. In the final section of this book, we discuss best practices and strategies for deployment of such solutions.Although the primary audience of the book is information technology(IT) professionals, we begin this book with the following warning: building a business intelligence solution as an IT project without sponsorship ofor buy-in by the lines of business is likely to end in very limited success orcareer-limiting failure. For many of you, the non-technical portions of thisbook in the best practices section might initially be of the least interest,because your interest is centered in IT and implementing technology platform solutions. However, applying techniques described in that sectioncould determine whether your project is viewed as successful.Business Intelligence andTransactional ApplicationsTransactional applications generally provide business intelligence to business users through reports that reveal current data in transactional tables.Oracle’s E-Business Suite of applications, PeopleSoft applications, JDEdwards applications, and Siebel Customer Relationship Managementapplications all provide this level of business intelligence. Reporting isselected and deployed based on key business requirements (KBRs) andmost commonly displayed as key performance indicators (KPIs) in a dashboard using portal technology.Most companies also deploy business intelligence solutions that rely on acomplementary data warehousing strategy when reporting and analysisbecomes more complex and summary level information is appropriate.Oracle’s PeopleSoft and JD Edwards’ EnterpriseOne applications are oftensurrounded by the PeopleSoft Enterprise Performance Management (EPM)data warehouse to enable such reporting through analytical applications.Oracle’s Siebel Customer Relationship Management (CRM) applicationsare similarly often surrounded by Business Analytics Applications builtupon a relationship management warehouse model. Other application vendors (such as SAP with their Business Warehouse) have such data warehouse models that often are deployed on Oracle databases.

05 919216 ch01.qxp11/21/069:59 PMPage 5Chapter 1 Oracle Business IntelligenceThe Oracle E-Business Suite leverages more of a blended approach todelivering business intelligence applications as many of these applicationsrely on data in summary levels of transactional tables. The Enterprise Planning and Budgeting application, a more complex analytical application,leverages Oracle OLAP technology in a separate multi-dimensional cube.At the time of publication of this book, Oracle has described manyaspects of Project Fusion, Oracle’s future single set of transactional applications that provide a migration path for current deployments of theE-Business Suite, PeopleSoft, JD Edwards, and Siebel CRM applications.The business intelligence solutions provided for t

Oracle Business Intelligence Enterprise Edition 18 BI (XML) Publisher 20 Oracle Portal 20 . 02_919216 ftoc.qxp 11/21/06 10:28 PM Page ix. Chapter 2 Oracle's Transactional Business Intelligence 27 Transactional Business Intelligence 28 Business Terminology 28 Oracle's Daily Business Intelligence 30 How DBI Works 33 Varieties of DBI 34 .