Practical Guide to SAP HANA and Big Data Analytics

Thank you for purchasing this book from Espresso Tutorials!

Like a cup of espresso coffee, Espresso Tutorials SAP books are concise and effective. We know that your time is valuable and we deliver information in a succinct and straightforward manner. It only takes our readers a short amount of time to consume SAP concepts. Our books are well recognized in the industry for leveraging tutorial-style instruction and videos to show you step by step how to successfully work with SAP.

Check out our YouTube channel to watch our videos at https://www.youtube.com/user/EspressoTutorials.

If you are interested in SAP Finance and Controlling, join us at http://www.fico-forum.com/forum2/ to get your SAP questions answered and contribute to discussions.

Related titles from Espresso Tutorials:

 Rob Frye, Joe Darlak, Dr. Bjarne Berg: The SAP® BW to HANA Migration Handbook

 Dominique Alfermann, Stefan Hartmann, Benedikt Engel: SAP® HANA Advanced Modeling

 Christian Savelli: SAP® BW on SAP HANA

 Deepa Rawat: Practical Guide to Advanced DSOs in SAP®

 Frank Riesner, Klaus-Peter Sauer: SAP® BW/4HANA and BW on HANA

 Bert Vanstechelman: The SAP® HANA Implementation Guide

Foreword

In today’s world, technologies are evolving at an extraordinary pace and new market opportunities are developing just as quickly. Companies face the challenge of adapting to unfamiliar technologies and markets with different rules and key players. While businesses adapt, IT departments face similar struggles with constantly changing business requirements and new technologies. In these hectic times, it is essential to have a good business intelligence foundation in order to keep track of both new and existing business.

SAP is one of the big players in business software and is currently developing new products and promoting new product suites at a speed previously unknown to most customers. Additionally, traditional SAP-based companies now need to combine their SAP systems with non-SAP-based software more than ever before, e.g. in the context of Big Data. New possibilities to maintain and provide company software, such as cloud platforms make architecture decisions more complicated, but also enable new business scenarios. The complexity lies in understanding the different architecture possibilities and deciding on the best option for your company.

This book aims to deliver clear recommendations for building a solid architecture based on the latest SAP HANA technologies; with an additional focus on the combination with Big Data platforms. We provide a detailed assessment of several possible architecture scenarios, a guideline on how to decide on one option or another, principles for processes, and the organization around such an architecture.

The target audience of this book is mainly SAP BI and Big Data architects, as well as IT architects. However, we welcome anyone else to dive with us into the wide world of SAP HANA BI opportunities. Readers should have a fundamental understanding of how a data warehouse functions, and the associated technologies, as well as a familiarity with Big Data environments.

Personal dedication

This book is the result of a great deal of encouragement, support and contribution from friends, families, and colleagues. Their numerous ideas, hints, and discussions helped us to greatly enrich each chapter. We would specifically like to mention Bikas Panigrahi who provided valuable input during the course of writing this book.

Last but not least, we want to say thank you to the Espresso Tutorials team, especially to Alice Adams, who patiently advised us in formalizing and completing the book.

We have added a few icons to highlight important information. These include:

Tip

Example

Attention

Finally, a note concerning the copyright: all screenshots printed in this book are the copyright of SAP SE. All rights are reserved by SAP SE. Copyright pertains to all SAP images in this publication. For the sake of simplicity, we do not mention this specifically underneath every screenshot.

1 Introduction

In this chapter, we highlight the reasons for writing this book and we explain the scope and content. We conclude the chapter by defining frequently used terms.

1.1 Intention

A key element for successful company management is a solid Business Intelligence (BI) solution, including a constant, up-to-date and holistic view of all business processes. A comprehensive BI platform provides decision makers at all levels of management with crucial reports and analyses for evaluating the current state of their organization, specifically the area they are responsible for. The underlying data basis has changed significantly in recent years. At the beginning, intra-company data was loaded via batch processing on a nightly basis. Now, BI (according to the definition in this book) includes external data, the processing of data in real-time, on-the-fly calculation of statistical models, and quick, user-friendly, well-defined reports; all with only a few clicks.

This evolution stems mostly from technical capabilities created by large-scale, in-memory computing, volume and velocity provided by distributed computing clusters, scalability, and advancements in UI out-of-the-box solutions. These technological developments have led to completely new business models, such as social media platforms, and cloud providers, and also to new technical solutions such as Big Data platforms, sensor analytics and real-time reporting. With these technical developments, you have the possibility for a much more elaborate business process analysis, also taking external information into account.

Suddenly, companies have the ability to tap into large social networks, delivering product opinions and trends on a scale never seen before. Weather data can be constantly evaluated in order to calculate better transport routes in logistics, or take storms and earthquakes into account for business and risk calculations.

However, in order to fully benefit from these new BI opportunities, you need a solid technical foundation together with a pervasive organizational change management within the company. Several questions need to be answered and decisions have to be made:

 Which technologies are best used within my company in order to reap all the described benefits?

 How do these technologies integrate in a consistent manner?

 What architectural principles should be followed during implementation, especially with a combination of technologies?

 How do I have to adapt my organization to fully leverage the capabilities of these new technologies?

These are just some examples of the fundamental questions that you need to address. Today, many vendors supply tools for comprehensive BI reporting and analysis, as well as for Big Data and in-memory computing. One of the vendors providing a consistent, integrated BI solution is SAP. Their portfolio, originally only simple SAP BW technology, has been significantly expanded over recent years. You now find new and improved products for reporting (e.g. SAP Lumira, Analysis for Office, SAP Design Studios etc.), for in-memory computing (e.g. SAP HANA), for predictive analytics (e.g. SAP Predictive Analytics suite), and for data warehousing and Extraction, Transformation and Load (ETL) operations (e.g. SAP BW, SAP Data Services etc.). In addition, connectors to Big Data environments have been continuously enhanced.

In this book, we provide a clear view of the latest technology trends in the SAP-based BI area, especially in conjunction with non-SAP Big Data technologies. Building on our project and lab experience, we give answers and recommendations on how various technological components can be combined in the most efficient and architecturally sound way in order to fulfill your company’s analytics needs.

1.2 Objective

Our goal in this book is to provide clear recommendations for building a solid architecture based on the latest SAP HANA technologies; combination with Big Data platforms included. We provide a detailed assessment of several possible architecture scenarios, a guideline on how to decide on one or the other, principles for processes, and the organization around such architecture.

1.3 Target audience

This book is aimed at SAP BI and Big Data architects, as well as IT personnel responsible for future-proof analytics solutions; but anyone is welcome to dive with us into the wide world of SAP HANA BI opportunities. Readers should have a fundamental understanding of how a data warehouse functions, and its associated technologies, as well as the technologies related to Big Data environments.

1.4 In scope/out of scope

This publication aims to establish a solid decision basis for your company’s SAP HANA-based BI architecture by providing recommendations and best practices following the latest SAP product and technology trends. These recommendations include:

1 general architecture scenarios consisting of an SAP HANA native data warehouse, an SAP BW/4HANA data warehouse (including SAP S/4HANA Embedded Analytics), a Big Data platform with a focus on SAP integrative use cases, an Advanced Analytics solution, and mixed scenarios;

2 individual architecture scenarios with a mix of the above-mentioned solution components;

3 a decision matrix to assist in selecting the best architecture components and how to reach this decision, and

4 structuring your organization and its processes in order to facilitate the breadth and complexity of your technical architecture with the different components.

The recommendations for the defined scope result from project experience, lab sessions, and the authors’ extensive work with these technologies.

This book does not cover non-SAP-based data warehouse solutions (excluding Big Data-based solutions), or detailed front-end integration evaluation. We focus purely on the latest SAP technologies and specifically exclude outdated SAP BW objects (such as InfoCube), SAP HANA 1.0 attribute and analytical views, and SAP ERP/Suite on HANA with SAP HANA Live.

1.5 Content

This book is organized in a logical sequence to help you explore the various architecture options, associated technologies and organizational principles related to SAP HANA BI architectures. We begin with an overview of relevant architectural elements and technologies in SAP HANA BI and Big Data platforms (in conjunction with SAP HANA). Next, we build on these elements by combining them with possible architectural scenarios and options. This includes a decision matrix of how to find the right scenario for your individual needs. Finally, based on the new technologies and architectures relevant for your individual needs, we look at the impact on your organization itself, and the related processes. We close with an overall summary and an overview of further topics resulting from the points covered in this book.

Let us have a brief look at the chapter content.

Chapter 2 looks at the latest technologies used within an SAP-based BI Architecture. As the core element of any current SAP data warehouse architecture is SAP HANA, we named the chapter accordingly. It contains core SAP HANA features and introduces the reader to the latest Big Data technologies and to the current SAP HANA-based integration possibilities with Big Data environments. We continue by explaining possible cloud scenarios and service models before we switch over to the possible front-end tools that operate with SAP and Big Data platforms. The last part of this chapter looks at the various solutions available for ETL in its different forms for Big Data platforms, as well as for data warehouses.

Chapter 3 combines the components presented in Chapter 2 and merges them into solid architectural scenarios. Determining which scenario best fits your individual needs depends on your previously identified requirements. The first section of this chapter discusses details of an SAP BW/4HANA-based architecture scenario. We continue with an SAP HANA native scenario, describing the advantages and challenges when implementing an SAP HANA native data warehouse. These two sections lead to a detailed discussion on how SAP HANA can most efficiently be merged with a Big Data platform. In addition, we focus on predictive analytics platforms utilizing the SAP HANA database and cloud implementations for all scenarios. We finish with best practices for a migration and a decision matrix, as well as best practices for deciding on the right architecture scenario.

Chapter 4 continues with organizational and procedural changes resulting from the move to a new BI architecture. We introduce the topic of landscape enablement, which includes subjects such as sizing, BI roadmap visioning, interfaces, and components. The next section moves into the area of governance, especially Data Governance with its required roles, responsibilities, and processes. Next, we discuss parallel development in a highly integrated environment, end-to-end testing, and debugging. We conclude with recommendations for security, authorization, and change processes, including training approaches for your existing team.

Chapter 5 provides a summary of the previous chapters and takes a look at further topics that should be investigated. We also outline what we see next on the horizon for SAP HANA BI eco systems. This chapter specifically refers to the latest SAP developments; e.g. SAP Leonardo and SAP Data Hub.

1.6 Definition of terms

Most of our technological terms are explained in Chapter 2, but there are additional, fundamental terms that we outline here.

Let us start with Business Intelligence (BI). Within the context of this book, BI is defined as all methods for gathering, analyzing, evaluating and reporting data. Furthermore, we view Big Data platforms, used for data analysis, as part of Business Intelligence. For the purposes of this book, the term Business Intelligence includes all technologies related to data gathering, analysis and evaluation.

As part of BI, data warehouses are data storage platforms, optimized for the analysis of structured data which has been gathered from several sources. They usually combine data in order to fulfill individual reporting needs.

The term Extraction, Transformation and Loading (ETL) describes processes used for gathering data from several sources and writing them into a platform optimized for reporting. As the term implies, data is extracted from a source system, transformed and enriched with additional information, and then loaded into the target system. In recent years, especially with in-memory and Big Data platforms becoming more popular, the term has been slightly changed to ELT (Extract, Load, Transform). This involves the same processes, but in a different order, resulting in a better use of power in in-memory and Big Data environments for the execution of transformations with massive amounts of data.

Finally, the most important explanations in this book are for the terms BI architecture and architecture scenarios. We define architecture as an overall framework, which provides standards and policies to structure BI tools and associated developments, thereby helping to define the mainstays of your future BI ecosystem. We specifically exclude infrastructure requirements from this term, as it is not the focus of this book. We define architectural options for SAP HANA BI-related scenarios as individual options for fulfilling your specific reporting needs. In doing so, we introduce solid architectures while combining selected tools and technologies. In the best case, a described scenario can cover all your reporting or analysis requirements without requiring further tools and technologies.

2 Building blocks of an SAP HANA architecture

This chapter introduces you to the specific building blocks of the architecture scenarios that we explore and use for our architectural discussions throughout this book. We present you with SAP-specific technology and tools, as well as non-SAP tools that we believe fit well into a modern overall BI architecture. Each section provides an explanation of the tools, recommendations for their usage and further information.

Overview of SAP HANA related tools

Below, we provide an overview of the tools we discuss in this chapter, and give links to information on technologies that are relevant for architectural (SAP HANA) designs.

SAP HANA core features

SAP HANA server-side components

The components of an SAP HANA server such as index, name, statistic server, etc.

See SAP HANA Advanced Modeling

SAP HANA engines

Similar functionalities are bundled into engines

See Section 2.1.2

SAP HANA rules framework

Business rules can be defined by the end user

https://blogs.sap.com/2016/12/19/hana-rules-framework-hrf-blog-of-blogs

Extended applications services advanced (XSA)

The XSA engine sustains several SAP HANA internal applications and also builds the basis for web apps

See Section 2.1.3

SAP HANA deployment infrastructure

HDI uses containers to package functionality into one single application

See Section 2.1.4

SAP HANA Views

SAP HANA views are used to combine data for reporting

See Section 2.1.5

Libraries (PAL, BFL, AFL, etc.)

Libraries represent standard functionality that can be reused within SAP HANA

See Section 2.1.6

Building your SAP HANA BI architecture

Big Data technologies

Tools for managing and running a Big Data platform

See Section 2.7

Predictive analytics tools

Tools used for finding patterns in, and making predictions on, the data

See Section 2.4

SAP BO/HANA in the cloud

Using cloud computing for running SAP HANA and SAP BusinessObjects

See Section 2.8

Front-end tools

Tools for the visualization and reporting of data (e.g. SAP Lumira)