BI and Big Data Management

BI and Big Data Management

Ulrich Hambuch

––––––––

Translated by Philipp Strazny 

BI and Big Data Management

Written By Ulrich Hambuch

Copyright © 2017 Ulrich Hambuch

All rights reserved

Distributed by Babelcube, Inc.

www.babelcube.com

Translated by Philipp Strazny

Cover Design © 2017 Ulrich Hambuch

Babelcube Books and Babelcube are trademarks of Babelcube Inc.

Imprint

© / Copyright: 2017 Ulrich Hambuch

E-Mail: [email protected]

Web: http://www.infogenesis.de

First edition

English translation: Philipp Strazny

Cover, Illustrations: Ulrich Hambuch,

Cover image: http://www.pixabay.com

Images: http://www.pixabay.com and http://freeimages.com

Fig. 23 with permission from: ORAYLIS GmbH

This work and all its parts are subject to copyright. Any use outside of the confines of copyright is inadmissible without agreement of publisher and author. This applies in particular to electronic or other reproduction, translation, distribution, and publication.

Bibliographical data of the Deutsche Nationalbibliothek (German National Library):

Die Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic information are available online under http://dnb.d-nb.de.

If you do not change direction, you may end up where you are heading.

Lao Tzu

MeYouWe

The magical App for improved cooperation and human development.

––––––––

Introduction

Companies increasingly recognize that the analysis of business information (business intelligence) can generate decisive competitive advantages.

Thus, there is more emphasis on looking for strategies and techniques to make valuable business process data visible, available, and interpretable.

In addition, the compliance guidelines BCBS 239, Basel II and III, SOX, and Solvency II have led to legal requirements for a minimum level of quality in reporting and planning data and processes. The establishment of enterprise-wide data management thus continues to be one of the major challenges for IT and management in the years to come.

Data quality is an integral success factor in the establishment of an optimal information infrastructure. A 2002 study from The Data Warehousing Institutes (TDWI) calculates that poor data quality in the US cost about $622 billion.

A basic requirement for adequate data quality is standardized and integrated data and information management, and companies taking steps to reach this goal recognize the crucial role of metadata.

Metadata describe data. They abstract from specific applications and thus establish data neutrality. This allows data to be integrated and used in other contexts.

Many projects in the context of decision support information systems, in particular business intelligence systems (BI) or big data initiatives, fail due to insufficient data quality. Data quality deficiencies have ramifications ranging from the need for post hoc data correction over reduced acceptance of the BI system to suboptimal decisions and insufficient support of operative business processes.

Gartner market research stated in 2006: Poor data quality costs a typical organization 20% of revenue.... A 2011 study by the Würzburg researcher BARC determined that poor data quality has various negative effects. It causes workers to be less satisfied when they have to spend a lot of time on unnecessary data cleansing. 61% of the surveyed also report increasing costs from poor data quality. 47% noted a decrease in customer satisfaction.

The worldwide financial and economic crisis after 2007 can retrospectively also be regarded as a data quality crisis. Despite far-reaching compliance requirements, many financial service companies have not been able to aggregate and prepare their risk data in a way to adequately control their risks, and they are still struggling in 2017. Besides factors such as homogeneous conceptual understanding, a modernized process and system architecture, and data governance, adequate and comprehensive data quality management as well as a maximally integrated metadata management also play a crucial role in efficient and reliable data management.

In the era of Big Data, data is viewed as the new oil and the available data volume worldwide multiplies every year. The requirements for transparency and data stream quality continue to increase, because these are considered essential for partially or completely new applications in decision support and other areas.

Figure 1: Volume forecast for digital data generated per year from 2005 to 2020 worldwide (in exabyte). Source: Digital Universe study.

Metadata and metadata management are important aids for ensuring adequate data quality. Metadata fall into roughly two categories:

Table 1: Metadata categories

Data abstraction, i.e. the generation and use of appropriate metadata, could be a suitable tool for getting a handle on growing data volumes. However, enterprises and government institutions hesitate to invest in relevant projects and infrastructures for successful data management, because they generally focus primarily on data collection and