Business and Scientific Workflows E-Book

Wei Tan

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Serie: IEEE Series on Systems Science and Engineering
• Sprache: Englisch

Seitenzahl: 352

Veröffentlichungsjahr: 2013

Beschreibung

Focuses on how to use web service computing and service-based workflow technologies to develop timely, effective workflows for both business and scientific fields

Utilizing web computing and Service-Oriented Architecture (SOA), Business and Scientific Workflows: A Web Service–Oriented Approach focuses on how to design, analyze, and deploy web service–based workflows for both business and scientific applications in many areas of healthcare and biomedicine. It also discusses and presents the recent research and development results.

This informative reference features application scenarios that include healthcare and biomedical applications, such as personalized healthcare processing, DNA sequence data processing, and electrocardiogram wave analysis, and presents:

• Updated research and development results on the composition technologies of web services for ever-sophisticated service requirements from various users and communities
• Fundamental methods such as Petri nets and social network analysis to advance the theory and applications of workflow design and web service composition
• Practical and real applications of the developed theory and methods for such platforms as personalized healthcare and Biomedical Informatics Grids
• The authors' efforts on advancing service composition methods for both business and scientific software systems, with theoretical and empirical contributions

With workflow-driven service composition and reuse being a hot topic in both academia and industry, this book is ideal for researchers, engineers, scientists, professionals, and students who work on service computing, software engineering, business and scientific workflow management, the internet, and management information systems (MIS).

Leseprobe

Contents

Cover

Series Page

Title Page

Copyright

Dedication

Foreword

Preface

Chapter 1: Introduction

1.1 Background and Motivations

1.2 Overview of Standards

1.3 Workflow Design: State of the Art

1.4 Contributions

Chapter 2: Petri Net Formalism

2.1 Basic Petri Nets

2.2 Workflow Nets

2.3 Colored Petri Nets

Chapter 3: Data-Driven Service Composition

3.1 Problem Statement

3.2 Data-Driven Composition Rules

3.3 Data-Driven Service Composition

3.4 Effectiveness and Efficiency of the Data-Driven Approach

3.5 Case Study

3.6 Discussion

3.7 Summary

3.8 Bibliographic Notes

Chapter 4: Analysis and Composition of Partially-Compatible Web Services

4.1 Problem Definition and Motivating Scenario

4.2 Petri Net Formalism for BPEL Service, Mediation, and Compatibility

4.3 Compatibility Analysis via Petri Net Models

4.4 Mediator Generation Approach

4.5 Bibliographic Notes

Chapter 5: Web Service Configuration with Multiple Quality-of-Service Attributes

5.1 Introduction

5.2 Quality-of-Service Measurements

5.3 Assembly Petri Nets and Their Properties

5.4 Optimal Web Service Configuration

5.5 Implementation

5.6 Summary

5.7 Bibliographic Notes

Chapter 6: A Web Service-Based Public-Oriented Personalized Health Care Platform

6.1 Background and Motivation

6.2 System Architecture

6.3 Web Service Composition with Branch Structures

6.4 Web Service Composition with Parallel Structures

6.5 Demonstrations and Results

6.6 Summary

Chapter 7: Scientific Workflows Enabling Web-Scale Collaboration

7.1 Service-Oriented Infrastructure for Science

7.2 Scientific Workflows in Service-Oriented Science

7.3 Summary

Chapter 8: Network Analysis and Reuse of Scientific Workflows

8.1 Social Computing Meets Scientific Workflow

8.2 Network Analysis of myExperiment

8.3 ServiceMap: Providing Map and GPS Assisting Service Composition in Bioinformatics

8.4 Summary

Chapter 9: Future Perspectives

9.1 Workflows in Hosting Platforms

9.2 Workflows Empowered by Social Computing

9.3 Workflows Meeting Big Data

9.4 Emergency Workflow Management

Abbreviations List

References

Index

IEEE Press

445 Hoes Lane

Piscataway, NJ 08854

IEEE Press Editorial Board 2012

John Anderson, Editor in Chief

Kenneth Moore, Director of IEEE Book and Information Services (BIS)

Technical Reviewer

Professor Xudong He

School of Computer and Information Services

Florida International University

Copyright © 2013 by The Institute of Electrical and Electronics Engineers, Inc.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey. All rights reserved.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data is available.

ISBN: 978-1-11-817133-2

From Wei TanDedicated to my beloved parents,Mr. Manliang Tan and Mrs. Xiaozhen Lin,for their long-lasting support.

From MengChu Zhou

In memory of Mr. Shenglong Zhou,my fatherwho had constantly inspired and encouraged me to strive toward a better life,better society, and better world.

Foreword

In 2001, as the first high-speed networks were deployed, the inimitable George Gilder observed that “when the network is as fast as the computer's internal links, the machine disintegrates across the net into a set of special purpose appliances.” More than a decade later, our networks are faster and more reliable than ever. It is now entirely feasible to decompose not only our computer systems but also our software systems and applications—and then to reconstitute them as services that can be accessed over the network and combined into a myriad of applications. By so doing, we can achieve efficiencies and a flexibility unimaginable when our hardware, software, and data had to be laboriously acquired, integrated, and maintained at a single location.

But just how are we to combine services to build our applications? In this new world, programmers must select and combine services to create workflows that meet specific requirements. But this new world is not more disciplined than the old. Services are defined and operated by many providers who may differentiate themselves according to the services they offer, their implementations of those services, or the price or performance of their implementations. Faced with a choice between different services and service implementations, how should programmers identify service instances that will meet goals for functionality, correctness, performance, and price? How do they combine services without acquiring knowledge of their implementation? How to determine what qualities of services they can expect from the resulting workflows? Without proper answers to these questions, the experience of creating and operating distributed workflows can be more frustrating than liberating.

These are the questions that Tan and Zhou take on in this book. They do not necessarily provide definitive answers—that is not yet a feasible goal. But they provide the reader with the knowledge and tools required to understand the questions, to evaluate new technologies in the rapidly evolving world of service-oriented workflows, and to make sense of results emerging from academic research. Bridging the theoretical and the practical concerns, Tan and Zhou describe methods for reasoning about workflows, methodologies for constructing workflows from services, and substantial applications of these methodologies.

A particularly attractive feature of this book, in my opinion, is its integrated view of services in business and science—two domains with increasingly similar requirements but still far too few connections. It irritates my colleagues in the science community when I talk about capturing and accelerating their business processes—by which I mean not their accounting practices, but the data collection, lifecycle management, and other informatics activities that take so much of their time. But in both business and science, automation of the routine activities is a key to competitiveness. Science certainly has much to learn from business in this regard. We should also expect that demanding science applications will motivate new methods and tools that can be useful to business. I hope that this book can contribute to a conversation between these two communities.

As our old computer systems disintegrate around us, to use Gilder's phrasing, an understanding of the service-oriented workflow will become increasingly important to programmers, software engineers, managers, and researchers. For some, the most important element of this understanding will be knowledge of tools, programming methodologies, and technical standards. For others, it will be learning to reason about workflows with a view to establish a formal foundation for the creation of a new type of software system. For yet others, it will be gaining insight into how service-oriented workflows work in practice. This book has something to offer each of these audiences, so read and enjoy.

Ian FosterDirector, Computation Institute,University of Chicago, USA;

Arthur Holly Compton Distinguished Service Professor,Department of Computer Science, University of Chicago, USA;

Distinguished Fellow, Argonne National Laboratory, USA

Preface

This book describes the technologies to build better workflows in the context of services computing, and the applications in both business and scientific domains. Originating from office automation, workflows are the computer models and automation mechanisms of business processes. Nowadays, workflow management is usually an indispensable subsystem in enterprise information systems. In the meanwhile, Web services technology, emerging from the early 2000s, has laid out a new foundation, that is, the service-oriented architecture (SOA), for the componentization, reuse, and interoperability of enterprise software. The interplay of workflow technology and SOA has provided a way to reuse existing services, quickly compose them into workflows, and adapt business processes with evolving requirements. Moreover, workflows themselves can be exposed as new services that can be used and/or composed by others.

However, there is still a big gap between academic research and real applications. This book tries to present the authors' efforts on advancing service composition methods and applying them to both business and scientific software systems, with both theoretical and empirical contributions. Methodology-wise, lightweight and flexible approaches are needed to compose services together, without imposing the need to acquire too much metadata or adapt services' internal workings. Application-wise, research results need to be incorporated into software tools to benefit academic researchers, industrial practitioners, and service users.

This book addresses both methodology and application problems in service-based workflows. We present a lightweight data-driven service composition approach, a mediation-aided approach to analyze and compose services, and a quality-of-service aware Web service functional configuration method. In the application aspect, we present the application of Web service composition techniques in a health care service platform and Web services and scientific workflows in e-Science. In particular, we discuss the network analysis and reuse of scientific workflows and introduce the social network concept into the workflow area.

Compared with the prior state of the art, this book has a salient feature, that is, it covers both research and engineering aspects. Since its first author is from an industry lab and the second author is a university professor, this book provides a unique perspective of business and scientific workflows from both academia and industry.

Organization of the Book

Chapter 1 provides an overview of the theme of this book, that is, the design of business and scientific workflows under a services computing paradigm. First, it gives an introduction to Web service technology and workflow technology. Then, it presents the industry standards in both areas, that is, services computing and workflow. Afterwards, it surveys five research topics, that is, automatic service composition, mediation-aided service composition, verification of service-based workflows, support for decentralized execution, and scientific workflows.

Chapter 2 provides a preliminary introduction to the formalism of Petri nets, its extensions, and applications to workflow and service composition. We make the Petri net formalism a standalone chapter because it is used in Chapters 3–5 of this book.

Chapter 3 presents a lightweight, data-driven approach for service composition in a Petri net framework. We utilize data relations in both business and service domains, and add data mediation constructs to make the data model in these domains complete and coherent. We devise three composition rules, that is, sequential, parallel, and choice, based on the augmented data model. On the basis of the data relations and composition rules, we propose a formal method to derive all the possible composition candidates, given a service portfolio. A prototype system is developed and an example is given to validate our approach as well as the algorithm.

Chapter 4 addresses how to compose partially compatible services and meet user requirements. Partial compatibility refers to the situation where two (or more) Web services provide complementary functionality and can be linked together in principle; however, their interfaces and interaction patterns do not fit each other as desired. Given two services whose interface invocation constraints are described by Business Process Execution Language (BPEL), we analyze their compatibility and adopt mediation as a lightweight approach to make them compatible without changing their internal logic. We first transform BPEL programs into a service workflow net, which is a type of colored Petri nets. Based on this formalism, we analyze the compatibility of two services and then devise an approach to verify whether there exists any message mediator whose composition does not violate the constraints imposed by either side. The method for mediator generation is finally proposed to assist the automatic composition of partially compatible services.

Chapter 5 deals with the automatic configuration of services under practical constraints. First, all Web services are discovered according to the customized or application-specific functional requirements. Second, by analyzing function decomposition and function selection on the service interface information, a complete service functional dependency configuration net based on Petri nets is built. Third, the quality-of-service (QoS) attributes for the whole configuration are chosen and computed. A transformation method is presented to convert nonlinear aggregation functions to linear ones. Relative importance or value trade-offs of different attributes are represented through subjective preference or perception. Fourth, the QoS attribute value for each real Web service is gained. An association algorithm translates and compiles QoS attributes. Finally, the linear programming problem is set and solved based on a Petri net mechanism to identify the best composition.

Chapter 6 presents a Public-oriented Health care Information Service Platform (PHISP). It can provide individuals with health information management, and intelligent and personalized health care services; for some specific diseases, it supports basic remote health care and guardianship. In the platform, most of the functional modules are packaged in the form of services, and in order to realize the personalized customization and active recommendation of intelligent health care services for individuals, several key techniques for service composition are used, which can support branch and parallel control structures in the process models of composite services. The implementation status of the platform is also described.

Chapter 7 focuses on the application of workflow and Web services in e-Science. We start with an introduction to the data deluge phenomenon in science—specifically in biological sciences. Then we introduce the paradigm of service-oriented science (SOS) in which Web services are virtual access points to data and computational resources. We introduce the design and implementation of caGrid (Cancer Grid), which is a service-based data and computation infrastructure. Afterwards we discuss the requirements for developing scientific workflows for caGrid and explain how we fulfill these requirements. The caGrid Workflow Toolkit, an extension to the Taverna workflow system, is designed and implemented to ease building and running caGrid workflows. It provides users with support for various phases in using workflows: service discovery, composition and orchestration, data access, and secure service invocation. We also present real-life and service-based scientific workflows at the end of this chapter.

Chapter 8 takes a step further based on the work in caGrid and the caGrid Workflow Toolkit. The wide adoption of scientific workflows is hampered by the fact that scientists are unaware of the existence of services, and thus, are not able to effectively incorporate the best practices and tailor them in their new experiments. Our solution to this challenge is to introduce the social network concept into scientific workflows. We try to address the issue of network analysis and reuse of scientific workflows in a why-what-how approach. Why: first we review the advances in social network technology and its far-reaching impact on science. What: we then present a network analysis on “myExperiment,” an online biological workflow repository. By examining the relationship among workflows and services, we reveal the usage pattern of services in scientific workflows, and how this knowledge can be extracted to facilitate reuse. How: based on a network-based model called ServiceMap, we aim to provide a GPS-like support to (1) help domain scientists better understand various usage patterns of the existing services and (2) provide a system-level support to recommend possible service compositions.

Chapter 9 summarizes the contribution of this book and highlights a few future research directions.

The following figure illustrates the organization of the contents in this book.

How to Read This Book

This book can be used as a reference book for either researchers in areas of business process management and services computing, or engineers who want to develop service-based workflow systems in business, scientific, and medical applications. More broadly, researchers and practitioners in areas such as software engineering, distributed computing, and management information systems (MIS) should find the topics of their interest.

All readers are encouraged to read Chapters 1 and 9, since they introduce the state of the art and point out some future directions, respectively. Chapters 2–5 are for readers who are more interested in service composition methodologies; readers who are familiar with the Petri net formalism can skip Chapter 2 and directly go to Chapters 3, 4 or 5. Chapters 6–8 are for those who are more interested in applications; particularly, Chapter 8 discusses the incorporation of social network concepts into service composition. From Chapter 3, every chapter is self-contained except the network analysis in Chapter 8 which is built on top of the software described in Chapter 7.

Acknowledgments

From Wei Tan:

I appreciate the help from people with whom I collaborated, although it would be impossible to list all of them.

My sincere thanks go to Professor Yushun Fan, my Ph.D. supervisor who brought me into the exciting area of workflow management in 2002. I am grateful to my colleagues in IBM China Research Laboratory (CRL), especially Dr. Zhong Tian, who assisted me during my internship in 2005–2006. They motivated me to perform the work presented in Chapters 3 and 4 and awarded me the IBM Ph.D. Fellowship in 2006.

My special thanks go to Professor Ian Foster, known as “the father of the Grid,” at the University of Chicago and Argonne National Laboratory, who was my mentor and colleague during 2008–2010. I gained immense knowledge from his vision and constructive discussions with him. I would also like to thank two renowned scholars in the area of e-Science, Professor Carole Goble, at the University of Manchester, and Professor David De Roure, at the University of Oxford, for their collaboration on scientific workflows.

I also thank my colleagues in caBIG project, especially Ravi Madduri, Dinanath Sulakhe, Stian Soiland-Reyes, and Alexandra Nenadic. I enjoyed the collaboration with Professor Jia Zhang at Carnegie Mellon University, Silicon Valley Campus, on the topic of network analysis of scientific workflows. Some of this collaboration is reflected in Chapter 8. The work in Chapters 7 and 8 was funded in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. N01-CO-12400, and the Google Summer of Code program (2009–2010).

I would like to thank my manager at IBM T. J. Watson Research Center, Ms. Liana Fong, who supported me while working on this book. My friends, Dr. Bo Liu, Dr. Yanhua Du, Dr. Jianwu Wang, Mr. Keman Huang, Dr. Zi Ye, Dr. Wei Tang, Dr. Xitong Li, and Ms. Lina Chen helped proofread some parts of the book, which I greatly appreciate.

Finally I would like to thank Professor MengChu Zhou who is my mentor and the coauthor of this book. He is a renowned scholar in Petri nets and discrete event systems, and I really enjoyed our collaboration on the use of Petri nets in various studies on workflow and service composition.

From MengChu Zhou:

There have been numerous collaborations behind this book and its related work. It would have been impossible to achieve without the help of the following collaborators, some of whom are already mentioned in the first author's message.

Dr. Pengcheng Xiong and Professor Yushun Fan collaborated with me for a number of years; we promoted the application of Petri nets to the area of Web composition successfully. We applied the disassembly Petri nets to the optimal configuration of Web services subject to environmental changes by considering multiple QoS measures, and adopted Petri net siphon and related theory for compatibility analysis, and a Petri net approach to the identification and resolution of protocol-level service composition mismatches. Some of our collaborative work is reflected in Chapter 5.

I enjoyed the collaboration with Dr. Pengwei Wang, Dr. Guanjun Liu, Dr. Ping Sun, Dr. Xianfei Tang, Professor Zhijun Ding, and Professor Changjun Jiang at the Key Laboratory of Embedded System and Service Computing, Ministry of Education, Tongji University. In recent years, Professor Changjun Jiang built up and led a world-class research group who made and are still making many important contributions in the areas of Web service theory and applications:

1. development of interactive Web service composition method based on Petri nets;

2. development of automated Web service composition method based on Horn clauses and Petri nets;

3. development of automated Web service composition method supporting both parallel and conditional branch structures;

4. design and implementation of a public-oriented personalized healthcare platform based on service-oriented architectures and Web service composition methods; and

5. invention of interactive Petri nets to describe interacting Web services and workflow systems.

Some of these research results are reflected in Chapter 6.

I have enjoyed the great support from my family for many years. Albert Zhou, my older son, currently a college student, helped proofread some parts of the book, which I greatly appreciated.

The work presented in this book was in part supported by the National Basic Research Program of China (2010CB328101 and 2011CB302804), the National Natural Science Foundation of China (NSFC) (60773001, 61074035, 61173016, and 61034004), and the National Science Foundation under Grant No. CMMI-1162482.

Wei TanIBM T. J. Watson Research Center, USA

MengChu ZhouNew Jersey Institute of Technology, USA and Tongji University, China

Chapter 1

Introduction

With the emergence of service-oriented architecture (SOA), Web services are gaining momentum as key elements in enterprise information systems. Meanwhile, building business and scientific workflows using service composition has become an important method for system integration and process reengineering. Therefore, workflow-driven service composition is now a hot topic in both academia and industry. This book focuses on how to design, analyze, and deploy Web service-based workflows for business, scientific, and medical applications. This chapter discusses the state of the art in both technologies, that is, Web services and workflow management, with a focus on their impact on each other.

1.1 Background and Motivations

1.1.1 Web Service and Service-Oriented Architecture

In a 1996 report, Schulte and Natis of Gartner Inc. first used the term service-oriented architecture (SOA) to describe a style of multitier computing that helped organizations share logic and data among multiple applications and usage modes [1].

Thomas Erl, who is recognized as a major contributor in the area of SOA, describes it as an architecture that is open, agile, extensible, federated, and composable—one that is composed of autonomous, QoS-capable, vendor diverse, interoperable, discoverable, and potentially reusable services [2]. He further summarizes the following eight principles of SOA [3]:

Standardized Service Contract. Services should expose their interface and their level of service explicitly, in a standard way. That is the only information needed by a user.