Service Orchestration as Organization

Building Multi-Tenant Service Applications in the Cloud

Malinda Kapuruge, Jun Han, Alan Colman

• Publisher: Morgan Kaufmann
• Year: 2014
• ISBN:
  9780128010976
• Topic: Ciencia de la computación
• Subtopic: Redes de computadoras

Service orchestration techniques combine the benefits of Service Oriented Architecture (SOA) and Business Process Management (BPM) to compose and coordinate distributed software services. On the other hand, Software-as-a-Service (SaaS) is gaining popularity as a software delivery model through cloud platforms due to the many benefits to software vendors, as well as their customers. Multi-tenancy, which refers to the sharing of a single application instance across multiple customers or user groups (called tenants), is an essential characteristic of the SaaS model.

Written in an easy to follow style with discussions supported by real-world examples, Service Orchestration as Organization introduces a novel approach with associated language, framework, and tool support to show how service orchestration techniques can be used to engineer and deploy SaaS applications.

• Describes the benefits as well as the challenges of building adaptive, multi-tenant software service applications using service-orchestration techniques
• Provides a thorough synopsis of the current state of the art, including the advantages and drawbacks of the adaptation techniques available
• Describes in detail how the underlying framework of the new approach has been implemented using available technologies, such as business rules engines and web services

Part One

Outline

1 Introduction

2 Motivational Scenario

3 Literature Review

1

Introduction

Business organisations have been increasingly adopting business process management (BPM) as a part of their core business strategy due to benefits such as improved business–information technology (IT) alignment and optimised operational efficiency. Service-oriented architecture (SOA) has emerged as an architecture style that provides a loosely coupled, reusable and standards-based way to integrate distributed systems. Both SOA and BPM are mutually complementary development approaches for enterprise architectures. By combining the benefits of these complementary developments, service orchestration approaches are providing methodologies and technologies to enable the creation of integrated IT systems that meet the needs of rapidly changing enterprises and market places.

Keywords

BPM; adaptability; SOA; orchestration; cloud computing; SaaS

Business organisations have been increasingly adopting business process management (BPM) as a part of their core business strategy due to benefits such as improved business–information technology (IT) alignment and optimised operational efficiency [1]. Service-oriented architecture (SOA) has emerged as an architecture style that provides a loosely coupled, reusable and standards-based way to integrate IT systems [2]. Both SOA and BPM are mutually complementary development approaches for enterprise architectures [3, 4]. By combining the benefits of these complementary developments, service orchestration approaches provide methodologies and technologies to enable the creation of integrated IT systems that meet the needs of rapidly changing enterprises and marketplaces.

Over the past few years, service orchestration methodologies have evolved. New business and IT paradigms have emerged. The online market places increasingly become open, diverse and complex as more computer-mediated services become available to consumers, and as demand for those services increases [5]. The repurposing of systems to open up new revenue channels is an emerging norm. In such service market places, the changes to systems become unpredictable and frequent while consumers expect the downtime to implement these changes to be zero.

These developments have challenged the existing service orchestration methodologies. This book introduces a service orchestration methodology called Serendip to address those challenges. In this chapter, we introduce the research context, research goals and an overview of the proposed approach. Sections 1.1 and 1.2 introduce the context of this research. In Section 1.1, we provide an overview of BPM and how BPM is applied in service-oriented systems. In Section 1.2, we discuss service orchestration in the context of novel requirements and challenges. The research goals are introduced in Section 1.3. An overview of the approach is presented in Section 1.4, whereas Section 1.5 lists the key contributions of this book. Finally, Section 1.6 provides an overview of the book structure.

1.1 Business process management

According to Gartner [6], the BPM¹ market in 2009 totalled US$1.9 billion in revenue, compared with $1.6 billion in 2008, which was an increase of 15%. This means that there has been a vast increase in adopting BPM as part of the core business strategies of business organisations.

Due to the cross-disciplinary nature, BPM has multiple definitions. These definitions have changed over time and varied from one discipline to another. One of the widely recognised definitions of BPM in IT was given by van der Aalst [7]. According to van der Aalst, ‘Business Process Management (BPM) includes methods, techniques, and tools to support the design, enactment, management, and analysis of operational business processes.’

This definition leads to the question, what is a business process? Curtis [8] in a broader context defines a process as a ‘partially ordered set of tasks or steps undertaken towards a specific goal.’ In the business context, Hammer and Champy [9] define a business process as a ‘set of activities that together produce a result of value to the customer.’ In the literature, there are many other definitions with slight variations. Irrespective of these variations, the fundamental characteristics of a business process is the ordering of activities that is required to achieve a business requirement/goal.

The attraction of BPM to business organizations is primarily driven by the requirement of managing the business processes and continuously optimising and leveraging their IT assets. BPM provides the management model to efficiently achieve this objective by using IT. Such efficiency becomes an important requirement as businesses demand shorter cycle times for executing processes [10]. Hence, the main objective of BPM can be seen as the provision of efficient business IT support, compared to its non-IT or manual counterparts, which helps the organisation to function in a productive manner [11].

The adoption of BPM in businesses also formalises the understanding of business processes. Formalised business process descriptions provide a unified view on the processes of a business organisation. A unified view helps identify the gaps and the required improvements in business strategies. Consequently, the business strategists can fill the gaps or find better ways of accomplishing tasks [11]. The Barings bank failure [12] is a good example that highlighted the disastrous effects on a business of unidentified gaps in business processes. The improvements and adoption of computing theories, notations and verification methodologies into BPM have assisted the communication, understanding and verification of defined business processes. Furthermore, the advancing capabilities of BPM suites such as logging, monitoring, visualisation and process analysis also assisted the adoption of BPM.

The ever-increasing competition within business environments forces business organisations to change their business processes frequently. According to Gartner, ‘more than two-thirds of organisations using BPM do so because they expect that they will have to change business processes at least twice per year’ [6]. The report also stated ‘18% of companies have said that they needed to change processes at least monthly and 10% said that their processes changed weekly.’ Manually adapting and verifying such frequent changes can be impractical; hence, proper IT support is a must.

In conclusion, the factors such as requirement for increased efficiency and the ability to support frequent changes have collectively forced business organizations to practice BPM methodologies as part of their core business strategy.

1.1.1 BPM in practice

Business organisations practice BPM in different ways. These practices are selected depending on the business and technical requirements of the organisation.

The most rudimentary use of IT support involves the computer-assisted drafting of process flows or business process definitions. Usually, a software tool is used by business strategists or policymakers to quickly draft the activity flows or the process definitions and distribute among a set of involved partners, e.g. business owners, line managers, clerks and labourers. The objective of this practice is to convey a specific business strategy captured via a process definition. The partners abide by the process definition. If a change is required, then the process definition is reloaded to the software tool to apply due changes. Later, the changed process definition is redistributed among the partners. The process definitions used in this practice are usually diagrammatic. The idea is to let humans understand the business strategy quickly and easily. There is no intention to automate the enactment via machines. During the early stages, business organisations developed their own proprietary notations to describe such process definitions. Both business strategists and partners were required to learn these notations. Later, in order to standardise the way such processes are modelled and communicated among different partners, many standard process modelling notations were proposed [13–17]. Yet, the process execution was carried out manually without any automation.

Due to the lack of efficiency in such manual execution, business organisations required executable languages [7, 18, 19]. Such languages allow machines to understand and automate the execution of the business processes. Many business process instances can be instantiated and executed simultaneously with little or no human involvement. The advent of business-to-business communication intensified the need for automated process execution. Subsequently, the emerging process executable languages demand business process execution engines. This requirement is backed by leading vendors such as Microsoft, IBM, BEA and SAP who created a number of process execution engines, usually as part of a complete BPM suite. This development paves the way to placing the business process execution engines and the BPM suites at the heart of the enterprise systems [20].

One of the potent catalysts for increased business process automation is the advancement in distributed computing [21]. In distributed computing, multiple automated systems that are geographically distributed communicate with each other over a network. With the advancement of SOA [2, 5] and wide adoption of service-oriented computing technologies such as Web services...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• List of Figures
• List of Tables
• List of Listings
• Preface
• About the Authors
• Part One
• Part Two
• Part Three
• Bibliography
• Appendix A. SerendipLang Grammar
• Appendix B. RoSAS Description
• Appendix C. Organiser Operations
• Appendix D. Adaptation Scripting Language
• Appendix E. Schema Definitions