Emerging Trends in the Evolution of Service-Oriented and Enterprise Architectures

This chapter presents an introduction to emerging trends in the evolution of service-oriented and enterprise architectures. The primary aim of this book is to highlight some of the most recent research results in the field. Brief descriptions of the chapters included in the book are provided.

The evolution of Services Oriented Architectures (SOA) presents many challenges due to their complex, dynamic and heterogeneous nature. We describe how SOA design principles can facilitate SOA evolvability and examine several approaches to support SOA evolution. SOA evolution approaches can be classified based on the level of granularity they address, namely, service code level, service interaction level and model level. We also discuss emerging trends, such as microservices and knowledge-based support, which can enhance the evolution of future SOA systems.

The implementation of service-oriented architectures is mostly driven by the motivation to create a flexible and maintainable IT. Whether this goal can be achieved or not strongly depends on the design quality of the services. For that reason, the services within a service-oriented architecture have to be created with care. In the past, several quality attributes and quality indicators were identified that provide information about the design quality of a service. These quality indicators were described with focus on method-driven services based on SOAP. However, today, services are often designed in a resource-oriented way using REST or similar approaches to enable technology-independent interactions. For that reason, this chapter maps the existing quality attributes and quality indicators onto resource-oriented web services. As result, architects and developers get a toolset to design resource-oriented web services in a service-oriented architecture systematically in a quality-oriented manner. The quality indicators are illustrated by means of a resource-oriented web service in the context of a service-oriented SmartCampus system developed at the Karlsruhe Institute of Technology. The scenario shows that the application of the quality indicators limits the design scope and accelerates making design decisions.

Software systems based upon Service-Oriented Architecture (SOA) are often large, heterogeneous and difficult to understand. Evolving such systems presents some unique challenges. For example, it is critical to understand the impacts on trust relationships and security as SOA systems evolve. A substantial body of work exists on the idea of knowledge elicitation and management through the creation of knowledge models, which are created to represent the conceptual knowledge of experts. Knowledge modeling based upon concept maps is an efficient process and knowledge representation scheme that holds potential to assist planning in evolving SOA systems. This chapter contains two examples of knowledge modeling in support of SOA system evolution. The first example is an academic study that illustrates the use of knowledge modeling to create a software security assurance case. The second example, which is the main focus of this chapter, pertains to the ongoing evolution of a large, real-world Sustainment Management System software suite named PAVER™. This software is being modified to allow third-party add-in functionality to interact with the base system and to create a SOA federation with other enterprise systems. This article contains a description of a knowledge elicitation and modeling effort to identify trust concerns as this increasingly large and complex federation evolves.

Fractal concepts are often said to be recursively self-similar across multiple levels of abstraction. In this paper, we describe our experience with the fractal nature of SOA designs for sustainment management tools as these tools evolve into even more dynamic, federated systems that are orchestrated over the web. This chapter summarizes insights gained from more than twenty years of software development, maintenance, and evolution of a major pavement engineering tool named PAVER™. We consider both theoretical and experiential aspects of SOA federations at three levels of abstraction: (1) a loosely coupled federation of enterprise systems with PAVER™ as one member, (2) a tightly coupled federation of two pavement management tools (PAVER™ and PCASE) where each has a separate domain identity and development team, and (3) an emerging federation of plugin tools, which provide additional pavement engineering functionality and can come from competing civil engineering firms. These plugin tools exist at different levels of abstraction within the level of the main system and are, again, fractal. We organize the presentation of our experiences in this domain by describing how SOA elements including Ontologies, Discovery, Composition, and Orchestration are fractal whether we are looking at algorithms or persistent state. We also define and describe a third orthogonal fractal dimension: Evolution. Although the details of the implementation solutions at the differing levels of abstraction can be substantially different, we will show that the underlying principles are strikingly similar in what problems they need to solve and how they generally go about solving them.

The digital transformation of our society changes the way we live, work, learn, communicate, and collaborate. The digitization of software-intensive products and services is enabled basically by four megatrends: Cloud Computing, Big Data Mobile Systems, and Social Technologies. This disruptive change interacts with all information processes and systems that are important business enablers for the current digital transformation. The Internet of Things, Social Collaboration Systems for Adaptive Case Management, Mobility Systems and Services for Big Data in Cloud Services environments are emerging to support intelligent user-centered and social community systems. Modern enterprises see themselves confronted with an ever growing design space to engineer business models of the future as well as their IT support, respectively. The decision analytics in this field becomes increasingly complex and decision support, particularly for the development and evolution of sustainable enterprise architectures (EA), is duly needed. With the advent of intelligent user-centered and social community systems, the challenging decision processes can be supported in more flexible and intuitive ways. Tapping into these systems and techniques, the engineers and managers of the enterprise architecture become part of a viable enterprise, i.e. a resilient and continuously evolving system that develops innovative business models.

This chapter proposes a lightweight enterprise architecture framework which serves the demands of enterprise architects being confronted with a digital transformation scenario with an agile development approach. This framework therefore emphasizes the adaptability and the possibility for propagation of change throughout the defined architecture instead of addressing all possible concerns of all stakeholders. The framework itself is roughly based on TOGAF 9.1 and uses definitions of its content metamodel and follows ADM.

In all customer facing businesses, time to market is a key differentiator. The quicker a business is, the more successful it is likely to be. And today, more and more companies, even those not explicitly in technology, need to master technology so they can move quickly enough to survive (Weill and Woerner in MIT Sloan Management Review 56(4):27, 2015 [1]). For new, digitally native businesses, that’s no problem. They’re built for speed. But more established businesses, even successful ones, have for many years delivered technology solutions to their employees and customers on lengthy release schedules that no longer make sense in today’s accelerated environment. Based on our research and client work, we have developed and refined a two-speed architecture that lets more mature companies compete effectively with the upstarts.

Technological advances, changes in regulations and increasing globalization of the economy demand high adaptability from enterprises in many areas. Enterprise Architecture Management provides organizations with an integrated view enabling such adaptability. In this respect, development and management of the capabilities receive attention, as the term is associated with flexibility, dynamics and variation. On the contrary, little effort has been put towards developing and modeling capabilities. This chapter focuses on the Capability-Driven Development (CDD) method, which is a novel approach for designing capabilities to tackle the challenges of rapidly changing enterprise environments by modeling the application context. The results presented in this chapter are (i) a description of the state of research in capability development methods, (ii) a component-wise structured capability modeling method based on business processes, goals and concepts of an enterprise, (iii) a demonstration of the method application in a use case from the utilities industry and (iv) observations made during the capability development and strategy use.

Triggered by the progressive change from an industrial to an digital information society, not only social but also economic conditions are modified. Fast shifting business models and ever shorter product lifecycles are just few reasons why modern enterprises need a strategy how to deal with those unpredictable changes in order to stay competitive. Therefore, the concept of capability-driven management gets more and more attention by executives and scientists. In the last decade IS and management journals as well as conferences were publishing an increasing number of capability related articles, but a common understanding corresponding the identification of capabilities, their management, types or elements seems to be not existing. This work encapsulates the body of capability literature to provide an overview about capability research investigations over the last 15 years.

The discipline of Enterprise Architecture Management started using a model-driven approach. In contrary to the model-driven approaches, our approach follows strives to tap also the information contained in the operational systems that support IT-Service-Management. Therefore, this paper aims at indicating the increased capabilities of Enterprise Architecture Analytics and Decision Support through the use of a data-driven approach. It will give fundamental insights in the current research work of enterprise architecture management analytics as well as decision support based on this quantitative data.

Digitalization, shorter product cycles, oversupply of markets and the increasing customer requirements both determine and affect the movement from an industrial to an information society. As a result companies are faced with new challenges to keep their market position, transparency and efficiency. Enterprises overcome these challenges by implementing strategies. In order to implement strategies successfully and achieve desired goals enterprises should have certain capabilities. Thus, the demand for a methodical capability management approach is growing. This chapter introduces a process for identifying, structuring, and maintaining enterprise capabilities. The guide is based on an integrated capability approach that results from a number of scientific investigations and practical experiences performed over years. Comprised of four building blocks, the capability management guide represents a flexible “engineering” approach for capability catalog developers, designers and evaluators.