Next-Gen Digital Services. A Retrospective and Roadmap for Service Computing of the Future

Distributed information systems and applications are generally described in terms of components and interfaces among them. How these component-based architectures have been designed and implemented evolved over the years, giving rise to the so-called paradigm of Service-Oriented Computing (SOC). In this chapter, we will follow a 25-years-long journey on how design methodologies and supporting technologies influenced one each other, and we discuss how already back in the late 90s the ancestors of the SOC paradigm were there, already paving the way for the technological evolution recently leading to microservice architectures and serverless computing.

Automated Service Composition is one of the “grand challenges” in the area of Service-Oriented Computing. Mike Papazoglou was not only one of the first researchers who identified the importance of the problem, but was also one of the first proposers of formulating it as an AI planning problem. Unfortunately, classical planning algorithms were not sufficient and a number of extensions were needed, e.g., to support extended (rich) goal languages to capture the user intentions, to plan under uncertainty caused by the non-deterministic nature of services; issues that where formulated (and, partially addressed) by Mike, being one of his key contributions to the service community.

In this chapter, we look at the development of the original vision of automated service composition as AI planning, going from planning with extended (rich) goals, further developing into composition under uncertainty, extending it to other domains (and reformulating the service composition as composition of sensors and actuators), and then showing possible alternative techniques for highly scalable service composition at the expense of the richness of the domain representation.

The need for acknowledging and managing sustainability as an essential quality of software systems has been steadily increasing over the past few years, in part as a reaction to the implications of “software eating the world”. Especially the widespread adoption of the Everything as a Service (*aaS) model of delivering software and (virtualized) hardware through cloud computing has put two sustainability dimensions upfront and center. On the one hand, services must be sustainable on a technical level by ensuring continuity of operations for both providers and consumers despite, or even better, while taking into account their evolution. On the other hand, the prosuming of services must also be financially sustainable for the involved stakeholders.

In this work, we discuss the need for a software architecting approach that encompasses in a holistic manner the other two dimensions of software sustainability as well, namely the social and environmental aspects of services. We highlight relevant works and identify key challenges still to be addressed in the context of software systems operating across different models for cloud delivery and deployment. We then present our vision for an architecting framework that allows system stakeholders to work in tandem towards improving a set of sustainability indicators specifically tailored for the *aaS model.

This chapter investigates the use of Computational Intelligence (CI) to tackle two challenges in the area of services. The first is involved with providing efficient decision support for migrating from monolithic to service-oriented software, while the latter addresses automatic service composition, which is a special form of service migration. Migration to service-oriented architecture (SOA) is influenced by a number of different and intertwined factors. These factors are identified through literature review and expert consultation. Different CI models, such as Fuzzy Influence Diagrams and Fuzzy Cognitive Maps, are employed to organize the factors and study their behavior. Various simulations are conducted that enable decision makers to execute what-if scenarios and take informed decisions as to whether to migrate or not to SOA, as well as to study the decisive factors contributing in favor or against this migration. Service synthesis is a tedious task considering on one hand the plethora of available services and on the other their different, often conflicting characteristics. Automation of this task is therefore a critical issue which deserves attention. In this context, the challenge of automatic service synthesis is addressed through specific methods and techniques based on Evolutionary Computation to achieve such automation to the best possible extent.

Pattern languages are well-established in the software architecture community. Many different aspects of creating a software architecture are addressed by such languages. Thus, several pattern languages have to be considered when building a particular architecture. But these pattern languages are isolated, i.e. it is hard to determine the relevant patterns to be applied from the different pattern languages. Moreover, the sum of patterns from different languages may be huge, i.e. restriction to relevant patterns is desirable. In this contribution we envision an encompassing tool, the pattern atlas, that supports building complex systems based on pattern languages. The analogy to cartography motivates the name of the tool.

Workflows have gained momentum in the last two decades, in industry as well as in modern sciences, as a means for modeling and enacting processes. The value of workflow specifications does not end once they are enacted. Indeed, such specifications encapsulate knowledge that documents business processes or scientific experiments, and are, therefore, worth storing and sharing to ultimately be reused or repurposed. We focus, in this paper, on service-based workflows, in which the steps are implemented by operations that are provided by services. In doing so, we examine two problems that are inherent to workflow reuse, namely search and preservation. The first problem deals with the issue of finding workflows, or fragments thereof, that are of interest to the user. The second problem, on the other hand, deals with the volatility of services. It examines solutions that can be used to curate a workflow specification that utilizes a service that is no longer available. As well as examining the above two problems, we close the paper by discussing future issues that need to be addressed in the field.

The maturity of a new generation of information technologies, including the internet of things (IoT), wearables, cloud computing, Artificial Intelligence (AI) and machine learning, has led to the advent of smart domains, such as smart manufacturing, smart logistics, and smart healthcare. Smart healthcare brings unlimited opportunities to solve many of the problems of traditional medical systems, with the ultimate goal of realizing 4P medicine (Predictive, Preventive, Personalized, Participative). However, to realize this ambitious vision in such a highly regulated multi-disciplinary and sensitive domain, a mine of challenges needs to be effectively and efficiently addressed. A smart health digital platform that integrates all relevant (semi-) structured and unstructured health-related data is fundamental. The platform should incorporate a variety of care data, including vital medical information from medical records, current medication, imaging studies, lifestyle, genetic, demographic, psychological & psychosocial and patient-provided health data from exercise or health monitoring applications and medical pathways. These will lead to improving post-operative planning, reduce medical risks and costs, and generate more accurate therapy and increased Quality of Life (QoL) for patients. The main contribution of this article is a reference architecture for a smart digital platform for personalized prevention and patient management that acts as a roadmap for further R&D in this domain.

For the past twenty years, Service Oriented Computing has changed the way in which information technology was understood. The approach involves not only technological advances that have influenced the development of Software Engineering, such as Service Oriented Architecture, Web services, Service Choreography, or Microservices. In addition, it has also provided the pillars for the development of Cloud Computing, which has transformed how the business in Information and Communication Technology is developed. In that context, this work focuses on how Service Oriented Computing can also drive the integration of humans in the Internet of Things and Crowd Sensing loops by enabling them to act as service providers. The key to this is the deployment of services on mobile devices, in particular smartphones. The enormous penetration of these devices in today’s society, together with the personal nature of the information they handle, open a new horizon for the development of services. Through them individuals are able to make personal information available to others. This paper depicts Human Microservices, an architecture that allows humans to be considered as service providers, and discusses the open challenges in the field that conforms one of the next frontiers for Service Oriented Computing.

We present a vision for the next generation of process technology based on cognitive augmentation. Starting from current process technology, we show how by augmenting layers of cognitive intelligence to combine advances in machine-automation, crowdsourcing and more importantly adaptation and reasoning, we can advance support for emerging requirements of highly changing environments. We believe the challenges lie in the synergy between human and machine, in understanding how to orchestrate and combine their contributions. This vision paper sets forth a roadmap for future research by introducing a framework for cognitive augmentation, identifying the relevant research and technologies, and discussing its application amidst real-world use cases.

In the era of Web of Things and services, context-aware services (CASs) are emerging as an important technology for building innovative smart applications. CASs enable the information integration from both the physical and virtual world, which affects the way human live. However, it is still challenging to build CASs, due to lack of context provisioning management approach and lack of generic approach for formalizing the development process. In this paper, we briefly introduce a large research project, ContextServ, which provides a platform for model-driven development of CASs based on a UML-based modelling language. We discuss the literature and also highlight several future research opportunities for context-aware service research and development.

In the service world in which we live, businesses and organizations are currently moving away from the traditional product-based business model toward the provision of services. The IT sector is by no means indifferent to this tendency and has for some time been providing services rather than software and/or hardware products. In this context, there is a need to improve the quality and productivity of service organizations through the creation of knowledge related to the concept of service ant the specific training of professionals for this sector. This chapter provides an analysis of the knowledge and skills that an IT Engineer should acquire to adapt to the world of services and presents an answer to these needs.

Edge computing is a fundamental enabler for the proliferation of the Internet of Things (IoT). Resources, including compute and storage, are increasingly located at the edge of the network and bridge the gap between the cloud and IoT entities. Edge computing enables low-latency, privacy-awareness, and resilient applications. Many of the applications are in fact business process-based and are distributed over edge as well as cloud resources. Edge devices can be used to analyze high-volume IoT data streams for on-premises monitoring and alerting, and at the same time to aggregate and forward these data to other premises or the cloud for long-term storage and analytics. Many operational and business challenges can be solved by running applications on edge resources or on-premises of Edge-Cloud infrastructure. However, the broad range of IoT application requirements concerning latency, QoS, or fault-tolerance, combined with the heterogeneous and dynamic nature of edge networks, make it particularly challenging to develop, configure, deploy, and operate such applications. In this paper, we discuss some of the research issues with span the domains of business processes engineering and edge computing.

This paper discusses the necessary steps and mechanisms that would allow servitizing the Internet-of-Things (IoT). Servitization exposes functionalities as services allowing potential users to consume these functionalities regardless of the development technologies. To ensure successful use of IoT functionalities, restrictions that hinder this use are identified and then, addressed from a servitization perspective. These restrictions are referred to as no-semantics, silo, and no-reasoning with focus on the first restriction in this paper. Customer-Facilitator-Provider interaction model is developed in the paper allowing to define who does what in the context of IoT servitization.

In the Web of Things (WoT) context, an increasing number of objects provide functions as RESTful services (resources), that can be composed with other existing resources, to create value-added processes (compositions). However, to form a composition, selecting the suitable resources is becoming more challenging, due to: (1) the growing number of resources providing identical functions, which calls for the use of Quality of Resource (QoR) to distinguish between them, and (2) the transient nature of resource availability as a result of objects’ sporadic connectivity in the WoT environments. In this chapter, we present a QoR-driven resource selection approach that forms i-compositions (with i \(\in \mathbb {N}^{*}\)) offering different implementation alternatives. This is done using a selection strategy adaptor that considers QoR constraints and Inputs/Outputs matching of related resources, as well as resource availability and users’ different needs (e.g., optimal compositions having the highest scores, and optimistic compositions having acceptable scores but obtained in more satisfactory delays). Analysis are made to evaluate our resource quality model against existing ones, and experiments are conducted in different environments setups to study the performance of our work.

The Internet of Things incorporates billions of sensors, cameras, RFID and other machines that observe and/or affect the physical world, as well as IoT applications that harvest and analyse IoT data in the cloud, edge computers, and/or the IoT devices themselves. To realise its full potential IoT must ensure the security of the IoT data and related applications that support the IoT-based services and products that are provided to their consumers. Although IoT devices, networks, and computing resources support robust security standards and include related mechanisms that can secure IoT data within the scope of these IoT components, compositions of such point security solutions often fail to ensure the security of IoT data across the IoT ecosystem. In this article we will discuss the main challenges in securing IoT data acquisition, communication, analysis, actuation and illustrate the need for IoT security solutions that are both holistic and lightweight. In addition, we propose a holistic and lightweight IoT security mechanism that via a novel combination of contextualisation with homomorphic encryption prevents harmful outcomes from lack of IoT data security.

The data economy is inevitably rising while data is becoming the ‘new fuel’ driving the world economy as discussed in many studies from a number of viewpoints including economics, technology development, digitisation, and data analytic technologies. Although the significance of data trading is recognised as a critical part of data economy, how data should be described, protected, packaged and priced as data product for trading has not been fully explored. In this chapter we present a service-based approach for data resources to data product transformation and mechanism for data product trading.

Managing data related to natural sciences poses new and challenging problems as it is impossible to represent reality on a one-to-one scale, and imprecision has to be taken into account, both in data memorization and in its processing. Machine learning has been a key enabler in the context of information extraction from natural sciences data. However, data-driven results are strongly affected by the volume, the sparsity and different types of imprecision in the available sources. Therefore, it becomes pivotal to associate both to data and to data-driven services information about their quality, in order to effectively interpret the results. Different levels of granularity and multiple data modalities captured from the same processes could coexist, due to technological constraints or other intrinsic limiting factors. In addition, different levels of granularity might be also the result of application requirements, and outcomes at multiple levels of precision needs to be provided. Affinities of quality issues in domains such as chemistry, biology, and geoinformatics are discussed in the paper.

The growth of social-media platforms has been remarkable in terms of both number of users and volume of content generated. Twitter now reports approximately 166M daily active users who generate in excess of 500M tweets. Such volumes pose major challenges when it comes to providing analytics services and sentiment analyses for specific issues. As citizens tend to freely express their sentiments on social platforms, Twitter has inherently become an indispensable source for the public discourse in a wide variety of topics. Carrying out sentiment analysis on a timely manner on streamed tweets is undoubtedly a demanding endeavor. In this paper, we propose a Spark-based Twitter sentiment analysis software architecture that receives online streamed messages and compiles analytics. We outline the main elements of our proposal and discuss how they collectively help address the challenges involved in this big-data processing task. In particular, our framework: i) exploits the Spark machine-learning library to classify Greek tweets in a timely-manner, ii) manages streamed tweets in synergy with contemporary queuing and in-memory data systems, and iii) determines with high accuracy whether a sentiment is expressed by a genuine account. We report on the findings while experimenting with a novel model for sentiment analysis we created for Greek and ascertain the effectiveness of our proposed architecture.