Modern science relies on workflow technology to capture, process, and analyse data obtained from scientific instruments. Scientific workflows are precise descriptions of experiments in which multiple computational tasks are coordinated based on the dataflows between them. Orchestrating scientific workflows presents a significant research challenge: they are typically executed in a manner such that all data pass through a centralised computer server known as the engine, which causes unnecessary network traffic that leads to a performance bottleneck. These workflows are commonly composed of services that perform computation over geographically distributed resources, and involve the management of dataflows between them. Centralised orchestration is clearly not a scalable approach for coordinating services dispersed across distant geographical locations. This thesis presents a scalable decentralised service-oriented orchestration system that relies on a high-level data coordination language for the specification and execution of workflows. This system's architecture consists of distributed engines, each of which is responsible for executing part of the overall workflow. It exploits parallelism in the workflow by decomposing it into smaller sub-workflows, and determines the most appropriate engines to execute them using computation placement analysis. This permits the workflow logic to be distributed closer to the services providing the data for execution, which reduces the overall data transfer in the workflow and improves its execution time. This thesis provides an evaluation of the presented system which concludes that decentralised orchestration provides scalability benefits over centralised orchestration, and improves the overall performance of executing a service-oriented workflow.

Service-orientation and the underlying concept of service-oriented architectures are a means to successfully address the need for flexibility and interoperability of software applications, which in turn leads to improved IT support of business processes. With a growing level of diffusion, sophistication and maturity, the number of services and interdependencies is gradually rising. This increasingly requires companies to implement a systematic management of services along their entire lifecycle. Service lifecycle management (SLM), i.e., the management of services from the initiating idea to their disposal, is becoming a crucial success factor. Not surprisingly, the academic and practice communities increasingly postulate comprehensive IT support for SLM to counteract the inherent complexity. The topic is still in its infancy, with no comprehensive models available that help evaluating and designing IT support in SLM. This thesis presents a reference architecture for SLM and applies it to the evaluation and designing of SLM IT support in companies. The artifact, which largely resulted from consortium research efforts, draws from an extensive analysis of existing SLM applications, case studies, focus group discussions, bilateral interviews and existing literature. Formal procedure models and a configuration terminology allow adapting and applying the reference architecture to a company’s individual setting. Corresponding usage examples prove its applicability and demonstrate the arising benefits within various SLM IT support design and evaluation tasks. A statistical analysis of the knowledge embodied within the reference data leads to novel, highly significant findings. For example, contemporary standard applications do not yet emphasize the lifecycle concept but rather tend to focus on small parts of the lifecycle, especially on service operation. This forces user companies either into a best-of-breed or a custom-development strategy if they are to implement integrated IT support for their SLM activities. SLM software vendors and internal software development units need to undergo a paradigm shift in order to better reflect the numerous interdependencies and increasing intertwining within services’ lifecycles. The SLM architecture is a first step towards achieving this goal.:Content Overview List of Figures....................................................................................... xi List of Tables ...................................................................................... xiv List of Abbreviations.......................................................................xviii 1 Introduction .................................................................................... 1 2 Foundations ................................................................................... 13 3 Architecture Structure and Strategy Layer .............................. 57 4 Process Layer ................................................................................ 75 5 Information Systems Layer ....................................................... 103 6 Architecture Application and Extension ................................. 137 7 Results, Evaluation and Outlook .............................................. 195 Appendix ..........................................................................................203 References .......................................................................................... 463 Curriculum Vitae.............................................................................. 498 Bibliographic Data............................................................................ 499

Les spécificités du secteur de la construction engendrent un contexte de travail coopératif que nous pouvons qualifier d'incertain par nature. L'activité de chantier se caractérise par un mode de production in situ. Dès lors, nombreux sont les dysfonctionnements qui peuvent apparaître. Nous citerons, par exemple, les problèmes liés à la nature du sol, aux intempéries ou encore ceux qui sont propres aux interactions entre les intervenants (ex. fourniture en matériaux, interfaces entre les corps de métier...). Aussi, la coordination repose sur un mélange subtil entre interactions implicites et explicites, où l'organisation prend de multiples configurations (hiérarchique, adhocratique ou transversale), et où la qualité du processus collectif repose sur l'autonomie et le sens des responsabilités de chacun des intervenants. Nous faisons l'hypothèse qu'un tel contexte est largement fondé sur la notion de confiance, car celle-ci a la capacité de réduire la perception du risque et de permettre l'action dans un environnement marqué par de nombreuses incertitudes. En conséquence, nous suggérons un rapprochement entre les outils d'assistance à la coordination et la notion de confiance, et nous proposons une nouvelle approche du pilotage de l'activité collective à partir de la représentation de la confiance. Ce travail de doctorat se structure autour de la notion de «confiance dans le bon déroulement de l'activité». Nous suggérons que celle-ci est dépendante de chacune des dimensions de l'activité collective : sa progression, les acteurs chargés de son exécution, les ouvrages en résultant (et leur difficulté de mise en œuvre), ainsi que les documents nécessaires à sa réalisation. Aussi, notre méthode consiste d'abord en l'identification des divers critères de confiance, consolidés par une étude de terrain. Puis, nous établissons un modèle mathématique destiné à évaluer cette confiance à partir des informations issues d'un contexte de coopération. Sur base de ces éléments, s'en suit la proposition d'un prototype dénommé Bat'iTrust, reposant sur une architecture logicielle multi-vues et orientée services. Pour l'utilisateur, la navigation au sein de ce prototype est guidée par un tableau de bord centré sur le concept de confiance. Enfin, la validation de cette proposition repose sur des enquêtes et une phase d'expérimentation qui nous ont permis de confronter nos résultats à des sujets expérimentaux représentatifs du domaine.

Data exchange between different information systems is a complex issue. Each system, designed for a specific purpose, is defined using a vocabulary of the specific business. While Web services allow interoperations and data communications between multiple systems, the clients of the services must understand the vocabulary of the targeting data resources to select services or to construct queries. In this thesis we explore an ontology-based approach to facilitate clients’ queries in the vocabulary of the clients’ own domain, and to automate the query processing. A governmental inter-department data query process has been used to illustrate the capability of the semantic approach.

This thesis research was funded by the Southeast National Marine Renewable Energy Center (SNMREC) at Florida Atlantic University. Its objective is the development of Quality of Service (QoS) mechanisms for the wireless communications architecture used by the Prognosis and Health Monitoring (PHM) subsystem. There are numerous technical challenges that the PHM Communications Subsystem tries to solve. Due to ocean platform mobility from waves, currents, and other environmental factors, signal quality can vary significantly. As a result, the wireless link between the electric generator platform and shore systems will have variable quality in terms of data rate, delay, and availability. In addition, the data traffic that flows from generator sensors and PHM applications to the shore systems consists of numerous types of messages that have different QoS demands (e.g. delay) and priority that depends on the message type, user ID, sensor location, and application-dependent parameters. The PHM Communications subsystem must handle effectively high priority messages, such as alarms, alerts, and remote control commands from shore systems. It also performs QoS in the application layer, so it can read the contents of every message to prioritize them. In order to perform QoS in the application layer the PHM subsystem relies on Java Servlet multithreaded technology and different queuing techniques to control message transmission order. Furthermore, it compresses all traffic that comes from the ocean-based electric generator/turbine platform to reduce the load on the wireless link. The PHM Communications subsystem consists of three components: the wireless link, the Link Manager, and the Web Services Network Proxy. We present experimental results for the Web Services Network Proxy and demonstrate the effectiveness of XML data compression and semantic-based message scheduling over a link with variable capacity. / by Timur Tavtilov. / Thesis (M.S.C.S.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

Os mercados estão se tornando independentes de barreiras geográficas e as indústrias têm procurado novas configurações de sistemas produtivos (SPs), passando de estruturas centralizadas para estruturas distribuídas, deslocando suas plantas produtivas para países com reservas de energia e baixos custos operacionais. Para permitir a coordenação e gerenciamento deste tipo de SP disperso, aproveita-se dos avanços das tecnologias mecatrônicas e de informação, as quais permitem uma maior cooperação entre as partes do sistema e entre os atores (clientes, operadores, administradores, etc.) envolvidos. Cada parte do SP disperso que também é um SP tem seu grau de autonomia operacional. Esse tipo de sistema apresenta novos problemas de integração e coordenação de componentes, que têm que ser superados para se chegar a uma efetiva implementação. A falta de dados de testes já realizados com estruturas distribuídas dificulta o desenvolvimento prático de SPs dispersos. Este trabalho inicialmente adota uma arquitetura de controle para a negociação entre usuários de um SP disperso. Para a implementação da arquitetura foram desenvolvidos modelos computacionais explorando o potencial da rede de Petri e do PFS (production flow schema) para sistematizar a construção dos modelos. Pela análise dos modelos com base nas propriedades da rede de Petri avaliou-se a arquitetura de controle e estabeleceu-se as especificações que foram adotadas para sua implementação prática. A implementação e os testes foram realizados considerando os subsistemas autônomos de um sistema flexível de montagem que emula um SP disperso. Os estudos, análises e testes realizados foram fundamentais para adquirir maior experiência prática relacionado a concepção, projeto, implementação e operação de arquiteturas de controle distribuído aplicadas a SPs dispersos. / The markets are becoming independent of geographic and industry have sought new configurations of productive systems, from centralized structures to distributed structures, shifting their production plants to countries with energy reserves and low operating costs. To allow the coordination and management of this type of dispersed productive system, takes advantage of advances in mechatronics and information technologies, which allow greater cooperation between parts of the system and among stakeholders (customers, operators, administrators, etc.) involved. Each part of disperse productive system, that is also a productive system, has its own level of operational autonomy. This type of system presents new problems of integration and coordination of components that must be overcome to achieve effective implementation. The lack of data from tests already carried out with distributed structures hinders the practical development of disperse productive systems. This work initially adopts a control architecture for negotiation between users of a disperse productive system. For the implementation of the architecture were developed computational models exploring the potential of Petri networks (PN) and the production flow schema (PFS) to systematize the construction of models. For the analysis of models based on the PN properties we evaluated the control architecture and established the specifications that were used for its practical implementation. The implementation and the tests were performed considering the autonomous subsystems of a flexible mounting system that emulates a disperse productive system. The studies, analysis and tests were essential to acquire more practical experience related to conception, design, implementation and operation of distributed control architectures applied to disperse productive systems.

A evolução dos processos de negócio para uma visão de serviços alavancou um novo modelo computacional, o modelo orientado a serviços. Nesse modelo, os processos de negócio são modelados e implementados sob a ótica de serviços. O governo mostra-se como um domínio potencial de implantação de soluções orientadas a serviços. Embora as organizações governamentais estejam adotando o uso de serviços a fim de alcançar a interoperabilidade de sistemas de informação de governo, os serviços são geralmente criados a partir dos princípios elementares, sem considerar o reuso de soluções orientadas a serviços concebidas por outras entidades públicas. Assim, esta pesquisa tem como propósito fornecer diretrizes para auxiliar a especificação de serviços de governo eletrônico baseadas em padrões de serviços, para subsidiar o desenvolvimento de sistemas de governo, alinhados aos benefícios de computação orientada a serviços e o reuso de soluções para o governo eletrônico. Esta tese apresenta o MESe-gov, um modelo para especificação de serviços de governo eletrônico e o DESe-gov, um conjunto de diretrizes para especificação de serviços de governo eletrônico. Também é proposto um ciclo de vida de serviço para a especificação de novos serviços a partir dos padrões de serviços. A concepção de serviços, aliada ao conceito de padrões de serviços, ajuda engenheiros de software identificar elementos funcionais recorrentes e reduzir a redundância de esforços para a concepção de serviços com propósitos similares. Nesta pesquisa, foram realizados estudos de casos em que foram aplicadas as diretrizes a partir dos serviços existentes na área financeira do governo. Como resultado, os estudos de casos mostram que as diretrizes auxiliam a especificação de padrões de serviços. / The evolution of business processes for an insight on services has boosted a new computational model, the service-oriented model. The business processes in the service-oriented computational model are modelled and implemented from the perspective of services. The government appears to be a high potential scenario for the deployment of service-oriented applications. Although government organizations are adopting the use of services in order to achieve interoperability of government systems, those services are usually created from basic principles, without considering the reuse of service-oriented solutions adopted by other public entities. Thus, this research aims to provide guidelines to assist the specification of e-government services based on service patterns, to support the development of government systems, aligned with the benefits of service-oriented computing and the reuse of solutions for e-government. This thesis presents the MESe-gov, a model for the services specification of electronic government and the DESe-gov, a set of guidelines for the services specification of electronic government. A service lifecycle is also proposed for the specification of new services from service patterns. The conception of services combined with the concept of service patterns can help software engineers to identify recurrent functional elements and reduce redundant efforts in the conception of services with the same purposes. In this research, case studies were conducted in which the guidelines from existing services in the governments financial area were implemented. As a result, the case studies show that the guidelines help to specify service patterns.

O número de dispositivos conectados e presentes na vida dos usuários é expressivo e a realização de atividades utilizando múltiplos dispositivos já é uma realidade. Entretanto, o uso integrado e sinérgico desses recursos, a que se denomina interação distribuída, tem encontrado empecilhos nas limitações dos sistemas em prover e atender interações dos usuários que atravessem as fronteiras do dispositivo. A diversidade de plataformas, bem como das áreas de pesquisa envolvidas são também forças contrárias à interação distribuída. O objetivo desta pesquisa, portanto, está em definir uma arquitetura que facilite a construção de aplicações observando aspectos de uma interação distribuída, na qual usuário e dispositivo não se acoplem unicamente no modelo de interação 1 usuário:1 dispositivo. Definiu-se uma arquitetura orientada a serviço, após pesquisa exploratória e sistemática acerca de frameworks e middlewares que promovem a integração entre dispositivos, e verificou-se sua viabilidade por provas de conceito. Como resultados, obteve-se um padrão de transições necessários para a construção de diversos cenários de uso e uma arquitetura orientada a serviços P2P dirigida a eventos que permite a implementação e execução desses cenários. / The number of connected devices and present in the lives of users is currently expressive and performing activities using multiple devices is already a reality. However, the integrated and synergistic use of these resources, which is called distributed interaction, has found obstacles in the limitations of current systems to provide and attend user interactions that cross the boundaries of the device. A variety of platforms, also involved research areas are contrary forces to the distributed interaction too. The aim of this research is therefore to define an architecture that facilitates the building of applications observing aspects of a distributed interaction, where user and device does not only bind in the interaction model 1 user:1 device. A service-oriented architecture has been defined, after exploratory and systematic research about frameworks and middlewares that promote integration among devices, and its feasibility has been verified by proofs of concept. As a result, we have obtained a pattern of transitions necessary for the construction of various usage scenarios and a P2P event-driven service-oriented architecture that allows implementation and execution of these scenarios.

L'informatique pervasive (ou ubiquitaire) est un support pour des environnements contenant denombreux objets (ou dispositifs) disséminés, équipés d'électronique et interconnectés. Ces dispositifsfournissent un accès distant à une multitude de fonctionnalités qui nous aident dans notre vie quotidienne.Les Architectures Orientées Services sont adaptées à la conception de logiciels pervasifs. En effet,chaque dispositif fournit son propre ensemble de fonctionnalités sous la forme de services. Ainsi, enl'absence de mécanisme complémentaire, les utilisateurs se trouvent limités à utiliser les servicesisolément alors que leurs besoins correspondent à des scénarios qui impliquent une composition demultiples services offerts par plusieurs appareils.Dans cette thèse, nous défendons qu'un système pervasif doit : d'une part, permettre aux utilisateursd'exprimer facilement leurs besoins en créant des scénarios et d'autre part, proposer à ses utilisateursune représentation et des moyens de gestion de leur contexte afin qu'ils puissent tirer le meilleur parti deleur environnement et de ses changements. De plus, la présence de plusieurs utilisateurs implique lanécessité de collaborer. Par ailleurs, l'exécution de scénarios doit être résiliente aux changementsenvironnementaux et aux actions des utilisateurs. Elle doit ainsi s'adapter dynamiquement et, si possible,tirer profit du contexte et des changements de l'environnement.Notre contribution, nommée SaS (Scenarios as Services), répond à ces objectifs. Elle propose uneapproche interopérable capable de s'adapter à l'environnement. Elle fournit une représentation persistanteet personnalisable du contexte et inclut un langage de description de scénarios destiné aux utilisateurs.Ces scénarios sont facilement contrôlables, personnalisables et réutilisables. Elle planifie l'exécution pasà pas des scénarios, afin de s'adapter aux changements de l'environnement et de bénéficier desavantages de la mobilité des utilisateurs (exécution d'un scénario, dans la durée, sur plusieurs lieux).Enfin, elle inclut le partage de scénarios qui permet aux utilisateurs de collaborer. Un prototype de SaS,basé sur des normes industrielles (telle qu'OSGi), prouve la faisabilité de notre contribution et nouspermet de l'évaluer sur un cas d'étude simple. / Pervasive (or ubiquitous) computing is a paradigm for environments containing distributedinterconnected devices that embed electronics. These devices provide a remote access to numerousfunctionalities that assist us in our daily life. Service-Oriented Architectures are suitable to design softwarefor pervasive environments. Indeed, each device provides its own set of functionalities as services.Without any extra mechanism, users can only use a single service at a time. Nevertheless, their needsusually correspond to scenarios which involve a composition of multiple services, provided by multipledevices.In this thesis, we advocate that a pervasive system must, on the one hand, enable users to easily expresstheir needs through scenario creation and, on the other hand, propose to users a representation of theircontext so that they can benefit from both their environment and its changes. In addition, the presence ofseveral users implies that users must be able to collaborate.Our contribution, named SaS (Scenarios as Services), fulfils these requirements. It proposes aninteroperable approach that adapts to its environment. It provides users with a customizable andpersistent representation of their context and includes a scenario description language targeted to users.Scenarios are easy to control, customize and reuse. SaS schedules the step-by-step execution ofscenarios to adapt to environmental changes and benefit from user mobility (scenario execution split over time on successive distinct sites). Finally, SaS includes scenario sharing mechanisms which are abasis for collaboration. A prototype of SaS, based on industrial standards (e.g., OSGi), proves thefeasibility of our contribution and serves for its evaluation on a simple use case.

Linha de produtos de software (LPS) corresponde a uma das mais bem sucedidas formas de reúso, pois permite a reutilização de requisitos e arquitetura. Embora o desenvolvimento, manutenção e evolução de uma LPS ainda possua um custo alto quando comparado ao desenvolvimento de sistemas únicos (single systems), um lucro significativo pode ser obtido com a venda de diversos produtos derivados da LPS. No projeto de uma LPS analisa-se os sistemas coletivamente, ou seja, o domínio. Geradores de aplicações são ferramentas capazes de gerar artefatos a partir de uma especificação, e no caso de se ter a especificação de um domínio, é possível gerar aplicações para esse domínio. Web services representam uma tecnologia promissora para disponibilização de serviços na Web e desenvolvimento de software com arquitetura flexível e de fácil manutenção. Neste trabalho é proposta uma abordagem de desenvolvimento de linha de produtos com arquitetura orientada a serviços, na qual a geração de produtos é apoiada por um gerador de aplicações. A abordagem chama-se SoProL-WS e possui o objetivo de reduzir os custos e prazos de desenvolvimento da LPS e facilitar a sua manutenção, evolução e derivação de seus membros. SoProLWS apresenta as atividades e artefatos necessários para partir dos requisitos da LPS, projetar, implementar, configurar um gerador de aplicações e gerar seus membros a partir do gerador ou por meio de uma configuração manual. Além disso, é apresentado um estudo de caso com o desenvolvimento de uma linha de produtos de leilões Web seguindo os passos da abordagem, bem como são discutidas as alternativas de projeto relevantes para esse tipo de desenvolvimento / Software product lines (SPL) is a successful reuse technique that fosters requirements and architecture reuse. Although SPL costs with development, maintenance and evolution increases when compared to single system development, significant profit can be obtained by selling many products derived from SPL. In a SPL project, systems are analised collectively like a domain. Application generators are tools capable of generating artifacts based on an specification, in case of a domain specification, it is possible to generate applications in this domain. Web services represent a technology to make services available over a network and to develop a flexible and adaptable software architecture. This work presents an approach, called SoProL-WS, to develop SPL using service oriented architecture, where product derivation is supported by an applicator generator. The aim of this approach is to enhance flexibility, reuse and consequently decrease SPL development costs. In addition, this work presents a case study where SoProL-WS is applied to develop a Web auctions SPL