A TV digital é hoje uma realidade em muitos países e no Brasil as pesquisas já estão atingindo um estágio avançando. Muita expectativa gira em torno desta tecnologia, que oferece uma ampla variedade de serviços multimídia, abrindo as portas deste modo para criativas propostas e novas possibilidades de relacionamento com o usuário final. Esta complexidade e as novas opções de serviços que podem ser oferecidos demandam, obrigatoriamente, uma arquitetura adequada de gerenciamento, capaz de controlar e otimizar os terminais desta rede bem como os serviços oferecidos pela mesma. Neste trabalho são apresentados os requisitos funcionais e não funcionais necessários para a construção de um sistema genérico de gerenciamento de set-top boxes e serviços em TV digital, baseados no estudo de pesquisas semelhantes e nas particularidades deste ambiente. Também para este sistema genérico são detalhados todos os casos de uso de interesse. Além disso, uma arquitetura híbrida é proposta para a solução deste problema, utilizando para tanto padrões abertos e considerando alguns cenários possíveis de implantação no mundo real. Nesta arquitetura, é adotado o modelo clássico de gerenciamento gerente-agente, através do uso de um servidor de gerência, responsável pelo armazenamento e análise de todos os dados de gerenciamento bem como pelo disparo de comandos e recebimento de respostas (síncronas e/ou assíncronas) dos agentes. Tanto na definição deste sistema genérico como no desenho da arquitetura final, procurou-se na medida do possível adotar a utilização de princípios clássicos de gerenciamento que já são aplicados em várias arquiteturas existentes, inclusive de mercado. Deste modo, foi possível encontrar uma solução para um problema complexo através do uso de conceitos amplamente conhecidos, o que facilita o entendimento final. / Nowadays, digital TV systems are a reality in many countries and in Brazil the research in this field is reaching an advanced stage. Much excitement is expected from this technology, which may offer a wide variety of multimedia services, opening the doors for creative proposals and new possibilities of relationships with the final user. This complexity and the new options of services that can be offered demand a suitable architecture for management, which should be able to control and optimize the set-top boxes of this network as well as of the services offered by it. In this work, the requirements for the building of a generic management system for the digital TV environment are presented, based in the study of similar research and considering the specifics involved. Besides that, a hybrid architecture based on open standards is proposed for the solution of this problem. In this architecture, it is adopted the classical manager-agent model, through the use of a management server, which is responsible for the storage and analysis of all the management data as well for the triggering of commands and receptions of answers and traps sent by the agents. The conception and design of this architecture has followed as much as possible, classic principles of network management. With this approach, it was possible to find a simple solution to a very complex problem.

[pt] Middlewares têm sido utilizados para desenvolver aplicações distribuídas em domínios muito diferentes. Cada um desses domínios possui pré-requisitos diversos relacionados à comunicação entre as partes do ambiente distribuído. Muitas vezes, os pré-requisitos de diferentes domínios são contraditórios, fazendo com que a implementação de um middleware que os atenda em mais de um domínio de aplicação se torne muito complexa. Uma das opções para resolver esse problema é criar o middleware de forma que ele possa ser adaptado de acordo com as necessidades de cada aplicação. Neste trabalho, estudamos algumas das técnicas disponíveis para a construção de middlewares adaptáveis, indo desde mudanças simples na política de transmissão de dados entre os nós de um sistema distribuído, até adaptações mais abrangentes na implementação do sistema para permitir diferentes contextos de execução. Como estudo de caso, propomos um modelo de componentização para um middleware já existente, implementado em Lua, com a finalidade de capacitá-lo a acomodar adaptações de sua implementação de uma forma simples e estruturada. Também realizamos alguns experimentos para avaliar a aplicabilidade e eficácia do novo middleware desenvolvido. / [en] Middlewares have been used in order to develop distributed applications in a variety of domains. Each of these domains has different requirements related to the communication among parts of the distributed system. Many times the requirements of different domains are contradictory, turning the implementation of a middleware are that fulfills them in more than one the application domain in into a very complex problem. One of the options to solve this issue is to create the middleware are in a way that it can be adapted accordingly to the needs of each application. In this study we survey some of the available techniques for the construction of adaptive middleware, ranging from simple changes in the transmission of data between nodes in a distributed system to more thorough adaptations in a system in order to allow for diferent application execution contexts. As a case study we propose a component model for an existing middleware, implemented in Lua, with the goal of enabling the middleware are to accommodate adaptations in its implementation in a simple and structured manner. We also created experiments in order to evaluate the applicability and effectiveness of the middleware.

Magistrinio darbo projektinėje dalyje buvo sukurta nuotolinė žinių testavimo sistema „Cool Test Tool“. Ji realizuota remiantis kliento – serverio architektūra. Projektavimo metu buvo pasirinkta „Remote Objects“ technologija su TCP protokolu. Todėl šio darbo tikslas – išanalizuoti CORBA, Remote Objects ir Web Services veikimo principus, ištirti jų resursų išnaudojimą ir išrinkti efektyviausią technologiją, tinkančią projektui. Šios technologijos pasirinktos todėl, kad jas paprasta pritaikyti esamam projektui. / Many of today’s enterprise computing systems are powered by distributed object middleware. Such systems, which are common in industries such as telecommunications, finance, manufacturing, and government, often support applications that are critical to particular business operations. Because of this, distributed object middleware is often held to stringent performance, reliability, and availability requirements. Fortunately, modern approaches have no problem meeting or exceeding these requirements. The goal of this research was to familiarize with middleware technology especially with CORBA, Microsoft .NET Remoting and Microsoft .NET Web Services. In experimental part was proved that the most efficient technology is Microsoft .NET Remoting with TCP protocol. That means our choice was correct during development of our project “Cool Test Tool”.

Scientific simulations running on High End Computing machines in domains like Fusion, Astrophysics, and Combustion now routinely generate terabytes of data in a single run, and these data volumes are only expected to increase. Since such massive simulation outputs are key to scientific discovery, the ability to rapidly store, move, analyze, and visualize data is critical to scientists' productivity. Yet there are already serious I/O bottlenecks on current supercomputers, and movement toward the Exascale is further accelerating this trend. This dissertation is concerned with the design, implementation, and evaluation of middleware-level solutions to enable high performance and resource efficient online data analytics to process massive simulation output data at large scales. Online data analytics can effectively overcome the I/O bottleneck for scientific applications at large scales by processing data as it moves through the I/O path. Online analytics can extract valuable insights from live simulation output in a timely manner, better prepare data for subsequent deep analysis and visualization, and gain improved performance and reduced data movement cost (both in time and in power) compared to the conventional post-processing paradigm. The thesis identifies the key challenges for online data analytics based on the needs of a variety of large-scale scientific applications, and proposes a set of novel and effective approaches to efficiently program, distribute, and schedule online data analytics along the critical I/O path. In particular, its solution approach i) provides a high performance data movement substrate to support parallel and complex data exchanges between simulation and online data analytics, ii) enables placement flexibility of analytics to exploit distributed resources, iii) for co-placement of analytics with simulation codes on the same nodes, it uses fined-grained scheduling to harvest idle resources for running online analytics with minimal interference to the simulation, and finally, iv) it supports scalable efficient online spatial indices to accelerate data analytics and visualization on the deep memory hierarchies of high end machines. Our middleware approach is evaluated with leadership scientific applications in domains like Fusion, Combustion, and Molecular Dynamics, and on different High End Computing platforms. Substantial improvements are demonstrated in end-to-end application performance and in resource efficiency at scales of up to 16384 cores, for a broad range of analytics and visualization codes. The outcome is a useful and effective software platform for online scientific data analytics facilitating large-scale scientific data exploration.

Mobile computing devices like handhelds are becoming ubiquitous and so is computing embedded in cyber-physical systems like cameras, smart sensors, vehicles, and many others. Further, the computation and communication resources present in these settings are becoming increasingly powerful. The resulting, rich execution platforms are enabling increasingly complex applications and system uses. These trends enable richer execution platforms for running ever more complex distributed applications. This thesis explores these opportunities (i) for cooperative mobile platforms, where the combined resources of multiple computing devices and the sensors attached to them can be shared to better address certain application needs, and (ii) for distributed platforms where opportunities for cooperation are further strengthened by virtualization. The latter offers efficient abstractions for device sharing and application migration that enable applications to operate across dynamically changing and heterogeneous systems without their explicit involvement. An important property of cooperative distributed platforms is that they jointly and cooperatively provide and maintain the collective resources needed by applications. Another property is that these platforms make decisions about the resources allocated to certain tasks in a decentralized fashion. In contrast to volunteer computing systems, however, cooperation implies the commitment of resources as well as the commitment to jointly managing them. The resulting technical challenges for the mobile environments on which this thesis is focused include coping with dynamic network topology, the runtime addition and removal of devices, and resource management issues that go beyond resource usage and scheduling to also include topics like energy consumption and battery drain. Platform and resource virtualization can provide important benefits to cooperative mobile platforms, the key one being the ability to hide from operating systems and applications the complexities implied by collective resource usage. To realize this opportunity, this thesis extends current techniques for device access and sharing in virtualized systems, particularly to improve their flexibility in terms of their ability to make the implementation choices needed for efficient service provision and realization in the mobile and embedded systems targeted by this work. Specifically, we use middleware-based approaches to flexibly extend device and service implementations across cooperative and virtualized mobile platforms. First, for cooperating platforms, application-specific overlay networks are constructed and managed in response to dynamics at the application level and in the underlying infrastructure. When virtualizing these platforms, these same middleware techniques are shown capable of providing uniform services to applications despite platform heterogeneity and dynamics. The approach is shown useful for sharing and remotely accessing devices and services, and for device emulation in mobile settings.

As cidades estão a seu tempo e a seu modo, modernizando os serviços prestados à população. Entre os diversos fatores que estão contribuindo para esta evolução estão a diversificação e proliferação de sensores, nos diversos domínios de serviços das cidades, e os novos canais de comunicação com os munícipes, entre eles, as redes sociais e mais recentemente os sistemas crowdsensing, motivados pelos anseios sociais, por melhores serviços públicos e pela popularização dos dispositivos móveis. Nesta direção, a eficiência administrativa é um fator essencial, uma vez que as cidades estão se mostrando mais complexas na medida em que cresce a população nas áreas urbanas. A utilização de técnicas de sistemas distribuídos para que múltiplos domínios de serviços usufruam da mesma infraestrutura computacional, pode auxiliar na eficiência das cidades, evitando gastos administrativos duplicados e até mesmo, possibilitando a correlação de eventos entre os serviços, favorecendo a identificação de fatores de causalidades e assim, a tomada de decisões administrativas mais objetivas e precisas. Neste contexto, este trabalho concentra-se na análise de um middleware direcionado à gestão de cidades para coleta, integração e interpretação dos dados de sensores, pertencentes aos serviços disponíveis da própria cidade, junto com os dados do sensoriamento colaborado pelos cidadãos. Para avaliação do conceito foi investigado o cenário de monitoração da conservação de vias públicas. Após 3 meses de coletas de dados por um sistema de sensoriamento automático, totalizando mais de 360 mil pontos e também mais de 90 relatórios pelo sensoriamento participativo, verificou-se que um sistema distribuído pode realizar a interpretação de séries históricas, engajar os munícipes apoiar a manutenção dos serviços da cidade e também indicar objetivamente aos gestores públicos os pontos que devem ser prioritariamente atendidos. Aliar ferramentas pelas quais o cidadão pode, de acordo com sua necessidade, convicção e altruísmo, exercer influência nos gestores públicos com o suporte de informação contínua e critérios objetivos das redes de sensores, pode estimular a continua excelência dos serviços públicos. / The cities in their own way and time are improving their services provided to the population. Among several factors that are contributing to this trend are the diversification and proliferation of sensors in various services domains of cities and new communication ways with citizens, for instance, social networks and more recently, crowdsensing systems, motivated by social expectations for better public services and the popularity of mobile devices. In this direction, administrative efficiency is a key factor, since the cities are proving more complex with increasing the population in urban areas. Techniques of distributed systems to share the same computing infrastructure to multiple service domains, can assist in the efficiency of cities, avoiding duplicate administrative costs and even allowing event correlation between services, providing the identification of causality factors, thus making management decisions more objective and accurate. In this context, this research focuses on analysis of a middleware directed to city management for collection, integration and interpretation of sensors data, present in city services, along with the sensing data contributed by citizens. For concept evaluation, was investigated the scenario of conservation of public streets. After 3 months of data collection by an automatic sensing system comprising more than 360 thousand points and also 94 reports of collaborative sensing, it was found that a distributed system can perform the interpretation of historical series; engage the citizens to support maintenance of city services and indicate objectively the points that should primarily be fix by public managers. Combining tools, which citizens can, according to their need, conviction, altruism, exert their own influence in public management and the continuous information support to objective criteria of sensor networks, can stimulate the continued excellence of public services.

[pt] Sistemas distribuídos frequentemente apresentam um comportamento em tempo de execução diferente do esperado pelo programador. A análise estática, somente, não suficiente para a compreensão do comportamento e para o diagnóstico de problemas nesses sistemas, em razão da sua natureza nao determinística, reflexo de características inerentes como concorrência, latência na comunicação e falha parcial. Sendo assim, torna-se necessário um melhor entendimento das interações entre os diferentes componentes de software que formam o sistema, para que o desenvolvedor possa ter uma melhor visão do comportamento do sistema durante sua execução. Neste trabalho, apresentamos uma ferramenta que faz a reconstrução das interações entre os componentes de uma aplicação distribuída, oferecendo uma visão das linhas de execução distribuídas e permitindo o acompanhamento das sequências de chamadas remotas e a análise das relações de causalidade. Essa ferramenta também faz a persistência do histórico dessas interações ao longo do tempo, correlacionando-as a arquitetura do sistema e aos dados de desempenho. Assim, a ferramenta proposta auxilia o desenvolvedor a melhor compreender cenários que envolvem comportamentos indevido do sistema e a restringir o escopo da análise do erro, facilitando a busca de uma solução. / [en] Distributed systems often present a runtime behavior different than what is expected by the programmer. Static analysis is not enough to understand the runtime behavior and to diagnoses errors. This difficulty is caused by the non-deterministic nature of distributed systems, because of their inherent characteristics, such as concurrency, communication latency and partial failure. Therefore, it’s necessary a better view of the interactions between the system’s software components in order to understand its runtime behavior. In this work we present a tool that rebuilds the interactions among distributed components, presents a view of distributed threads and remote call sequences, and allows the analysis of causality relationships. Our tool also stores the interactions over time and correlates them to the system architecture and to performance data. The proposed tool helps the developer to better understand scenarios involving an unexpected behavior of the system and to restrict the scope of error analysis, making easier the search for a solution.

[pt] Os avanços nas tecnologias de redes de computadores, telecomunicações, e dispositivos móveis portáteis têm gerado uma demanda por aplicações e serviços que sejam apropriados para ambientes com conectividade intermitente e mobilidade dos dispositivos. Uma questão central para a viabilização de tais aplicações em ambientes como estes são as possíveis soluções para o Gerenciamento de Mobilidade, isto é, a manutenção automática da conectividade entre componentes de um sistema em cenários nos quais os dispositivos mudam de endereço IP dinamicamente, à medida que se reconectam a diferentes domínios de rede. Partindo de uma investigação inicial das principais soluções existentes, este trabalho apresenta a implementação de uma solução na camada de aplicação baseada no protocolo SIP. Em seguida, apresenta a adaptação de um sistema publish/subscribe existente para fazer com que o mesmo utilize a solução implementada, com o objetivo de prover suporte à mobilidade e à desconexão dos produtores e consumidores de eventos, bem como para a travessia de firewalls e NATs, permitindo que o sistema publish/subscribe possa ser usado na Internet, e não somente em redes locais. O modelo publish/subscribe foi escolhido como estudo de caso, pois oferece interações assíncronas e anônimas que se adéquam muito bem aos requisitos de um cenário de mobilidade. A adaptação deste sistema partiu da investigação de alguns dos principais requisitos que um sistema pub/sub deve atender para tratar mobilidade e desconexão de dispositivos. O referido sistema foi testado para diferentes cenários, e seu desempenho foi avaliado para diferentes configurações, e comparado ao do sistema publish/subscribe convencional. / [en] Advances in computer networks, telecommunications, and portable mobile devices have increased the demand for applications and services that are suitable for environments with intermittent connectivity and mobility of devices. A central question for the viability of such applications on environments like these are the possible solutions for Mobility Management, i.e., the automatic maintenance of the connectivity between system components in scenarios where the devices change their IP addresses dynamically as they reconnect to different network domains. Starting from an initial investigation of the main solutions for Mobility Management, this dissertation presents the development of a solution on the application layer based on the SIP protocol. Then, it presents the adaptation of an existing publish/subscribe system to make it use the developed Mobility Management solution, in order to provide support for the mobility and disconnection of event producers and consumers, and also NAT and firewall traversal, enabling the system to be used on the Internet, and not only on local networks. The publish/subscribe model has been chosen as a case study for the implemented solution, because it provides asynchronous and anonymous interactions that are very well suited to the requirements of a mobility scenario. The adaptation of the publish/subscribe system started from an investigation of the main requirements a system like this should meet in order to support mobility and disconnection of devices. The referred system has been tested for different scenarios, and its performance has been evaluated for different configurations, and it has been compared to the conventional publish/subscribe system.

[pt] Aplicações que usam CORBA como plataforma de comunicação geralmente possuem restrições ao serem executadas em ambientes compostos por mais de um domínio administrativo. Este fato é particularmente verdade quando as aplicações precisam atravessar firewalls/NAT. Além do mais, não existe atualmente uma solução padronizada e adotada por todos os ORBs, obrigando as aplicações que utilizam este enfatizar{middleware} a adotarem soluções proprietárias que muitas vezes não são adequadas ao ambiente em que as aplicações funcionam (e.g. impossibilidade de abertura de portas no firewall). Este trabalho apresenta e avalia três soluções para a travessia de firewall/NAT por aplicações distribuídas que utilizam CORBA como camada de comunicação, cada uma explorando as vantagens de uma situação específica. Exemplos de tais situações são a possibilidade de configuração do firewall ou a possibilidade de abertura de conexões TCP para fora da rede. / [en] Applications that use CORBA as the communication layer often face some restrictions for multi-domain deployment. This is particularly true when they have to face firewall/NAT traversal. Furthermore, nowadays there is no well-accepted unique or standardized solution adopted by all ORBs, compelling applications using this type of middleware to use proprietary solutions that sometimes do not address the environment restrictions in which they are deployed (e.g. impossibility to open firewall ports). This work presents and compares three solutions for firewall/NAT traversal by CORBA-based distributed applications, each one suitable for a specific situation and exploring its advantages. Examples of such situations are the possibility of open firewall ports or the possibility of start a TCP connection to the outside network.

De nos jours, l'internet des objets (IoT) connaît un intérêt considérable tant de la part du milieu universitaire que de l'industrie. Il a contribué à améliorer la qualité de vie, la croissance des entreprises et l'efficacité dans de multiples domaines. Cependant, l'hétérogénéité des objets qui peuvent être connectés dans de tels environnements, rend difficile leur interopérabilité. En outre, les observations produites par ces objets sont générées avec différents vocabulaires et formats de données. Cette hétérogénéité de technologies dans le monde IoT rend nécessaire l'adoption de solutions génériques à l'échelle mondiale. De plus, elle rend difficile le partage et la réutilisation des données dans d'autres buts que ceux pour lesquels elles ont été initialement mises en place. Dans cette thèse, nous abordons ces défis dans le contexte des applications de santé. Pour cela, nous proposons de transformer les données brutes issues de capteurs en connaissances et en informations en s'appuyant sur les ontologies. Ces connaissances vont être partagées entre les différents composants du système IoT. En ce qui concerne les défis d'hétérogénéité et d'interopérabilité, notre contribution principale est une architecture IoT utilisant des ontologies pour permettre le déploiement d'applications IoT sémantiques. Cette approche permet de partager les observations des capteurs, la contextualisation des données et la réutilisation des connaissances et des informations traitées. Les contributions spécifiques comprennent : * Conception d'une ontologie " Cognitive Semantic Sensor Network ontology (CoSSN) " : Cette ontologie vise à surmonter les défis d'interopérabilité sémantiques introduits par la variété des capteurs potentiellement utilisés. CoSSN permet aussi de modéliser la représentation des connaissances des experts. * Conception et mise en œuvre de SeMoM: SeMoM est une architecture flexible pour l'IoT intégrant l'ontologie CoSSN. Elle s'appuie sur un middleware orienté message (MoM) pour offrir une solution à couplage faible entre les composants du système. Ceux-ci peuvent échanger des données d'observation sémantiques de manière flexible à l'aide du paradigme producteur/consommateur. Du point de vue applicatif, nous sommes intéressés aux applications de santé. Dans ce domaine, les approches spécifiques et les prototypes individuels sont des solutions prédominantes ce qui rend difficile la collaboration entre différentes applications, en particulier dans un cas de patients multi-pathologies. En ce qui concerne ces défis, nous nous sommes intéressés à deux études de cas: 1) la détection du risque de développement des escarres chez les personnes âgées et 2) la détection des activités de la vie quotidienne (ADL) de personnes pour le suivi et l'assistance à domicile : * Nous avons développé des extensions de CoSSN pour décrire chaque concept en lien avec les deux cas d'utilisation. Nous avons également développé des applications spécifiques grâce à SeMoM qui mettent en œuvre des règles de connaissances expertes permettant d'évaluer et de détecter les escarres et les activités. * Nous avons mis en œuvre et évaluer le framework SeMoM en se basant sur deux expérimentations. La première basée sur le déploiement d'un système ciblant la détection des activités ADL dans un laboratoire d'expérimentation pour la santé (le Connected Health Lab). La seconde est basée sur le simulateur d'activités ADLSim développé par l'Université d'Oslo. Ce simulateur a été utilisé pour effectuer des tests de performances de notre solution en générant une quantité massive de données sur les activités d'une personne à domicile. / Nowadays, the adoption of the Internet of Things (IoT) has received a considerable interest from both academia and industry. It provides enhancements in quality of life, business growth and efficiency in multiple domains. However, the heterogeneity of the "Things" that can be connected in such environments makes interoperability among them a challenging problem. Moreover, the observations produced by these "Things" are made available with heterogeneous vocabularies and data formats. This heterogeneity prevents generic solutions from being adopted on a global scale and makes difficult to share and reuse data for other purposes than those for which they were originally set up. In this thesis, we address these challenges in the context of healthcare applications considering how we transform raw data to cognitive knowledge and ontology-based information shared between IoT system components. With respect to heterogeneity and integration challenges, our main contribution is an ontology-based IoT architecture allowing the deployment of semantic IoT applications. This approach allows sharing of sensors observations, contextualization of data and reusability of knowledge and processed information. Specific contributions include: * Design of the Cognitive Semantic Sensor Network ontology (CoSSN) ontology: CoSSN aims at overcoming the semantic interoperability challenges introduced by the variety of sensors potentially used. It also aims at describing expert knowledge related to a specific domain. * Design and implementation of SeMoM: SeMoM is a flexible IoT architecture built on top of CoSSN ontology. It relies on a message oriented middleware (MoM) following the publish/subscribe paradigm for a loosely coupled communication between system components that can exchange semantic observation data in a flexible way. From the applicative perspective, we focus on healthcare applications. Indeed, specific approaches and individual prototypes are preeminent solutions in healthcare which straighten the need of an interoperable solution especially for patients with multiple affections. With respect to these challenges, we elaborated two case studies 1) bedsore risk detection and 2) Activities of Daily Living (ADL) detection as follows: * We developed extensions of CoSSN to describe each domain concepts and we developed specific applications through SeMoM implementing expert knowledge rules and assessments of bedsore and human activities. * We implemented and evaluated the SeMoM framework in order to provide a proof of concept of our approach. Two experimentations have been realized for that target. The first is based on a deployment of a system targeting the detection of ADL activities in a real smart platform. The other one is based on ADLSim, a simulator of activities for ambient assisted living that can generate a massive amount of data related to the activities of a monitored person.