The Internet of Things is expected to bring large and promising spectrum of social goods in various domains. Several new challenges arise or are to be reconsidered within the IoT systems supporting these goods, among them the Quality of Service (QoS) issue. The goal of this paper is first to introduce our approach for an autonomic Middleware-level QoS management of IoT systems. As a contribution at the second maturity level of the autonomic computing paradigm such as defined by IBM, it is then to propose and validate, within an emulation testbed platform, a proof of concept-oriented architecture of a monitoring component allowing detecting QoS degradation symptoms. We also demonstrate the benefits that could be gained from simple network-inspired QoS-oriented adaptation actions.

Un bus de service autonome pour des systèmes distribués à base de services Avec l’évolution des technologies de l’internet, les applications et plus généralement lessystèmes distribués sont de plus en plus conçus en composant et interconnectant un ensemblede services distribués. Ces services pouvant être très hétérogènes, plusieurs approches etsolutions pour la gestion de l’intégration et l’interopérabilité ont été proposées. De toutes cespropositions, les bus de services (ESB – Enterprise Service Bus) ont été désignés commeétant la solution la plus adaptée. Toutefois, le problème avec les ESB est qu’ils sont déployésdans un contexte très évolutif et très dynamique; un contexte dans lequel un grand nombre deservices peuvent être fournis et utilisés de façon concurrente à travers le bus. L’utilisationconcurrente de ces services mais aussi des ressources sous-jacentes allouées au bus (mémoire,processeur, etc.) peut conduire à des événements imprévisibles tels qu’une surcharge du bus,une indisponibilité des services, des temps de réponse élevés, une diminution de la fiabilité,etc. Dans ce contexte, des solutions efficaces permettant de garantir ou d'améliorer à la fois laqualité de service et l'évolutivité offertes par les ESB sont nécessaires. Le but de cette thèse est de proposer les principes architecturaux pour la mise en place un busde service autonome (ASB) qui offre une solution de communication scalable guidée par lestransactions des systèmes interconnectés, mais aussi par les ressources disponibles. L’ASBoffre aussi un service d’intégration différentiée en fonction des exigences en termes de qualitéde service spécifiques aux systèmes interconnectés

Nowadays, the Internet of Things (IoT) paradigm is expanding immensely. It brings ubiquitous intelligence through the interconnection of equipment (sensors, devices, etc.) to the Internet, allowing the birth of a new era of applications in various fields. One of the main challenges to be considered in the IoT is the Quality of Service (QoS) issue. Aiming to support the vision of an autonomic Middleware-level QoS management for IoT systems, this paper proposes and validates a functional architecture of a monitoring component for the detection of QoS degradation. We also demonstrate the benefits that could be gained from simple QoS-oriented adaptation actions dynamically executed to overcome the detected degradations.

Modern dynamic distributed systems require to dynamically take into account at runtime the changes in users' needs and the execution environment variations in order to improve the quality of service. The evolution of distributed systems, through the smart management of their properties and the extension of the existing integration infrastructures, becomes a necessity. Autonomic computing allows the self-management of system properties at runtime, according to fluctuations in the environment and changes in users' requirements. However, the mechanisms for parallel and distributed execution of multiple self-management processes have not been addressed substantially. It is critical to coordinate the execution of several processes performed by different autonomic managers, while still guaranteeing specific and global goals achievement. We address this issue by proposing a software architecture that allows the coordination of multiple autonomic managers which handle several component-based and service-oriented collaborative software entities. This architecture offers a distributed cross-layer self-management solution through orchestration and choreography. Using both techniques, autonomic managers running on multiple locations and different layers will be able to achieve their goals in a consistent and cost-effective way. In this paper, we present a set of mechanisms intended to coordinate the distributed execution of a set of self-management processes in one or more layers. We have chosen an use case involving the self-management of autonomic data replication systems integrated via an autonomic service bus in order to illustrate our approach.

1. ESBay Case Study. 2. Service-Oriented and Cloud Computing Architectures. 3. SPaaS 1.0 Cookbook. 4. SSOAPaaS 1.0 Cookbook. 5. SSOAPaaS 2.0 Cookbook. 6. SSOAPaaS 3.0 Cookbook.

The growing complexity and scale of systems implies challenges to include Autonomic Computing capabilities that help maintaining or improving the performance, availability and reliability of nowadays systems. In dynamic environments, the systems have to deal with changing conditions and requirements; thereby the autonomic features need a better technique to analyze and diagnose problems, and learn about the functioning conditions of the managed system. In the medical diagnostic area, the tests have included statistical and probabilistic models to aid and improve the results and select better medical treatments. We propose a probabilistic approach to implement an analysis process. The base of our approach is building a Bayesian network as model representing runtime properties of the Managed Element and their relationships. The Bayesian network is initially built from monitored data of an Enterprise Service Bus platform under different workload conditions, by means a structure learning algorithm. We aim to improve the functionalities of an Enterprise Service Bus platform integrating monitoring and fault diagnosis capabilities. A case study is presented to prove the effectiveness of our approach.

Heterogeneous networks, including personal, home and collaborative enterprise networks are aimed at interconnecting diverse communities of users. In the case of large communities of mobile, distributed and collaborating users sharing their multimedia content, scalable and semantic approaches are needed to facilitate the efficient publication and discovery of data. In this work, we propose an approach aimed at efficiently providing the required functionalities for dynamically managing the resources and the multimedia content of interconnected home networks, as well as providing the required community sharing capabilities. Our solution consists in combining the benefits of both ontologies and peer-to-peer approaches, by taking advantage of the semantic capabilities of ontologies at local domain level and the scalable P2P solutions at inter-domains level. Our proposed framework is founded on a new architecture based on peer-to-peer event-based communication system and using mainly a home box entity based on a global multimedia ontology. This domain ontology allows publishers and subscribers to use a common semantic space to characterise production and consumption of resources and services. Our approach can also be applied for similar large interconnected systems scenarios such as topic discovery with JNDI and service exchange with UDDI.

Enterprise Service Bus (ESB) allows the integration of pervasive, distributed and networked systems more and more designed as a composition of atomic and heterogeneous services. Offered services have various types of Quality of Service (QoS) that can be unpredictable and may lead to undesirable situations such as service unavailability, high response time, decrease of reliability, etc. These kinds of situation may be unacceptable for service consumers or systems that require a more controllable reliability, efficiency, performance, etc. Being between the service providers and requesters, the goal of this paper is to show how efficient strategies can be associated to the ESB to give it the ability to guarantee or to take into account the satisfaction of non-functional requirements during the service mediation. Using a cloud-based deployment, we will show also how to guarantee the ESB scalability when a large number of concurrent services can be provided and consumed via the ESB.

The growing complexity and scale of systems implies challenges to include Autonomic Computing capabilities that help to maintain or improve the performance, availability and reliability characteristics. The autonomic management of a system can be defined deterministically based on experiment observations on the system and possible results of associated plans. However in dynamic environments with changing conditions and requirements, a better technique to diagnose observations and learn about the functioning conditions of the managed system is needed to guide the autonomic management. In the case of medical diagnostic, tests have included statistical and probabilistic models to aid and improve the results and select better medical treatments. In this paper we also adopt a probabilistic approach to define a Bayesian network from monitored data of an Enterprise Service Bus under different workload conditions. This model is used by the Autonomic Service Bus as a knowledge base to diagnose the cause of degradation problems and repair them. Experimental results assess the effectiveness of our approach.

With the development of Internet technologies, distributed and multimedia applications are more and more used. Because these applications have Quality of Service (QoS) expectations (bounded delay, guaranteed bandwidth, etc.), a set of QoS oriented protocols aimed at aiding end-to-end communications are proposed. Each one of these protocols offers one or several mechanisms. Selecting the best protocol with the best mechanism(s) according to the application requirements and network contexts is a challenge if we want to provide enough QoS to the final users. Also because contexts can evolve, for example when network characteristics change, the previous selected protocol can become obsolete. In this context, an adaptation is mandatory. In this paper, we show how the OSI transport layer can become autonomic in order to perform dynamically reconfiguration actions when it is needed. Our approach consists in including at the transport layer the MAPE loop of the autonomic computing proposal to monitor and analyze the communication in order to predict or to detect a QoS degradation induced by changes of network context, and to plan and execute dynamically new reconfiguration actions in order to keep the communication at the expected QoS level. To implement and evaluate our approach, we extend the new Multipath-TCP (MPTCP) protocol, which offers an extensible application oriented layer. This layer gives opportunity to deploy QoS-aware mechanisms. The test results show that the consumption time of autonomic selection and composition of these mechanisms at run time does not prevent enhancing the QoS, and encourage us to continue working in this approach.

The increasing number of interfaces using different access technologies in modern devices gives opportunities for enhancing the Quality of Service (QoS) delivered to multimedia and interactive data transfers involved in distributed applications. In the current Internet, the presence of "middleboxes" such as Network Address Translators (NAT) or firewalls hardly lets applications use any Transport protocol but the well-known Transmission Control Protocol (TCP). Currently under standardization at the IETF, the new Multipath-TCP (MPTCP) protocol uses several TCP flows to make use of the multiple interfaces available on the end terminals, thus improving both network availability and QoS, still being capable to cross over middle boxes. Although originally being fully reliable and fully ordered, its two sub-layers architecture gives opportunity to use QoS techniques over fully reliable paths. This paper studies the QoS benefits induced by the implementation of the "partial reliability" concept in MPTCP for interactive video applications. Two different mechanisms are experimented with the aim to enhance global quality of video transmission over paths close to 3G networks characteristics. The Internet Engineering Task Force (IETF) MPTCP working group's ns-2 implementation is used for the experimentations. The Evalvid toolkit over ns-2 is used to measure the QoS benefits expressed in terms of Peak Signal to Noise Ratio (PSNR).

The large diversity and heterogeneity of collaborating objects and services in modern networked environments such as personal, vehicular or home networks makes difficult the efficient configuration, deployment and management of composite dynamic systems without a major participation of human actors. This paper proposes an autonomic architecture based on decision models built on ontologies and aimed at self-configuring and self-adapting service-oriented and event-driven distributed systems. Our solution follows a top-down approach based on well-known and accepted standards and implemented as an ecosystem-wide ontology aimed at characterizing the properties of services and events related to consumers and producers entities participating in the autonomic collaborative environment.