Service-oriented architectures are a popular architectural paradigm for building software applications from a number of loosely coupled, distributed services. Through a set of procedural rules, workflow technologies define how groups of services coordinate with one another to achieve a shared task. A problem with workflow specifications is that often the patterns of interaction between the distributed services are too complicated to predict and analyse at design-time. In certain cases, the exact patterns of message exchange and the concrete services to call cannot be predicted in advance, due to factors such as fluctuating network load or the availability of services. It is a more realistic assumption to endow software components with the ability to make decisions about the nature and scope of their interactions at runtime. Multiagent systems offer a complementary paradigm: building software applications from a number of self interested, autonomous agents. This thesis presents an investigation into fusing the agency and service-oriented architecture paradigms, in order to facilitate flexible, workflow composition. Our approach offers an agent-based solution to service choreography and is founded on the concept of shared interaction protocols. By adopting an agent-based approach to service choreography, active autonomous agents can utilise the typically passive service-oriented architectures, found in Internet and Grid systems. In contrast with statically defined, centralised service orchestrations, decentralised agents can perform service choreography at runtime, allowing them to operate in scenarios where it is not possible to define the pattern of interaction in advance. Application to real scenarios is a driving factor behind this research. By working closely with a number of active Grid projects, namely AstroGrid and the Large-Synoptic Survey Telescope (LSST), a concrete set of requirements for scientific workflow have been derived, based on realistic science problems. This research has resulted in the MultiAgent Service Choreography (MASC) language to express scientific workflow, methodology for system building and a software framework which performs agent based Web service choreography, in order to enact distributed e-Science experiments. Evaluation of this thesis is conducted through case study, applying the language, methodology and software framework to solve a motivating set of workflow scenarios.

Ensemble learning is a promising direction of research in machine learning, in which an ensemble classifier gives better predictive and more robust performance for classification problems by combining other learners. Meanwhile agent-based systems provide frameworks to share knowledge from multiple agents in an open context. This thesis combines multi-agent knowledge sharing with ensemble methods to produce a new style of learning system for open environments. We now are surrounded by many smart objects such as wireless sensors, ambient communication devices, mobile medical devices and even information supplied via other humans. When we coordinate smart objects properly, we can produce a form of collective intelligence from their collaboration. Traditional ensemble methods and agent-based systems have complementary advantages and disadvantages in this context. Traditional ensemble methods show better classification performance, while agent-based systems might not guarantee their performance for classification. Traditional ensemble methods work as closed and centralised systems (so they cannot handle classifiers in an open context), while agent-based systems are natural vehicles for classifiers in an open context. We designed an open and social ensemble learning architecture, named J-model, to merge the conflicting benefits of the two research domains. The J-model architecture is based on a service choreography approach for coordinating classifiers. Coordination protocols are defined by interaction models that describe how classifiers will interact with one another in a peer-to-peer manner. The peer ranking algorithm recommends more appropriate classifiers to participate in an interaction model to boost the success rate of results of their interactions. Coordinated participant classifiers who are recommended by the peer ranking algorithm become an ensemble classifier within J-model. We evaluated J-model’s classification performance with 13 UCI machine learning benchmark data sets and a virtual screening problem as a realistic classification problem. J-model showed better performance of accuracy, for 9 benchmark sets out of 13 data sets, than 8 other representative traditional ensemble methods. J-model gave better results of specificity for 7 benchmark sets. In the virtual screening problem, J-model gave better results for 12 out of 16 bioassays than already published results. We defined different interaction models for each specific classification task and the peer ranking algorithm was used across all the interaction models. Our research contributions to knowledge are as follows. First, we showed that service choreography can be an effective ensemble coordination method for classifiers in an open context. Second, we used interaction models that implement task specific coordinations of classifiers to solve a variety of representative classification problems. Third, we designed the peer ranking algorithm which is generally and independently applicable to the task of recommending appropriate member classifiers from a classifier pool based on an open pool of interaction models and classifiers.

Orientador: Edmundo Roberto Mauro Madeira / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-09T05:05:03Z (GMT). No. of bitstreams: 1 Nakai_AlanMassaru_M.pdf: 7384395 bytes, checksum: 7ae9dd61b3367851a744bc93622be309 (MD5) Previous issue date: 2007 / Resumo: Uma cadeia produtiva é definida como o conjunto de atividades envolvidas na criação, transformação e distribuição de um produto, da matéria-prima ao consumidor final. Os participantes da cadeia produtiva podem trabalhar de forma integrada, com o objetivo de otimizar o desempenho coletivo, aumentando sua competitividade comercial. Do ponto de vista tecnológico, a natureza distribuída, autônoma e heterogênea dos participantes da cadeia produtiva dificultam a automação dos seus processos de negócio interorganizacionais. Este trabalho propõe uma infra-estrutura baseada em coreografias de serviços Web para coordenar as atividades que compõem os processos de negócio interorganizacionais das cadeias produtivas. Esta infra-estrutura implementa um modelo de coordenação que visa facilitar o projeto e a implantação dos processos de negócio interorganizacionais. Neste modelo, os processos são representados por coreografias WS-CDL, que são mapeadas para planos de coordenação executáveis descritos em BPEL. O trabalho também apresenta um protótipo da infra-estrutura, com o objetivo de validá-la / Abstract: A supply chain is the set of activities involved in the creation, transformation and distribution of a product, from raw material to the consumer. Supply chain's participants can work in an integrated way to optimize their performance and increase their commercial competitiveness. From the technological point of view, the distributed, autonomous and heterogeneous nature of supply chain's participants raises di±culties when we consider the automation of interorganizational processes. This work proposes an infrastructure based on Web services choreographies for coordination of the activities that compose the interorganizational business processes of the supply chains. This infrastructure implements a coordination model that aims to make easier the design and the deployment of the interorganizational business processes. In this model, processes are represented by WS-CDL choreographies, which are mapped to executable BPEL coordination plans. The work also presents a prototype of the infrastructure to validate it / Mestrado / Sistemas Distribuidos / Mestre em Ciência da Computação

This thesis project presents the web service choreography approach used for the composition of web services. It leverages the CHOReVOLUTION platform, a future-oriented and scalable platform, that is used to design and deploy web service choreographies. By using this platform, a use case that falls into the ITS domain is developed. This use case highlights the benefits of the web service choreography when used for the development of ITS applications. The necessary web services are designed and their interactions are defined through a choreography diagram that graphically represents how the services should collaborate together to fulfill a specific goal. By using the choreography diagram as input to the platform and by registering the web services on a web server, the choreography is deployed over the platform. The resulted choreography is tested in terms of services coordination. It is demonstrated that the platform can generate specific components that are interposed between the services and are able to take care of the services coordination for the use case created. Moreover, the execution time required to complete the choreography is measured, analyzed and reported under different conditions. Finally, it is shown that the execution time varies depending on the data that the services have to process and that the processing of huge data sets may lead to high execution times. / Detta examensarbete behandlar hur man med hjälp koreografering av webbtjänster kan komponera webbtjänster. Det använder sig av CHOReVOLUTION plattformen, en framåtblickande och skalbar plattform, som används för att designa och verkställa koreografering av webbtjänster. Med denna plattform skapas ett användningsfall inom ITS-området. Detta fall belyser fördelarna med webbtjänskoreografi i samband med utveckling av ITS- applikationer. De nödvändiga webbtjänsterna designas och deras samspel definieras genom ett diagram för koreografin, som på ett grafiskt vis presenterar hur tjänsterna skall kollaborera för att nå ett specifikt mål. Genom att mata plattformen med data från diagrammet, och genom att registrera webbtjänster på en webbserver, verkställs koreografin. Med resultatet testas koordineringen av tjänsterna. I detta examensarbete visas det att plattformen kan skapa specifika komponenter som interagerar med tjänsterna, samt sköta koordineringen av tjänster som krävs för detta användningsfall. Exekveringstiden mäts, analyseras och rapporteras under flera olika omständigheter. Det demonstreras också att exekveringstiden varierar beroende på den data som tjänsterna måste behandla, och hur behandlingen av mycket stora datamängder kan leda till långa exekveringstider.

Seleção de serviços em composições distribuídas considera principalmente a qualidade de serviço que atenda requisitos estabelecidos pelo usuário, como por exemplo, preço. No entanto, problemas relacionados a execução de composições de serviços podem ocorrer quando não se considera aspectos relacionados à rede e ao hardware, que afetam diretamente o desempenho da composição. Esse problema se agrava em composições coreografadas, pois a característica descentralizada requer um maior esforço para que essas informações possam ser consideradas em uma perspectiva global. Dessa forma, apesar da descentralização apresentar vantagens, é necessário que requisitos de qualidade de serviço da composição também sejam considerados em coreografias de serviços web para que a escolha de serviços para desempenhar um papel leve em consideração parâmetros importantes que podem afetar no desempenho da composição. Este trabalho apresenta um mecanismo, implementado sobre o framework OpenKnowledge, para selecionar serviços web em ambientes coreografados considerando primeiramente estimativas de atraso, taxa de perda e por fim considera a utilização de outros parâmetros, como utilização de CPU. Os primeiros experimentos em diferentes cenários de rede confirmaram as vantagens da proposta em relação a um seletor de serviços que ignora aspectos relacionados com a rede. Obteve-se ganhos de 20 a 97% no que diz respeito ao tempo total da execução da coreografia. Em seguida, experimentos inserindo utilização de CPU na escolha dos serviços confirmaram as vantagens de utilização de diferentes parâmetros para seleção de serviços em coreografias. / Selection of services in distributed compositions mainly considers the quality of service that meets requirements set by the user, such as price. However, problems related to the execution of service compositions may occur when aspects related to network and hardware, which directly affect the performance of the composition, are not taken in consideration. This problem is aggravated in choreographed compositions because the decentralized feature requires a greater effort so that this information may be considered in a global perspective. Thus, despite the decentralization advantages, it is necessary that the service quality requirements of the composition may also be considered in the choreography of web services so that the choice of services may play a role that takes into account important parameters, which can affect the performance of the composition. This dissertation presents a mechanism, implemented on the OpenKnowledge framework, to select web services in choreographed environments. First, considering delay estimates, loss rate and finally the use of other parameters, such as CPU utilization. The first experiments on different network scenarios confirmed the proposal advantages in relation to a selection mechanism that bypasses aspects of the network. It was obtained gains from 20% to 97% considering the total time of execution of the choreography. Then, experiments inserting CPU utilization in the choice of services confirmed the advantages of using different parameters for selection of services in choreography.

Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2017-05-19T13:20:49Z No. of bitstreams: 2 Tese - Raphael de Aquino Gomes - 2017.pdf: 11643713 bytes, checksum: 23f27a75ccf239e427777100a4ef504f (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-05-19T13:24:29Z (GMT) No. of bitstreams: 2 Tese - Raphael de Aquino Gomes - 2017.pdf: 11643713 bytes, checksum: 23f27a75ccf239e427777100a4ef504f (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-05-19T13:24:29Z (GMT). No. of bitstreams: 2 Tese - Raphael de Aquino Gomes - 2017.pdf: 11643713 bytes, checksum: 23f27a75ccf239e427777100a4ef504f (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-04-06 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / This thesis proposes a model-based approach to abstracting, simplifying, and automating cloud resource management decisions to deploy a set of service choreographies subject to non-functional constraints. Given a high-level description of service choreographies and related constraints, the approach autonomously performs resource estimation, selection, and allocation in a hybrid cloud environment with multiple cloud providers whilst decreases resource utilization costs and inter-services communication overhead. The main motivation for this work is because service choreographies are widely used for the development of solutions with complex needs, with service sharing among them. This scenario turns resource management a challenging task, mainly due to the different roles that a service assumes, the interference among constraints, and a large number of available resource types. This thesis also proposes an architecture that extends the approach with strategies to dynamic resource management to face constraint violations. This architecture was partially implemented in a prototype that was used in the proposed approach evaluation. / Esta tese apresenta uma abordagem baseada em modelos para abstrair, simplificar e automatizar as decisões de gerenciamento de recursos em nuvem ao implantar um conjunto de coreografias de serviços sujeitas a restrições não-funcionais. Dada uma descrição em alto nível das coreografias e das restrições relacionadas, a abordagem realiza de maneira autônoma a estimativa, a seleção e a alocação dos recursos em um ambiente de nuvem híbrida com múltiplos provedores, enquanto reduz os custos associados à utilização dos recursos e o atraso de comunicação entre os serviços. A principal motivação para o seu desenvolvimento se deve ao fato das coreografias de serviço serem amplamente utilizadas para o desenvolvimento de soluções com necessidades complexas, que geralmente compartilham serviços entre si. Isso faz com que o gerenciamento de recursos esteja sujeito a uma série de desafios, principalmente relacionados aos diferentes papéis que um serviço assume, à interferência que uma restrição causa em outra e à grande quantidade de tipos de recurso a serem considerados. A tese também propõe uma arquitetura que agrega à abordagem uma proposta para a automação das atividades relacionadas ao gerenciamento dinâmico de recursos como forma de reparar violações detectadas sobre as restrições. Esta arquitetura foi parcialmente implementada em um protótipo que foi utilizado na avaliação da abordagem.

Seleção de serviços em composições distribuídas considera principalmente a qualidade de serviço que atenda requisitos estabelecidos pelo usuário, como por exemplo, preço. No entanto, problemas relacionados a execução de composições de serviços podem ocorrer quando não se considera aspectos relacionados à rede e ao hardware, que afetam diretamente o desempenho da composição. Esse problema se agrava em composições coreografadas, pois a característica descentralizada requer um maior esforço para que essas informações possam ser consideradas em uma perspectiva global. Dessa forma, apesar da descentralização apresentar vantagens, é necessário que requisitos de qualidade de serviço da composição também sejam considerados em coreografias de serviços web para que a escolha de serviços para desempenhar um papel leve em consideração parâmetros importantes que podem afetar no desempenho da composição. Este trabalho apresenta um mecanismo, implementado sobre o framework OpenKnowledge, para selecionar serviços web em ambientes coreografados considerando primeiramente estimativas de atraso, taxa de perda e por fim considera a utilização de outros parâmetros, como utilização de CPU. Os primeiros experimentos em diferentes cenários de rede confirmaram as vantagens da proposta em relação a um seletor de serviços que ignora aspectos relacionados com a rede. Obteve-se ganhos de 20 a 97% no que diz respeito ao tempo total da execução da coreografia. Em seguida, experimentos inserindo utilização de CPU na escolha dos serviços confirmaram as vantagens de utilização de diferentes parâmetros para seleção de serviços em coreografias. / Selection of services in distributed compositions mainly considers the quality of service that meets requirements set by the user, such as price. However, problems related to the execution of service compositions may occur when aspects related to network and hardware, which directly affect the performance of the composition, are not taken in consideration. This problem is aggravated in choreographed compositions because the decentralized feature requires a greater effort so that this information may be considered in a global perspective. Thus, despite the decentralization advantages, it is necessary that the service quality requirements of the composition may also be considered in the choreography of web services so that the choice of services may play a role that takes into account important parameters, which can affect the performance of the composition. This dissertation presents a mechanism, implemented on the OpenKnowledge framework, to select web services in choreographed environments. First, considering delay estimates, loss rate and finally the use of other parameters, such as CPU utilization. The first experiments on different network scenarios confirmed the proposal advantages in relation to a selection mechanism that bypasses aspects of the network. It was obtained gains from 20% to 97% considering the total time of execution of the choreography. Then, experiments inserting CPU utilization in the choice of services confirmed the advantages of using different parameters for selection of services in choreography.