• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 6
  • 2
  • 1
  • 1
  • Tagged with
  • 9
  • 6
  • 6
  • 4
  • 3
  • 3
  • 3
  • 3
  • 3
  • 3
  • 3
  • 3
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Uso do padrão AMQP para transporte de mensagens entre atores remotos / Usage of the AMQP standard to transport messages among remote actors

Carmo, Thadeu de Russo e 08 May 2012 (has links)
O modelo de atores tem sido visto como uma abordagem alternativa à programação concorrente convencional, baseada em travas e variáveis de condição. Atores são agentes computacionais que se comunicam por troca de mensagens e que possuem uma caixa de correio e um comportamento. As mensagens destinadas a um ator são armazenadas na caixa de correio do ator e processadas de maneira assíncrona. Sistemas de middleware orientados a mensagens trabalham com troca assíncrona de mensagens e formam uma base que simplifica o desenvolvimento de aplicações distribuídas. Tais sistemas permitem interoperabilidade com baixo acoplamento e provêm suporte para tratamento robusto de erros em caso de falhas. Message brokers são frequentemente apresentados como uma tecnologia que pode mudar a maneira com que sistemas distribuídos são construídos. A especificação AMQP é uma proposta recente de padronização de um protocolo para message brokers. Neste trabalho exploramos a potencial sinergia entre um message broker e uma implementação do modelo de atores. Criamos uma versão modificada da implementação do modelo de atores do projeto Akka que utiliza um message broker AMQP como mecanismo de transporte de mensagens para atores remotos. / The actor model has been seen as an alternative for conventional concurrent programming based on locks and condition variables. Actors are computational agents that communicate by sending messages and have a mailbox and a behavior. The messages sent to an actor are stored in its mailbox and are asynchronously processed. Message oriented middleware systems work with asynchronous message exchange and create a base that simplifies the development of distributed applications. These systems have interoperability with low coupling and provide support for robust error handling in case of failures. Message brokers are often presented as a technology that can change the way distributed systems are built. The AMQP specification is a recent proposal of a standard protocol for message brokers. In this document we explore the potential synergy between a message broker and an implementation of the actor model. We created a modified version of the actor model implementation provided by the Akka project. Our modified implementation uses an AMQP message broker as the transport engine for messages to remote actors.
2

Uso do padrão AMQP para transporte de mensagens entre atores remotos / Usage of the AMQP standard to transport messages among remote actors

Thadeu de Russo e Carmo 08 May 2012 (has links)
O modelo de atores tem sido visto como uma abordagem alternativa à programação concorrente convencional, baseada em travas e variáveis de condição. Atores são agentes computacionais que se comunicam por troca de mensagens e que possuem uma caixa de correio e um comportamento. As mensagens destinadas a um ator são armazenadas na caixa de correio do ator e processadas de maneira assíncrona. Sistemas de middleware orientados a mensagens trabalham com troca assíncrona de mensagens e formam uma base que simplifica o desenvolvimento de aplicações distribuídas. Tais sistemas permitem interoperabilidade com baixo acoplamento e provêm suporte para tratamento robusto de erros em caso de falhas. Message brokers são frequentemente apresentados como uma tecnologia que pode mudar a maneira com que sistemas distribuídos são construídos. A especificação AMQP é uma proposta recente de padronização de um protocolo para message brokers. Neste trabalho exploramos a potencial sinergia entre um message broker e uma implementação do modelo de atores. Criamos uma versão modificada da implementação do modelo de atores do projeto Akka que utiliza um message broker AMQP como mecanismo de transporte de mensagens para atores remotos. / The actor model has been seen as an alternative for conventional concurrent programming based on locks and condition variables. Actors are computational agents that communicate by sending messages and have a mailbox and a behavior. The messages sent to an actor are stored in its mailbox and are asynchronously processed. Message oriented middleware systems work with asynchronous message exchange and create a base that simplifies the development of distributed applications. These systems have interoperability with low coupling and provide support for robust error handling in case of failures. Message brokers are often presented as a technology that can change the way distributed systems are built. The AMQP specification is a recent proposal of a standard protocol for message brokers. In this document we explore the potential synergy between a message broker and an implementation of the actor model. We created a modified version of the actor model implementation provided by the Akka project. Our modified implementation uses an AMQP message broker as the transport engine for messages to remote actors.
3

Distributed data analysis over meteorological datasets using the actor model

Sanchez, Jimmy Kraimer Martin Valverde January 2017 (has links)
Devido ao contínuo crescimento dos dados científicos nos últimos anos, a análise intensiva de dados nessas quantidades massivas de dados é muito importante para extrair informações valiosas. Por outro lado, o formato de dados científicos GRIB (GRIdded Binary) é amplamente utilizado na comunidade meteorológica para armazenar histórico de dados e previsões meteorológicas. No entanto, as ferramentas atuais disponíveis e métodos para processar arquivos neste formato não realizam o processamento em um ambiente distribuído. Essa situação limita as capacidades de análise dos cientistas que precisam realizar uma avaliação sobre grandes conjuntos de dados com o objetivo de obter informação no menor tempo possível fazendo uso de todos os recursos disponíveis. Neste contexto, este trabalho apresenta uma alternativa ao processamento de dados no formato GRIB usando o padrão Manager-Worker implementado com o modelo de atores fornecido pelo Akka toolkit. Realizamos também uma comparação da nossa proposta com outros mecanismos, como o round-robin, random, balanceamento de carga adaptativo, bem como com um dos principais frameworks para o processamento de grandes quantidades de dados tal como o Apache Spark. A metodologia utilizada considera vários fatores para avaliar o processamento dos arquivos GRIB. Os experimentos foram conduzidos em um cluster na plataforma Microsoft Azure. Os resultados mostram que nossa proposta escala bem à medida que o número de nós aumenta. Assim, nossa proposta atingiu um melhor desempenho em relação aos outros mecanismos utilizados para a comparação, particularmente quando foram utilizadas oito máquinas virtuais para executar as tarefas. Nosso trabalho com o uso de metadados alcançou um ganho de 53.88%, 62.42%, 62.97%, 61.92%, 62.44% e 59.36% em relação aos mecanismos round-robin, random, balanceamento de carga adaptativo que usou métricas CPU, JVM Heap e um combinado de métricas, e o Apache Spark, respectivamente, em um cenário onde um critério de busca é aplicado para selecionar 2 dos 27 parâmetros totais encontrados no conjunto de dados utilizado nos experimentos. / Because of the continuous and overwhelming growth of scientific data in the last few years, data-intensive analysis on this vast amount of scientific data is very important to extract valuable scientific information. The GRIB (GRIdded Binary) scientific data format is widely used within the meteorological community and is used to store historical meteorological data and weather forecast simulation results. However, current libraries to process the GRIB files do not perform the computation in a distributed environment. This situation limits the analytical capabilities of scientists who need to perform analysis on large data sets in order to obtain information in the shortest time possible using of all available resources. In this context, this work presents an alternative to data processing in the GRIB format using the well-know Manager-Worker pattern, which was implemented with the Actor model provided by the Akka toolkit. We also compare our proposal with other mechanisms, such as the round-robin, random and an adaptive load balancing, as well as with one of the main frameworks currently existing for big data processing, Apache Spark. The methodology used considers several factors to evaluate the processing of the GRIB files. The experiments were conducted on a cluster in Microsoft Azure platform. The results show that our proposal scales well as the number of worker nodes increases. Our work reached a better performance in relation to the other mechanisms used for the comparison particularly when eight worker virtual machines were used. Thus, our proposal upon using metadata achieved a gain of 53.88%, 62.42%, 62.97%, 61.92%, 62.44% and 59.36% in relation to the mechanisms: round-robin, random, an adaptive load balancing that used CPU, JVM Heap and mix metrics, and the Apache Spark respectively, in a scenario where a search criteria is applied to select 2 of 27 total parameters found in the dataset used in the experiments.
4

Distributed data analysis over meteorological datasets using the actor model

Sanchez, Jimmy Kraimer Martin Valverde January 2017 (has links)
Devido ao contínuo crescimento dos dados científicos nos últimos anos, a análise intensiva de dados nessas quantidades massivas de dados é muito importante para extrair informações valiosas. Por outro lado, o formato de dados científicos GRIB (GRIdded Binary) é amplamente utilizado na comunidade meteorológica para armazenar histórico de dados e previsões meteorológicas. No entanto, as ferramentas atuais disponíveis e métodos para processar arquivos neste formato não realizam o processamento em um ambiente distribuído. Essa situação limita as capacidades de análise dos cientistas que precisam realizar uma avaliação sobre grandes conjuntos de dados com o objetivo de obter informação no menor tempo possível fazendo uso de todos os recursos disponíveis. Neste contexto, este trabalho apresenta uma alternativa ao processamento de dados no formato GRIB usando o padrão Manager-Worker implementado com o modelo de atores fornecido pelo Akka toolkit. Realizamos também uma comparação da nossa proposta com outros mecanismos, como o round-robin, random, balanceamento de carga adaptativo, bem como com um dos principais frameworks para o processamento de grandes quantidades de dados tal como o Apache Spark. A metodologia utilizada considera vários fatores para avaliar o processamento dos arquivos GRIB. Os experimentos foram conduzidos em um cluster na plataforma Microsoft Azure. Os resultados mostram que nossa proposta escala bem à medida que o número de nós aumenta. Assim, nossa proposta atingiu um melhor desempenho em relação aos outros mecanismos utilizados para a comparação, particularmente quando foram utilizadas oito máquinas virtuais para executar as tarefas. Nosso trabalho com o uso de metadados alcançou um ganho de 53.88%, 62.42%, 62.97%, 61.92%, 62.44% e 59.36% em relação aos mecanismos round-robin, random, balanceamento de carga adaptativo que usou métricas CPU, JVM Heap e um combinado de métricas, e o Apache Spark, respectivamente, em um cenário onde um critério de busca é aplicado para selecionar 2 dos 27 parâmetros totais encontrados no conjunto de dados utilizado nos experimentos. / Because of the continuous and overwhelming growth of scientific data in the last few years, data-intensive analysis on this vast amount of scientific data is very important to extract valuable scientific information. The GRIB (GRIdded Binary) scientific data format is widely used within the meteorological community and is used to store historical meteorological data and weather forecast simulation results. However, current libraries to process the GRIB files do not perform the computation in a distributed environment. This situation limits the analytical capabilities of scientists who need to perform analysis on large data sets in order to obtain information in the shortest time possible using of all available resources. In this context, this work presents an alternative to data processing in the GRIB format using the well-know Manager-Worker pattern, which was implemented with the Actor model provided by the Akka toolkit. We also compare our proposal with other mechanisms, such as the round-robin, random and an adaptive load balancing, as well as with one of the main frameworks currently existing for big data processing, Apache Spark. The methodology used considers several factors to evaluate the processing of the GRIB files. The experiments were conducted on a cluster in Microsoft Azure platform. The results show that our proposal scales well as the number of worker nodes increases. Our work reached a better performance in relation to the other mechanisms used for the comparison particularly when eight worker virtual machines were used. Thus, our proposal upon using metadata achieved a gain of 53.88%, 62.42%, 62.97%, 61.92%, 62.44% and 59.36% in relation to the mechanisms: round-robin, random, an adaptive load balancing that used CPU, JVM Heap and mix metrics, and the Apache Spark respectively, in a scenario where a search criteria is applied to select 2 of 27 total parameters found in the dataset used in the experiments.
5

Distributed data analysis over meteorological datasets using the actor model

Sanchez, Jimmy Kraimer Martin Valverde January 2017 (has links)
Devido ao contínuo crescimento dos dados científicos nos últimos anos, a análise intensiva de dados nessas quantidades massivas de dados é muito importante para extrair informações valiosas. Por outro lado, o formato de dados científicos GRIB (GRIdded Binary) é amplamente utilizado na comunidade meteorológica para armazenar histórico de dados e previsões meteorológicas. No entanto, as ferramentas atuais disponíveis e métodos para processar arquivos neste formato não realizam o processamento em um ambiente distribuído. Essa situação limita as capacidades de análise dos cientistas que precisam realizar uma avaliação sobre grandes conjuntos de dados com o objetivo de obter informação no menor tempo possível fazendo uso de todos os recursos disponíveis. Neste contexto, este trabalho apresenta uma alternativa ao processamento de dados no formato GRIB usando o padrão Manager-Worker implementado com o modelo de atores fornecido pelo Akka toolkit. Realizamos também uma comparação da nossa proposta com outros mecanismos, como o round-robin, random, balanceamento de carga adaptativo, bem como com um dos principais frameworks para o processamento de grandes quantidades de dados tal como o Apache Spark. A metodologia utilizada considera vários fatores para avaliar o processamento dos arquivos GRIB. Os experimentos foram conduzidos em um cluster na plataforma Microsoft Azure. Os resultados mostram que nossa proposta escala bem à medida que o número de nós aumenta. Assim, nossa proposta atingiu um melhor desempenho em relação aos outros mecanismos utilizados para a comparação, particularmente quando foram utilizadas oito máquinas virtuais para executar as tarefas. Nosso trabalho com o uso de metadados alcançou um ganho de 53.88%, 62.42%, 62.97%, 61.92%, 62.44% e 59.36% em relação aos mecanismos round-robin, random, balanceamento de carga adaptativo que usou métricas CPU, JVM Heap e um combinado de métricas, e o Apache Spark, respectivamente, em um cenário onde um critério de busca é aplicado para selecionar 2 dos 27 parâmetros totais encontrados no conjunto de dados utilizado nos experimentos. / Because of the continuous and overwhelming growth of scientific data in the last few years, data-intensive analysis on this vast amount of scientific data is very important to extract valuable scientific information. The GRIB (GRIdded Binary) scientific data format is widely used within the meteorological community and is used to store historical meteorological data and weather forecast simulation results. However, current libraries to process the GRIB files do not perform the computation in a distributed environment. This situation limits the analytical capabilities of scientists who need to perform analysis on large data sets in order to obtain information in the shortest time possible using of all available resources. In this context, this work presents an alternative to data processing in the GRIB format using the well-know Manager-Worker pattern, which was implemented with the Actor model provided by the Akka toolkit. We also compare our proposal with other mechanisms, such as the round-robin, random and an adaptive load balancing, as well as with one of the main frameworks currently existing for big data processing, Apache Spark. The methodology used considers several factors to evaluate the processing of the GRIB files. The experiments were conducted on a cluster in Microsoft Azure platform. The results show that our proposal scales well as the number of worker nodes increases. Our work reached a better performance in relation to the other mechanisms used for the comparison particularly when eight worker virtual machines were used. Thus, our proposal upon using metadata achieved a gain of 53.88%, 62.42%, 62.97%, 61.92%, 62.44% and 59.36% in relation to the mechanisms: round-robin, random, an adaptive load balancing that used CPU, JVM Heap and mix metrics, and the Apache Spark respectively, in a scenario where a search criteria is applied to select 2 of 27 total parameters found in the dataset used in the experiments.
6

Extension of the behavior composition framework in presence of failures using recovery techniques and AKKA / Extension d'un cadre de composition de comportements en présence de pannes à l'aide de techniques de reprise et de AKKA

Hosnidokht, Mohammad January 2018 (has links)
Abstract: Fault tolerance is an essential property to be satis ed in the composition of services, but reaching a high level of fault tolerance remains a challenge. In the area of ubiquitous computing, the composition of services is inevitable when a request cannot be carried out by a single service, but by a combination of several services. This thesis studies fault tolerance in the context of a general behavior composition framework. This approach raises, rst, the problem of the synthesis of controllers (or compositions) in order to coordinate a set of available services to achieve a new service, the target service and, second, the exploitation of all compositions to make the new service fault tolerant. Although a solution has been proposed by the authors of the behavior composition framework, it is incomplete and has not been evaluated experimentally or in situ. This thesis brings two contributions to this problem. On one hand, it considers the case in which the service selected by the controller is temporarily or permanently unavailable by exploiting recovery techniques to identify a consistent state of the system from which it may progress using other services or leave it in a coherent state when none of the available services no longer allows progression. On the other hand, it evaluates several recovery solutions, each useful in services malfunction situations, using a case study implemented with the aid of Akka, a tool that facilitates the development of reactive, concurrent and distributed systems. / La tolérance aux fautes est une propriété indispensable à satisfaire dans la composition de services, mais atteindre un haut de niveau de tolérance aux fautes représente un défi majeur. Dans l'ère de l'informatique ubiquitaire, la composition de services est inévitable lorsqu'une requête ne peut être réalisée par un seul service, mais par la combinaison de plusieurs services. Ce mémoire étudie la tolérance aux fautes dans le contexte d'un cadre général de composition de comportements (behavior composition framework en anglais). Cette approche soulève, tout d'abord, le problème de la synthèse de contrôleurs (ou compositions) de façon à coordonner un ensemble de services disponibles afin de réaliser un nouveau service, le service cible et, ensuite, celui de l'exploitation de l'ensemble des compositions afin de rendre le nouveau service tolérant aux fautes. Bien qu'une solution ait été proposée par les auteurs de ce cadre de composition, elle est incomplète et elle n'a pas été évaluée expérimentalement ou in situ. Ce mémoire apporte deux contributions à ce problème. D'une part, il considère le cas dans lequel le service visé par le contrôleur est temporairement ou définitivement non disponible en exploitant des techniques de reprise afin d'identifier un état cohérent du système à partir duquel il peut progresser en utilisant d'autres services ou de le laisser dans un état cohérent lorsqu'aucun service, parmi ceux disponibles, ne permet plus de progression. D'autre part, il évalue plusieurs solutions de reprise, chacune utile dans des situations particulières de pannes, à l'aide d'une étude de cas implémentée en Akka, un outil qui permet aisément de mettre en oeuvre des systèmes réactifs, concurrents et répartis.
7

Extension of the behavior composition framework in presence of failures using recovery techniques and AKKA

Hosnidokht, Mohammad January 2018 (has links)
Abstract: Fault tolerance is an essential property to be satis ed in the composition of services, but reaching a high level of fault tolerance remains a challenge. In the area of ubiquitous computing, the composition of services is inevitable when a request cannot be carried out by a single service, but by a combination of several services. This thesis studies fault tolerance in the context of a general behavior composition framework. This approach raises, rst, the problem of the synthesis of controllers (or compositions) in order to coordinate a set of available services to achieve a new service, the target service and, second, the exploitation of all compositions to make the new service fault tolerant. Although a solution has been proposed by the authors of the behavior composition framework, it is incomplete and has not been evaluated experimentally or in situ. This thesis brings two contributions to this problem. On one hand, it considers the case in which the service selected by the controller is temporarily or permanently unavailable by exploiting recovery techniques to identify a consistent state of the system from which it may progress using other services or leave it in a coherent state when none of the available services no longer allows progression. On the other hand, it evaluates several recovery solutions, each useful in services malfunction situations, using a case study implemented with the aid of Akka, a tool that facilitates the development of reactive, concurrent and distributed systems. / La tolérance aux fautes est une propriété indispensable à satisfaire dans la composition de services, mais atteindre un haut de niveau de tolérance aux fautes représente un défi majeur. Dans l'ère de l'informatique ubiquitaire, la composition de services est inévitable lorsqu'une requête ne peut être réalisée par un seul service, mais par la combinaison de plusieurs services. Ce mémoire étudie la tolérance aux fautes dans le contexte d'un cadre général de composition de comportements (behavior composition framework en anglais). Cette approche soulève, tout d'abord, le problème de la synthèse de contrôleurs (ou compositions) de façon à coordonner un ensemble de services disponibles afin de réaliser un nouveau service, le service cible et, ensuite, celui de l'exploitation de l'ensemble des compositions afin de rendre le nouveau service tolérant aux fautes. Bien qu'une solution ait été proposée par les auteurs de ce cadre de composition, elle est incomplète et elle n'a pas été évaluée expérimentalement ou in situ. Ce mémoire apporte deux contributions à ce problème. D'une part, il considère le cas dans lequel le service visé par le contrôleur est temporairement ou définitivement non disponible en exploitant des techniques de reprise afin d'identifier un état cohérent du système à partir duquel il peut progresser en utilisant d'autres services ou de le laisser dans un état cohérent lorsqu'aucun service, parmi ceux disponibles, ne permet plus de progression. D'autre part, il évalue plusieurs solutions de reprise, chacune utile dans des situations particulières de pannes, à l'aide d'une étude de cas implémentée en Akka, un outil qui permet aisément de mettre en oeuvre des systèmes réactifs, concurrents et répartis.
8

Representation of asynchronous communication protocols in Scala and Akka

Eriksson, Joakim January 2013 (has links)
This thesis work investigates how to represent protocols for asynchronous communication in the Scala programming language and the Akka actor framework, to be run on Java Virtual Machine (JVM). Further restrictions from the problem domain - the coexistence of multiple protocol instances sharing the same Java thread - imply that neither an asynchronous call waiting for response nor anything else can block the underlying Java threads. A common way to represent asynchronous communication protocols is to use state machines. This thesis seeks a way to shrink the size of and to reduce the complexity of the protocol implementations by representing sequences of asynchronous communication calls (i.e. sequences of sent and received messages) as a type of procedure. The idea is find a way to make the procedures that contain asynchronous calls look like synchronous communication procedures by hiding the asynchronous details. In other words, the resulting procedure code should show what to do and not so much focus on how to overcome the impediment of the asynchronous calls. With the help of an asynchronous communication protocol toy example, this report shows how such an protocol can be implemented with a combination of a state machine and a procedure representation in Scala and Akka. The procedure representation hides away the asynchronous details by using the Scala capability to use CPS-transformed delimited continuations. As a sub-problem, this thesis also shows how to safely schedule asynchronous communication timeouts with help of Scala and Akka within the restrictions of the thesis problem domain.
9

物聯網與大數據平台之通訊架構設計與實作 / Design and Implementation of the Communication Architecture for IoT & Big Data Platform

胡學賓, Hu, Hsueh Pin Unknown Date (has links)
本研究針對物聯網與雲端大數據分析之不同程度的通訊需求,以微服務架構為基礎,設計一個四層式物聯網與大數據平台之通訊架構。面對物聯網之即時通訊需求,本研究採用MQTT通訊協定做為解決方案,而雲端大數據分析之通訊需求則採用Apache Kafka。 本研究基於參與者模型(Actor Model)所提出之「裝置代理人」,全面的解決了物聯網中異質通訊協定所產生的複雜性,同時解決了集中式物聯網閘道器所造成的系統複雜性與效能瓶頸,使物聯網閘道器能進行分散式部署,並且共享運算資源。

Page generated in 0.0197 seconds