Esta tese propõe um ambiente automático para desenvolvimento de simulação distribuída ASDA (Ambiente de Simulação Distribuída Automático), que tem como objetivo principal facilitar a utilização e desenvolvimento de simulação distribuída. As funcionalidades definidas no ASDA tornam-o diferente de todos os outros ambientes encontrados na literatura. A especificação do ASDA foi realizada através de um diagrama modular composto por sete módulos e também com o auxílio da ferramenta UML (Unified Modelling Language), através da utilização de três de seus diagramas: de casos de uso, de classes e de atividades. O ASDA permite aos usuários a utilização de simulação distribuída através da definição de uma nova simulação ou da replicação de um programa de simulação já desenvolvido. Se a opção for pelo desenvolvimento de um novo programa de simulação, o usuário deve fornecer o modelo e os parâmetros e o ambiente se encarrega de gerar o código do programa de simulação utilizando a abordagem que proporciona o melhor desempenho, levando em consideração as características do modelo e da plataforma. Além da especificação do ASDA, esta tese definiu um protótipo do ambiente com o objetivo de mostrar sua viabilidade de utilização. Neste protótipo, três módulos foram implementados, destacando-se o módulo Replicador, que utiliza a abordagem MRIP (Multiple Replication in Parallel). Esta tese contribui também com a definição de algumas diretrizes para a utilização da abordagem MRIP. A base para essa definição foram os resultados obtidos com a utilização do módulo Replicador / This thesis proposes an automatic environment for the development of distributed simulation ASDA (Ambiente de Simulação Distribuída Automático (in Portuguese), whose main goal is to make easier the use and development of distributed simulation. The ASDA functionality makes it different from all other environments found in the literature. The ASDA has been specified through a modular diagram, composed of seven modules built with the help of the UML (Unified Modelling Language) tool, using three of its diagrams: use case, class and activity. ASDA users can define the distributed simulation by means of the specification of a new simulation program or the replication of a simulation program already developed. If the user chooses to develop a new simulation program, he must only provide the model and the parameters. The environment will then generate the simulation program code using the approach that provides the best performance considering the model and platform characteristics. Besides the specification, this thesis presents a prototype of the ASDA environment with the goal of showing its viability. Three modules have been implemented for the prototype, highlighting the Replication module, which uses the MRIP (Multiple Replication in Parallel) approach. Another contribution of this thesis is the definition of a set of guidelines to the utilization of the MRIP approach. The basis to define these procedures was the results obtained with the utilization of the Replication module

The aim of this thesis is to exploit the technological capabilities of computer supported collaborative work (CSCW) in the field of collaborative Modelling and Simulation (M&S). The thesis focuses on addressing two main problems: (i) providing flexible means of consistency management in collaborative M&S, and (ii) the ability of providing platform and application independent services for collaborative M&S. In this work, some CSCW technologies and how some of the concepts can be incorporated in a distributed collaborative M&S environment, have been studied. An environment for component based simulation development and visualization, which provides support for collaborative M&S, has been designed. Some consistency policies that can be used in conjunction with distributed simulation and the High Level Architecture (HLA) have been investigated. Furthermore, the efficient utilization of HLA and XML in combination, as the foundation of a CSCW infrastructure has been proved. Two consistency policies were implemented utilizing HLA, a strict and an optimistic, in the distributed collaborative environment. Their performance was compared to the performance of a totally relaxed policy, in various collaboration situations. / QC 20101222

INCOSE defines Model-Based Systems Engineering (MBSE) as "the formalized application of modeling to support system requirements, design, analysis, verification, and validation activities beginning in the conceptual design phase and continuing throughout development and later life cycle phases." One very important development is the utilization of MBSE to develop distributed and hybrid (discrete-continuous) simulation modeling systems. MBSE can help to describe the systems to be modeled and help make the right decisions and partitions to tame complexity. The ability to embrace conceptual modeling and interoperability techniques during systems specification and design presents a great advantage in distributed and hybrid simulation systems development efforts. Our research is aimed at the definition of a methodological framework that uses MBSE languages, methods and tools for the development of these simulation systems. A model-based composition approach is defined at the initial steps to identify distributed systems interoperability requirements and hybrid simulation systems characteristics. Guidelines are developed to adopt simulation interoperability standards and conceptual modeling techniques using MBSE methods and tools. Domain specific system complexity and behavior can be captured with model-based approaches during the system architecture and functional design requirements definition. MBSE can allow simulation engineers to formally model different aspects of a problem ranging from architectures to corresponding behavioral analysis, to functional decompositions and user requirements (Jobe, 2008).

In this dissertation, we present an efficient algorithm, called P-Pruning algorithm, for data distribution management problem in High Level Architecture. High Level Architecture (HLA) presents a framework for modeling and simulation within the Department of Defense (DoD) and forms the basis of IEEE 1516 standard. The goal of this architecture is to interoperate multiple simulations and facilitate the reuse of simulation components. Data Distribution Management (DDM) is one of the six components in HLA that is responsible for limiting and controlling the data exchanged in a simulation and reducing the processing requirements of federates. DDM is also an important problem in the parallel and distributed computing domain, especially in large-scale distributed modeling and simulation applications, where control on data exchange among the simulated entities is required. We present a performance-evaluation simulation study of the P-Pruning algorithm against three techniques: region-matching, fixed-grid, and dynamic-grid DDM algorithms. The P-Pruning algorithm is faster than region-matching, fixed-grid, and dynamic-grid DDM algorithms as it avoid the quadratic computation step involved in other algorithms. The simulation results show that the P-Pruning DDM algorithm uses memory at run-time more efficiently and requires less number of multicast groups as compared to the three algorithms. To increase the scalability of P-Pruning algorithm, we develop a resource-efficient enhancement for the P-Pruning algorithm. We also present a performance evaluation study of this resource-efficient algorithm in a memory-constraint environment. The Memory-Constraint P-Pruning algorithm deploys I/O efficient data-structures for optimized memory access at run-time. The simulation results show that the Memory-Constraint P-Pruning DDM algorithm is faster than the P-Pruning algorithm and utilizes memory at run-time more efficiently. It is suitable for high performance distributed simulation applications as it improves the scalability of the P-Pruning algorithm by several order in terms of number of federates. We analyze the computation complexity of the P-Pruning algorithm using average-case analysis. We have also extended the P-Pruning algorithm to three-dimensional routing space. In addition, we present the P-Pruning algorithm for dynamic conditions where the distribution of federated is changing at run-time. The dynamic P-Pruning algorithm investigates the changes among federates regions and rebuilds all the affected multicast groups. We have also integrated the P-Pruning algorithm with FDK, an implementation of the HLA architecture. The integration involves the design and implementation of the communicator module for mapping federate interest regions. We provide a modular overview of P-Pruning algorithm components and describe the functional flow for creating multicast groups during simulation. We investigate the deficiencies in DDM implementation under FDK and suggest an approach to overcome them using P-Pruning algorithm. We have enhanced FDK from its existing HLA 1.3 specification by using IEEE 1516 standard for DDM implementation. We provide the system setup instructions and communication routines for running the integrated on a network of machines. We also describe implementation details involved in integration of P-Pruning algorithm with FDK and provide results of our experiences.

The increasing scale and complexity of network applications and protocols motivates the need for tools to aid in the understanding of network dynamics at similarly large scales. While current network simulation tools achieve large scale modeling, they do so by ignoring much of the intra-program state that plays an important role in the overall system's behavior. This work presents The Distributed Open Network Emulator, a scalable distributed network model that incorporates application program state to achieve high fidelity modeling. The Distributed Open Network Emulator, or DONE for short, is a parallel and distributed network simulation-emulation hybrid that achieves both scalability and the capability to run existing application code with minimal modification. These goals are accomplished through the use of a protocol stack extracted from the Linux kernel, a new programming model based on C, and a scaled real-time method for distributed synchronization. One of the primary challenges in the development of DONE was in reconciling the opposing requirements of emulation and simulation. Emulated code directly executes in real-time which progresses autonomously. In contrast, simulation models are forced ahead by the execution of events, an explicitly controlled mechanism. Relativistic time is used to integrate these two paradigms into a single model while providing efficient distributed synchronization. To demonstrate that the model provides the desired traits, a series of experiments are described. They show that DONE can provide super-linear speedup on small clusters, nearly linear speedup on moderate sized clusters, and accurate results when tuned appropriately. / Master of Science

As tecnologias da informação, telecomunicações e mobilidade aliadas às mudanças econômicas e sociais acarretaram uma grande reestruturação da indústria. Entre estas mudanças, verifica-se um maior nível de descentralização e especialização das unidades produtivas, o aumento da automação dos processos e, em conseqüência, uma maior quantidade e complexidade nas interações de seus sub-sistemas. De forma a lidar com esta complexidade e facilitar o estudo e projeto de novos sistemas, faz-se necessário o uso de modelos, que são analisados por exemplo, através de simulação. Entre elas destaca-se a simulação distribuída, a qual trata da evolução de situações/cenários do sistema em computadores fisicamente dispersos, conectados através de uma rede de comunicação, visando, por exemplo, a redução do tempo de simulação, a simulação de grandes modelos(composto por muitos elementos), maior tolerânica a falhas e mesmo a distribuição geográfica. Visando contribuir para uma maior aplicabilidade das técnicas de modelagem e simulação, em especial a distribuída, propõe-se nesta tese (1) um novo método para a modelagem hierárquica de sistemas produtivos; e (2) um novo algoritmo para a sincronização da evolução do tempo de simulação de diferentes simuladores interagindo através de redes de comunicação (LAN, WAN). No método de modelagem proposto, utiliza-se uma abordagem top-down para a decomposição do sistema, partindo-se de um nível de abstração para um de maior detalhamento, permitindo, assim, um maior nível de conhecimento quanto ao comportamento dos elementos e suas interações. No nível de detalhamento desejado, utiliza-se a Rede de Petri na modelagem dos elementos básicos do sistema, que são, assim como na orientação a objetos, denominados classes. Em seguida, através de uma abordagem bottom-up, estes modelos são agrupados, formando modelos mais complexos: componentes e aplicativos. A fim de garantir a interação entre estes elementos, foram definidos um conjunto de interfaces, bem como suas regras de relacionamento. Este método foi aplicado a um estudo de caso para comprovar sua eficácia. No que diz respeito ao algoritmo proposto para sincronizar os tempos de simulação, utiliza-se como subsidio o mecanismo de gerenciamento da transmissão de dados em redes conhecido como Token Ring. Um simulador de eventos, distribuído, foi implementado com a finalidade de validar o algoritmo proposto. / Evolution on the information technology, telecommunications and transport systems, associated to social and economic changes around the world have caused a significative reorganization of the industry. In this context, a high level of decentralization and specialization of the productive units, as well as an increment of the automation level used in productive processes have been verified. It results on the increase of the amount and the complexity of the enterprise subsystems interactions. Modeling techniques are used with simulation to deal with the complexity, to analysis, and to design new productive systems. Among the simulation approaches, distributed simulation is distinguished. It deals with the execution of simulation in physically dispersed computers connected through a LAN (Local Area Network), providing, for example, the reduction of the simulation time, huge simulation models (models with many elements), fault tolerance, as well as geographic dispersion. To contribute for the evolution of modeling and simulation techniques, in special the distributed one, it is proposed on this work: (1) a new method for the hierarchical modeling of productive systems; and (2) a new time synchronization algorithm used to manage the time evolution of a set of distributed simulation software. On the proposed modeling method it is used a top-down approach to decompose the system in basic elements, starting in a high-level abstraction model, and ending with a set of basic models with high level of detail. Then, these models are modeled using Petri net. As well as on object-oriented languages, each model is called class. After that, using a bottom-up approach, these basic models are grouped to generate more complex models: component and application. A set of interfaces, as well as its relationship rules had been defined to guarantee the interaction among these elements. This method was applied to a case study to confirm its effectiveness. About the time synchronization algorithm, the token ring protocol is used as subsidy. An event based distributed simulator was implemented with the purpose to validate the proposed algorithm.

As tecnologias da informação, telecomunicações e mobilidade aliadas às mudanças econômicas e sociais acarretaram uma grande reestruturação da indústria. Entre estas mudanças, verifica-se um maior nível de descentralização e especialização das unidades produtivas, o aumento da automação dos processos e, em conseqüência, uma maior quantidade e complexidade nas interações de seus sub-sistemas. De forma a lidar com esta complexidade e facilitar o estudo e projeto de novos sistemas, faz-se necessário o uso de modelos, que são analisados por exemplo, através de simulação. Entre elas destaca-se a simulação distribuída, a qual trata da evolução de situações/cenários do sistema em computadores fisicamente dispersos, conectados através de uma rede de comunicação, visando, por exemplo, a redução do tempo de simulação, a simulação de grandes modelos(composto por muitos elementos), maior tolerânica a falhas e mesmo a distribuição geográfica. Visando contribuir para uma maior aplicabilidade das técnicas de modelagem e simulação, em especial a distribuída, propõe-se nesta tese (1) um novo método para a modelagem hierárquica de sistemas produtivos; e (2) um novo algoritmo para a sincronização da evolução do tempo de simulação de diferentes simuladores interagindo através de redes de comunicação (LAN, WAN). No método de modelagem proposto, utiliza-se uma abordagem top-down para a decomposição do sistema, partindo-se de um nível de abstração para um de maior detalhamento, permitindo, assim, um maior nível de conhecimento quanto ao comportamento dos elementos e suas interações. No nível de detalhamento desejado, utiliza-se a Rede de Petri na modelagem dos elementos básicos do sistema, que são, assim como na orientação a objetos, denominados classes. Em seguida, através de uma abordagem bottom-up, estes modelos são agrupados, formando modelos mais complexos: componentes e aplicativos. A fim de garantir a interação entre estes elementos, foram definidos um conjunto de interfaces, bem como suas regras de relacionamento. Este método foi aplicado a um estudo de caso para comprovar sua eficácia. No que diz respeito ao algoritmo proposto para sincronizar os tempos de simulação, utiliza-se como subsidio o mecanismo de gerenciamento da transmissão de dados em redes conhecido como Token Ring. Um simulador de eventos, distribuído, foi implementado com a finalidade de validar o algoritmo proposto. / Evolution on the information technology, telecommunications and transport systems, associated to social and economic changes around the world have caused a significative reorganization of the industry. In this context, a high level of decentralization and specialization of the productive units, as well as an increment of the automation level used in productive processes have been verified. It results on the increase of the amount and the complexity of the enterprise subsystems interactions. Modeling techniques are used with simulation to deal with the complexity, to analysis, and to design new productive systems. Among the simulation approaches, distributed simulation is distinguished. It deals with the execution of simulation in physically dispersed computers connected through a LAN (Local Area Network), providing, for example, the reduction of the simulation time, huge simulation models (models with many elements), fault tolerance, as well as geographic dispersion. To contribute for the evolution of modeling and simulation techniques, in special the distributed one, it is proposed on this work: (1) a new method for the hierarchical modeling of productive systems; and (2) a new time synchronization algorithm used to manage the time evolution of a set of distributed simulation software. On the proposed modeling method it is used a top-down approach to decompose the system in basic elements, starting in a high-level abstraction model, and ending with a set of basic models with high level of detail. Then, these models are modeled using Petri net. As well as on object-oriented languages, each model is called class. After that, using a bottom-up approach, these basic models are grouped to generate more complex models: component and application. A set of interfaces, as well as its relationship rules had been defined to guarantee the interaction among these elements. This method was applied to a case study to confirm its effectiveness. About the time synchronization algorithm, the token ring protocol is used as subsidy. An event based distributed simulator was implemented with the purpose to validate the proposed algorithm.

A vast array of computer-based simulation tools are used to support engineering design and analysis activities. Several such activities call for the simulation of various coupled sub-systems in parallel, typically to study the emergent behavior of large, complex systems. Most sub-systems have their own simulation models associated with them, which need to interoperate with each other in a federated fashion to simulate system-level behavior. The run-time exchange of information between federate simulations requires a common information model that defines the representation of simulation concepts shared between federates. However, most federate simulations employ disparate representations of shared concepts. Therefore, it is often necessary to implement transformation stubs that convert concepts between their common representation to those used in federate simulations. The tasks of defining a common representation for shared simulation concepts and building translation stubs around them adds to the cost of performing a system-level simulation. In this thesis, a framework to support automation and reuse in the process of achieving interoperability between federate simulations is developed. This framework uses ontologies as a means to capture the semantics of different simulation concepts shared in a federation in a formal, reusable fashion. Using these semantics, a common representation for shared simulation entities, and a corresponding set of transformation stubs to convert entities from their federate to common representations (and vice-versa) are derived automatically. As a foundation to this framework, a schema to enable the capture of simulation concepts in an ontology is specified. Also, a graph-based algorithm is developed to extract the appropriate common information model and transformation procedures between federate and common simulation entities. As a proof of concept, this framework is applied to support the development of a federated air traffic simulation. To progress with the design of an airport, the combined operation of its individual systems (air traffic control, ground traffic control, and ground-based aircraft services) in handling varying volumes of aircraft traffic is to be studied. To do so, the individual simulation models corresponding to the different sub-systems of the airport need to be federated, for which the ontology-based framework is applied.

In this dissertation, routing protocols, load-balancing protocols, and efficient evaluation techniques for multi-hop mobile wireless networks are explored. With the advancements made in wireless communication and computer technologies, a new type of mobile wireless network, known as a mobile ad hoc network (MANET), has drawn constant attention. In recent years, several routing protocols for MANETs have been proposed. However, there still remains the need for mechanisms for better scalability support with respect to network size, traffic volume, and mobility. To address this issue, a new method for multi-hop routing in MANETs called Dynamic NIx-Vector Routing (DNVR) is proposed. DNVR has several distinct features compared to other existing on-demand routing protocols, which lead to more stable routes and better scalability. Currently, ad hoc routing protocols lack load-balancing capabilities. Therefore they often fail to provide good service quality, especially in the presence of a large volume of network traffic since the network load concentrates on some nodes, resulting in a highly congested environment. To address this issue, a novel load-balancing technique for ad hoc on-demand routing protocols is proposed. The new method is simple but very effective in achieving load balance and congestion alleviation. In addition, it operates in a completely distributed fashion. To evaluate and verify wireless network protocols effectively, especially to test their scalability properties, scalable and efficient network simulation methods are required. Usually simulation of such large-scale wireless networks needs a long execution time and requires a large amount of computing resources such as powerful CPUs and memory. Traditionally, to cope with this problem, parallel network simulation techniques with parallel computing capabilities have been considered. This dissertation explores a different type of method, which is efficient and can be achieved with a sequential simulation, as well as a parallel and distributed technique for large-scale mobile wireless networks.

O uso de ferramentas de simulação para validar projetos de sistemas digitais é uma prática comum, devido às vantagens que estas trazem ao desenvolvimento destes sistemas, tais como: custo, segurança, velocidade e acuracidade. Porém, a simulação seqüencial de alguns sistemas pode levar várias horas ou até mesmo dias, fazendo desta maneira surgir a necessidade de técnicas para acelerar tal procedimento. Uma solução encontrada para aumentar a velocidade de simulação pode estar no uso de técnicas de sistemas distribuídos, já que muitas vezes o próprio sistema real tem embutido em si um certo paralelismo, o que facilita os procedimentos de distribuição. Ao se tratar da simulação de sistemas distribuídos logo surge um dos grandes problemas inerentes a estes, o controle global do tempo, fazendo com que a sincronização entre os processos seja bastante complicada. Neste trabalho são estudados dois paradigmas de sincronização, o otimista e o conservativo. Tendo como base estes paradigmas, formularam-se duas técnicas para solucionar o problema de sincronização, no contexto da simulação multinível de sistemas digitais. Nos estudos realizados, utilizou-se como plataforma a API WinSock para Windows a fim de proporcionar a comunicação entre processos. Ao final é feita uma análise comparativa das versões desenvolvidas, as quais fizeram uso das técnicas de sincronização acima mencionadas. / The use of simulation tools to validate the design of digital systems is a common practice, due to the benefits these tools bring to the development of those systems: cost, security, velocity, and accuracy. However, the sequential simulation of some systems may take hours or even days, thus creating the need of techniques for speeding up this procedure. A solution for increasing the simulation speed may be the use of techniques based on distributed systems, since very often the real system has an implicit parallelism, which makes easier the aplication of distribution procedures. When dealing with the simulation of distributed systems, one of the big problems that arise is the global control of simulation time, which makes the synchronization among processes very complex. In this work two synchronization paradigms are studied: the optimist and the conservative ones. Based on these paradigms, two techniques for solving the problem of synchronization in the context of multi-level simulation of digital systems have been developed. In these studies, the API WinSock for Windows has been used for supporting the communication between processes. A comparative analysis of the versions we developed, that use the above mentioned synchronization techniques, is also presented.