Uma metodologia para desenvolver simulações distribuídas HLA usando DSEEP guiado a modelos com OPM e UML

• Ano de defesa: 2024
• Autor(a) principal: Schittler, João Gabriel da Cunha
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Ciência da Computação; UFSM; Programa de Pós-Graduação em Ciência da Computação; Centro de Tecnologia
• Palavras-chave em Português: Simulações distribuídas; Arquitetura orientada a modelos; Transformações entre modelos; Metodologia de objetos e processos; Distributed simulation; Model driven architecture; Model to model transformation; Object-process methodology; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.ufsm.br/handle/1/32633

Distributed simulations aim to recreate some real behavior using computer networks; they are often geographically distributed as well. One widely used approach to develop distributed simulations is to use a pre-established standard for network communication between simulators; one such standard is High-Level Architecture (HLA) (IEEE, 2010). However, the development of source code for such standards is often complex and error-prone, making code generators a good option for its development. Model Driven Architectures (MDA) (OMG, 2023) have been explored to generate efficient source code based on highlevel conceptual models. Nevertheless, using MDA for code generation creates another challenge: the specification and development of accurate high-level models of the simulation, which can be quite difficult if the project stakeholders cannot properly understand the models. This work, centered in HLA project development, explores the use of ObjectProcess Methodology (DORI, 2002) in a methodology for developing distributed simulations with a conceptual modeling step that is natural, understandable to stakeholders and that guarantees automatic transformation from high-level (human-friendly) specifications into HLA source code for the specified simulation. The proposed methodology is aligned to the Distributed Simulation Engineering and Execution Process (DSEEP) and maintains the possibility of using UML as a modeling language. This work also contains a series of experiments that aim to validate the contributions, showcasing that the developed methodology can achieve its objectives and that its use is appropriate for developing distributed simulations.