Um modelo auto-adaptativo para apoio ao offloading dinâmico em aplicações móveis

• Ano de defesa: 2018
• Autor(a) principal: Nakahara, Flávio Akira
• Orientador(a): Beder, Delano Medeiros
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• Palavras-chave em Português: Computação móvel nas nuvens; Suporte à tomada de decisão; Ciência de contexto
• Palavras-chave em Inglês: Mobile cloud computing; Decision taking support; Context awareness
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/10113

Mobile cloud computing is one of the main ways to augment the performance of resource-constrained mobile devices, bringing resources and services from computationally powerful remote servers in order to provide support to the execution of rich mobile applications. However, an efficient and intelligent use of cloud resources is required due to changing environment conditions and application variability usage. This dissertation presents CoSMOS - Context-Sensitive Model for Offloading System - a context-aware and self-adaptive offloading decision support model for mobile cloud computing systems, based on self-aware and self-expressive system architecture patterns. It employs decision-taking estimation based on application's time execution and energy consumption to decide efficiently when and which application methods should be offloaded in order to improve system's execution. Two practical study cases were used to evaluate the model's approach performance: a N-queen problem application, and MpOS's BenchImage. The results shown that the model is capable of inferring appropriate decisions with acceptable performance in a range of environment conditions.