Técnicas energeticamente eficientes para o posicionamento de aplicações em computação na borda

• Ano de defesa: 2022
• Autor(a) principal: Rios, Rafael Machado
• Orientador(a): Ferreto, Tiago Coelho
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Pontifícia Universidade Católica do Rio Grande do Sul
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação
• Departamento: Escola Politécnica
• País: Brasil
• Palavras-chave em Português: Computação na Borda; Eficiência Energética; Posicionamento de Aplicações
• Palavras-chave em Inglês: Edge Computing; Energy Efficiency; Application Placement
• Área do conhecimento CNPq: CIENCIA DA COMPUTACAO::TEORIA DA COMPUTACAO
• Link de acesso: https://tede2.pucrs.br/tede2/handle/tede/10962

Information and Communication Technology has become an important sector in the global economy and the tendency is that it keeps growing in the foreseeable future. This sector utilizes devices whose functioning wastes energy and produces heat, which is problematic for environmental, socioeconomical and technological reasons. Techniques for reducing energy consumption of such devices typically involve shutting them off or placing them in low-power states, which may impact overall system performance. In the Edge Computing paradigm, however, applications usually have strict requirements such as low latency and low response time. This work presents two application placement techniques with the objective of balancing the tradeoff between energy efficiency and system responsiveness. The first technique involves the creation of a priority-based scheduling algorithm with node-availability information, while the second utilizes a deployment-SLA and node availability-driven scheduler using a multi-objective genetic algorithm. Results show the effectiveness of both application placement techniques in reducing energy consumption while minimizing Service-Level Agreement violations.