Uma arquitetura de micro-PaaS em fog computing : orquestração baseada em container para aplicações IoT utilizando single board computer

• Ano de defesa: 2019
• Autor(a) principal: Espírito Santo, Walter do
• Orientador(a): Ribeiro, Admilson de Ribamar Lima
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Programa de Pós-Graduação: Pós-Graduação em Ciência da Computação
• Palavras-chave em Português: Computação; Computação em nuvem; Cluster (sistema de computador); Internet das coisas
• Palavras-chave em Inglês: Fog computing; Orchestration; Cluster; Container; Single board computing
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://ri.ufs.br/jspui/handle/riufs/18340

The Internet of Things (IoT) is an emerging technology paradigm in which ubiquitous sensors monitor, physical infrastructure, environments and people in real time to make decisions that improve systems efficiency and reliability while adding comfort and quality of life to society. IoT data storage and processing are commonly performed on cloud computing infrastructures, but issues such as computational resource constraints, high latency, and different quality of service (QoS) requirements related to IoT devices that eventually move cloud technologies around. towards fog computing, and the adoption of light virtualization solutions such as container-based technologies to address the diverse needs of different domains. The present study aims to propose and implement a micro Platform as a Service (micro-PaaS) architecture in fog computing, in a single board computer (SBC) cluster. The proposed architecture encompasses application orchestration using containers, applied to the Internet of Things and meeting QoS criteria such as high availability, scalability, load balancing and latency. Based on the proposed model, the micro-PaaS Fog was implemented with container virtualization technology using orchestration services in a cluster of Raspberry Pi devices for intelligent monitoring of water and energy consumption at Sergipe Campus Lagarto Federal Institute focal points. . The results showed that it is possible to implement a performance micro-PaaS Fog at a total cost of ownership (TCO) equivalent to 23% of a public platform as a service (PaaS), 26m average mote-fog latency and average recovery time. 1.33 second failures taking into account a confidence interval (CI) of 95%. This research also contributed to a systematic mapping to identify key features and requirements for efficient orchestration in fog computing environments. Through a comparative study, the following characteristics were identified as being the most commonly addressed in this area: heterogeneity, QoS management, scalability, mobility, federation, and interoperability.