Slice como serviço em edge computing e dispositivos IoT com restrição de recursos

• Ano de defesa: 2021
• Autor(a) principal: Azevedo, Rodrigo Ferraz
• Orientador(a): Verdi, Fábio Luciano
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus Sorocaba
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC-So
• Palavras-chave em Português: Cloud computing; Edge computing; IoT; Slice; NECOS; Microsserviços; Cloud network slice; Microservices
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/14835

A new concept that describes a federated framework that provides computing (processing), networking, and storage as a virtual slice framework, Cloud Network Slicing (CNS) enables new approaches to IoT applications and structuring resources at the edge of the network. Some characteristics of IoT devices, such as intermittent connection, battery power, low processing power, and other resource constraints, require new strategies to meet this new application profile. This research adopts the architecture defined in the NECOS Project and implements the functions for creating CNS in resource-constrained edge devices. The implementation was evaluated in Single Board Computer (SBC) hardware using lightweight virtualization solutions (microservices) and the results achieved confirmed the hypothesis of slice support in these devices with restricted resources, but with limitations to support multiple slices. Therefore, this work presents a brief study on the limitations of the NECOS project in resource-constrained environments, mainly in the IoT context, proposes and implements the extension of some NECOS components to support resource-constrained equipment, then analyzes the performance of the proposed on real IoT hardware using SBC.