Dissertação de Otávio Moraes de Carvalho

Contato

Detalhes do Evento

Date: 17/12/2018 14:00 – 16:00
Categorias: Defesas, Dissertação de Mestrado

Aluno: Otávio Moraes de Carvalho
Orientador: Prof. Dr. Philippe Olivier Alexandre Navaux

Título: A model for distributed data aggregation in edge and cloud environments
Linha de Pesquisa: Computação de Alto Desempenho e Sistemas Distribuídos

Data: 17/12/2018
Hora: 14h
Local: Prédio 43412 – AUD-1 (auditório 1) do Instituto de Informática.

Banca Examinadora:
-Prof. Dr. Lisandro Zambenedetti Granville (UFRGS)
-Prof. Dr. Tiago Coelho Ferreto (PUCRS)
-Prof. Dr. Rodrigo da Rosa Righi (UNISINOS)

Presidente da Banca: Prof. Dr. Philippe Olivier Alexandre Navaux

Abstract: The Internet of Things generates each time larger amounts of data through devices that range from personal smartphones sensors to large scale smart cities sensor deployments. Edge Computing and Fog Computing hardware resources nowadays have processing power that in the past was only found on Cloud Computing resources. These capabilities enable the possibility of aggregating data directly on the edge of the network, avoiding network traffic that was previously necessary to transfer data to the cloud. However, the ability to leverage each time more powerful edge nodes to decentralize data processing and aggregation is still a significant challenge for both industry and academia. Although large scale sensor deployments and powerful edge nodes are now available, the proper frameworks and processing patterns for these profiles of data are still under active research. In this work, it is analyzed the impact of a model for data aggregation in a large scale smart grid application dataset. The results obtained show that the implemented testbed application, through the usage of edge node aggregation and messaging windows, was able to achieve data aggregation rates of above 400 million measurements per second. These results were obtained using machines on 15 distinct geographic regions on the Microsoft Azure platform, for a total of 1366 machines in the largest evaluation scenario.

Keywords: Edge Computing, Fog Computing, Cloud Computing, Internet of Things, Distributed Computing.