Detalhes bibliográficos
Ano de defesa: |
2017 |
Autor(a) principal: |
Santos, Gabriel Brito dos |
Orientador(a): |
Não Informado pela instituição |
Banca de defesa: |
Não Informado pela instituição |
Tipo de documento: |
Dissertação
|
Tipo de acesso: |
Acesso aberto |
Idioma: |
por |
Instituição de defesa: |
Não Informado pela instituição
|
Programa de Pós-Graduação: |
Não Informado pela instituição
|
Departamento: |
Não Informado pela instituição
|
País: |
Não Informado pela instituição
|
Palavras-chave em Português: |
|
Link de acesso: |
http://www.repositorio.ufc.br/handle/riufc/28930
|
Resumo: |
In spite of the increasing processing power of handheld smart devices, their capacity to perform some tasks is always a few steps behind their contemporary desktop counterparts. Besides, mobile devices have limited power supplies, which leads software designers to always keep energy consumption in mind when dealing with such devices. An alternative to help overcome this issue is using the offloading technique, which allows a mobile device to offload an expensive task to another device, for the sake of performance or energy saving. This second device may be a remote server hosted in a public cloud, or in the same Wi-fi network as the first mobile device. Facing this problem, this dissertation presents CAOS D2D, a proposal for a framework which allows for a mobile device to offload tasks to other mobile devices, as well as acting as an offloading server too. The prototype implementation is based on the CAOS framework, aiming to extend it in a way that its elements were embedded in a mobile device.In order to evaluate the solution, aspects such as the reduction in execution times of applications when performing offloading and improvements of energy consumption were verified, using different Android devices, application runtime reports, and an equipment for in loco measurement of consumption power. In the analyzed scenarios, there were cases in which, in terms of execution times and energy consumption, the execution of tasks in offloading was more advantageous than the local execution. But also, due to factors such as the computational complexity of the task and the volume of data to be processed, there were also situations where the opposite became true. |