Detalhes bibliográficos
Ano de defesa: |
2016 |
Autor(a) principal: |
SANTANA, Nielson Avelino de
 |
Orientador(a): |
LINS, Fernando Antonio Aires |
Banca de defesa: |
CALLOU, Gustavo Rau de Almeida,
CYSNEIROS FILHO, Gilberto Amado de Azevedo,
DINIZ, Juliana Regueira Basto |
Tipo de documento: |
Dissertação
|
Tipo de acesso: |
Acesso aberto |
Idioma: |
por |
Instituição de defesa: |
Universidade Federal Rural de Pernambuco
|
Programa de Pós-Graduação: |
Programa de Pós-Graduação em Informática Aplicada
|
Departamento: |
Departamento de Estatística e Informática
|
País: |
Brasil
|
Palavras-chave em Português: |
|
Área do conhecimento CNPq: |
|
Link de acesso: |
http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7877
|
Resumo: |
One of the main drawbacks to the development of mobile applications with compute intensive capabilities, that is, applications based on technologies such as natural language processing, voice recognition and augmented reality, is the low energy-autonomy of cell phones. In the context of this restricted scenario set in the development of these applications, it rises the mobile cloud computing. The mobile cloud computing is a paradigm for mobile applications that extends the processing efficiency as either the battery energy life by leveraging the demanding tasks to the cloud. To this end, the proposal of this work is an integrated methodology for performance evaluation of applications in the mobile cloud environment. The proposed methodology is presented in two phases. The first phase presents the performance evaluation technique by measurement. It details the steps which includes the context problem comprehension, metrics collection tasks and the results demonstration. The second phase, in turn, presents the performance evaluation technique by simulation. Since the approaches are linked, the data collected in the first phase (measurement step) will be used as input for the development of the simulation models. Furthermore, the given methodology is applied in a case study, which consisted at the execution of the Linpack benchmark in the device and cloud scenarios, in order to assess whether the cloud can be more performatic over the local execution. The case study was considered also to check if the proposed methodology was adequate in test of a real world experiment. The experiment results showed that the mobile cloud computing execution can be more efficient than the local processing in the mobile environment. Although the promising results, it is still required further studies with other benchmark models besides the mathematical ones approached in this work. Finally, we conclude that the proposed methodology presented is adequate for the case study experiments. |