Publicado em: 28/07/2015

Dissertação de Mestrado – Emmanuell Diaz Carreño

UNIVERSIDADE FEDERAL DO RIO GRANDE DO SUL
INSTITUTO DE INFORMÁTICA
PROGRAMA DE POS-GRADUAÇÃO EM COMPUTAÇÃO

———————————————————

DEFESA DE DISSERTAÇÃO DE MESTRADO

Aluno: Emmanuell Diaz Carreño
Orientador: Prof. Dr. Philippe Olivier Alexandre Navaux
Título: Migration and Evaluation of a Numerical Weather Prediction Application in a Cloud Computing Infrastructure
Linha de Pesquisa: Processamento Paralelo e Distribuído
Data: 03/08/2015
Hora: 13:30

Local: Prédio 43412 – Sala 215 (sala de videoconferência), Instituto de Informática.

Banca Examinadora:
Prof. Dr. Jairo Panetta (INPE/CPTEC – por videoconferência)
Prof. Dr. Lucas Mello Schnorr (UFRGS)
Prof. Dr. Marco Antonio Zanata Alves (UFRGS)
Presidente da Banca: Prof. Dr. Philippe Olivier Alexandre Navaux

Abstract: The usage of clusters and grids has benefited for years the High Performance Computing (HPC) community. These kind of systems have allowed scientists to use bigger datasets and to perform more intensive computations, helping them into achieving results in less time but has also increased the cost associated to this area of science. As some e-Science projects are carried out also in highly distributed network environments, or using immense data sets that sometimes require grid computing, they are good candidates too for cloud computing initiatives. The Cloud Computing paradigm has emerged as a practical solution to perform large-scale scientific computing. The elasticity of the cloud and its pay-as-you-go model presents an interesting opportunity for applications commonly executed in clusters or supercomputers. In this context, the user does not need to buy infrastructure, the resources can be rented from a provider and used for a period of time. This thesis presents the challenges and solutions of migrating a numerical weather prediction (NWP) application to a cloud computing infrastructure. We performed the migration of this HPC application and evaluate its performance in a local cluster and in the cloud using different instance sizes. We analyze the principal characteristics of the application running in the cloud trying to get the most of it. The experiments demonstrate that, although processing and networking create a limiting factor, storing input and output datasets in the cloud presents an attractive option to share results and ease the deployment of a test-bed for a weather research platform. Results show that cloud infrastructure can be used as a viable HPC alternative for numerical weather prediction software.