Modelo Híbrido de Programação Paralela para uma Aplicação de Elasticidade Linear Baseada no Método dos Elementos Finitos

AUTOR(ES)
DATA DE PUBLICAÇÃO

2006

RESUMO

In the area of parallel processing there are two major programming paradigms: Shared Memory and Message Passing. Each of them fits into a specific physical model, but there are multiprocessor architectures whose mapping to one of these paradigms is not so simple. SMP clusters, for example, are built by connecting some shared memory machines through an interconnection network. Applications on SMP clusters can be programmed to use message passing among all processors. However, its possible to achieve better performance using a hybrid model with shared memory communication inside SMP nodes and message passing communication between them. In this work, a hybrid model was used to develop an engineering application based on the Finite Element Method in order to evaluate this model and to compare its performance with a pure message passing version of the same application.

ASSUNTO(S)

memória compartilhada ciencia da computacao programação paralela troca de mensagens método dos elementos finitos sistemas distribuídos

Documentos Relacionados