Parallel processing applied to a materials engineering problem

• Autor(a) principal: Quaresma, Paulo Jorge
• Data de Publicação: 2018
• Outros Autores: Medeiros, Pedro Duarte, Velhinho, Alexandre
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://doi.org/10.34627/rcc.v2i0.67

X-ray tomography is a technique that is being widely used in many areas. In one hand, the available tomographic devices produce massive amounts of data; on the other hand the researchers continually demand more information to be extracted from the raw data. In this work, the starting point was a sequential program developed by G. Vignoles of Bordeaux 1 University that includes several processing steps of the raw tomographic data. In order to reduce the execution time of the longer processing steps, different approaches have been tried like the optimization of some parts of sequential code (mainly those related with I/O operations), and the use of distributed-memory multiprocessors (clusters), where the use of a parallel file system for data sharing between the nodes was combined with a geometrical parallelization approach. In the last approach the MPI-2 library (including the parallel I/O primitives) was used.