Code optimization by using GPU applied to a Dataflow numerical simulation model

15th Brazilian Symposium on High Performance Computational Systems

计算机科学

Evangelista, Luiz E.S.^1 ; Fazenda, Alvaro L.^1 ; De Melo, Vincius V^1

Science and Technology Institute, Federal University of Sao Paulo (UNIFESP), São José Dos Campos, Avenida Cesare Mansueto Giulio Lattes 1201, 12247-014, Brazil^1

关键词: Coarse grains;    Code instrumentation;    Code optimization;    Computational codes;    Dataflow;    GPU-accelerated;    Original algorithms;    Performance Gain;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/649/1/012003/pdf
DOI  :  10.1088/1742-6596/649/1/012003

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This work aims to study techniques for parallel computing using GPU (Graphics Processing Unit) in order to optimize the performance of a fragment of computational code, implemented as a Dataflow system, which is part of a meteorological numerical model responsible for calculating the advection transportation phenomena. The possible algorithm limitations for GPU efficiency will also be addressed through an extensive code instrumentation. Considering the difficulties found on the original algorithm which implies a GPU accelerated code dealing with flow dependencies and coarse grain parallelism, the performance gain with GPU may be considered fair.