Fine-Grained Parallel Solution for Solving Sparse Triangular Systems on Multicore Platform Using OpenMP Interface

Abstract

This paper describes and analyses a novel method to improve the parallel performance for solving sparse triangular systems (spTRSV). The main objective of this study consists in reducing the total idle time of processors as well as the execution time. Also, the developed solution is suitable for sparse and band structures. To evaluate and validate our contribution, a series of experiments have been carried out on a multicore platform using OpenMP interface. Practical results show the efficiency of the proposed technique compared to the previous studies.