Open AccessGPU-accelerated fast implementation of shortest path algorithm in the noise simulation analysis system
Author(s) -
Yuxiao Qiu,
Anxin Zou,
Pengcheng Chen,
Luwen Xu
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/768/7/072035
Subject(s) - computer science , noise (video) , graphics processing unit , computation , algorithm , central processing unit , graphics , convergence (economics) , computational science , point (geometry) , matrix (chemical analysis) , process (computing) , parallel computing , computer graphics (images) , mathematics , artificial intelligence , computer hardware , geometry , economics , image (mathematics) , economic growth , materials science , composite material , operating system
A graphics processing unit (GPU) framework for the computation of the noise level between the noise source and the receiving point in the noise simulation analysis system is presented. The calculation process of the noise level often encounters a problem of slow convergence, and the calculation amount is large due to problems such as weighted point diffusion, transformation of the potential energy matrix, and gradient degradation. To circumvent these limitations, we devise a GPU-accelerated algorithm to calculate the shortest distance between the noise source and the receiving point, which has been shown to perform 11 times faster than the CPU method. Compare test results between GPU method and CPU method in different mesh density scenarios while maintaining their same calculation accuracy.