GPU-Accelerated Integral Equation Seismic Simulation

Alexey L. Gordov, Nikolay I. Khokhlov, Mikhail S. Malovichko

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

In this chapter, we investigate the use of GPUs to speed up the matrix–vector product within an iterative integral equation (IE) solver. Our implementation is based on CUDA. Numerical experiments suggest that GPU computing can accelerate the performance of the solver by a factor of 70 for the single-precision arithmetic and by a factor of 30 for the double precision. The material presented here can be of interest for iterative finite difference acoustic and electromagnetic solvers based on layered earth Green’s function preconditioners.

Original languageEnglish
Title of host publicationSmart Modelling for Engineering Systems - Proceedings of the International Conference on Computational Methods in Continuum Mechanics CMCM 2021
EditorsMargarita N. Favorskaya, Alena V. Favorskaya, Igor B. Petrov, Lakhmi C. Jain
PublisherSpringer Science and Business Media Deutschland GmbH
Pages283-294
Number of pages12
ISBN (Print)9789813346185
DOIs
Publication statusPublished - 2021

Publication series

NameSmart Innovation, Systems and Technologies
Volume215
ISSN (Print)2190-3018
ISSN (Electronic)2190-3026

Keywords

  • CUDA
  • GPU
  • Seismic acoustic equation

Fingerprint

Dive into the research topics of 'GPU-Accelerated Integral Equation Seismic Simulation'. Together they form a unique fingerprint.

Cite this