Hybrid MPI-OpenMP -VALU Parallelization to Accelerate the Solution of VSIE

Pengyuan Wang, Mang He, Haiwen Ke

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The vector arithmetic logic unit (VALU) in modern CPU can process four data concurrently. A hybrid parallelization that combining VALU and MPI-OpenMP is proposed to accelerate the solution of volume-surface integral equation (VSIE) by multilevel fast multipole algorithm (MLFMA) in this paper. And a new sparse approximate inverse (SAI) preconditioner is used to reduce the iterative steps.

Original languageEnglish
Title of host publication2022 IEEE 10th Asia-Pacific Conference on Antennas and Propagation, APCAP 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665489546
DOIs
Publication statusPublished - 2022
Event10th IEEE Asia-Pacific Conference on Antennas and Propagation, APCAP 2022 - Xiamen, China
Duration: 4 Nov 20227 Nov 2022

Publication series

Name2022 IEEE 10th Asia-Pacific Conference on Antennas and Propagation, APCAP 2022 - Proceedings

Conference

Conference10th IEEE Asia-Pacific Conference on Antennas and Propagation, APCAP 2022
Country/TerritoryChina
CityXiamen
Period4/11/227/11/22

Keywords

  • MLFMA
  • SAI
  • VALU
  • VSIE
  • accelerate

Fingerprint

Dive into the research topics of 'Hybrid MPI-OpenMP -VALU Parallelization to Accelerate the Solution of VSIE'. Together they form a unique fingerprint.

Cite this