Slides_Electromagnetic_Wave_Propagation_Strategy_3D_MPI

The simulation of physical, mathematical, and geological models is of great importance as it enables the prediction of complex system behaviors. Among the most simulated models are electromagnetic wave propagation, fluid dynamics, finite differences, finite elements, optimization, geothermal processes, and earthquakes, among others. Simulating an electromagnetic wave requires significant amounts of RAM and long execution times. Therefore, this research aims to optimize resource utilization through a computational strategy and asynchronous transfers. First, the 3D electromagnetic wave equation is discretized using the FDTD method. Second, an algorithm for wave propagation is derived from discretization. Third, the first strategy is implemented, involving domain decomposition across two GPUs connected via the PCI Express port using MPI. Fourth, the second strategy is applied, which incorporates asynchronous transfers. The results show that the first strategy improves execution times by up to 32% and enables simulations on domains twice the size allowed by RAM when compared to execution on a single GPU. The second strategy further enhances execution times by 10% compared to the first approach.

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2025-09-24
DOI
https://doi.org/10.5281/zenodo.17196022
Primary Topic
Diverse Scientific and Economic Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Slides_Electromagnetic_Wave_Propagation_Strategy_3D_MPI

Zenodo (CERN European Organization for Nuclear Research)
Diverse Scientific and Economic Studies
article

Slides_Electromagnetic_Wave_Propagation_Strategy_3D_MPI

article en

Abstract

The simulation of physical, mathematical, and geological models is of great importance as it enables the prediction of complex system behaviors. Among the most simulated models are electromagnetic wave propagation, fluid dynamics, finite differences, finite elements, optimization, geothermal processes, and earthquakes, among others. Simulating an electromagnetic wave requires significant amounts of RAM and long execution times. Therefore, this research aims to optimize resource utilization through a computational strategy and asynchronous transfers. First, the 3D electromagnetic wave equation is discretized using the FDTD method. Second, an algorithm for wave propagation is derived from discretization. Third, the first strategy is implemented, involving domain decomposition across two GPUs connected via the PCI Express port using MPI. Fourth, the second strategy is applied, which incorporates asynchronous transfers. The results show that the first strategy improves execution times by up to 32% and enables simulations on domains twice the size allowed by RAM when compared to execution on a single GPU. The second strategy further enhances execution times by 10% compared to the first approach.

Zenodo (CERN European Organization for Nuclear Research)
Universidad Industrial de Santander (CO)
Openalex Percentile: Top 35%
Diverse Scientific and Economic Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.