High-performance Parallel Solver for Integral Equations of Electromagnetics Based on Galerkin Method
For researchers in computational electromagnetics, this solver offers a memory-efficient and scalable solution for high-contrast media problems, though it is an incremental improvement over existing integral equation methods.
The paper presents a parallel solver for volumetric integral equations in electromagnetics using the Galerkin method, achieving 8x lower memory usage than analogous algorithms and perfect scalability on various hardware platforms, validated on a magnetotelluric sounding problem.
A new parallel solver for the volumetric integral equations (IE) of electrodynamics is presented. The solver is based on the Galerkin method which ensures the convergent numerical solution. The main features include: (i) the memory usage is 8 times lower, compared to analogous IE based algorithms, without additional restriction on the background media; (ii) accurate and stable method to compute matrix coefficients corresponding to the IE; (iii) high degree of parallelism. The solver's computational efficiency is shown on a problem of magnetotelluric sounding of the high conductivity contrast media. A good agreement with the results obtained with the second order finite element method is demonstrated. Due to effective approach to parallelization and distributed data storage the program exhibits perfect scalability on different hardware platforms.