Kinetic Algorithms for Modeling Conductive Fluids Flow
on High-Performance Computing Systems
RAS Academician B. N. Chetverushkina,*, A. V. Savelievb,c,**, and V. I. Savelievb,***
a Federal Research Center Keldysh Institute of Applied Mathematics, Russian Academy of Sciences, Moscow,
125047 Russia
b Immanuel Kant Baltic Federal University, Kaliningrad, 236041 Russia
c Lomonosov Moscow State University, Moscow, Russia
Correspondence to: *e-mail: office@keldysh.ru
Correspondence to: **e-mail: andrey.saveliev@desy.de
Correspondence to: ***e-mail: valeri.saveliev@desy.de
Received 19 August, 2019
Abstract—Processes in the dynamics of electrically conducting fluid flows in complex heat transfer systems are mathematically modeled in detail on high-performance parallel computing systems. The study is based on the kinetically consistent magnetogasdynamic approach adjusted to this class of problems. The kinetically consistent algorithm is well adapted to the architecture of high-performance computing systems with massive parallelism, so that complex heat transfer systems can be effectively studied with a high resolution. The approach, method, and algorithms are described, and numerical results are presented.
DOI: 10.1134/S1064562419060206