COMPUTATIONAL ASPECTS OF SOLVING GRID EQUATIONS ON GRAPHICS ACCELERATORS
Abstract
To predict emergencies and irreversible consequences of human activities, scientists use
mathematical modeling. When an emergency occurs, it is very important to minimize the decisionmaking
time. The development of the project solution can be based on the forecast of changes in
the modeled process. In the numerical solution of problems of hydrophysics and biological kinetics,
it becomes necessary to develop effective methods for solving systemic equations of large dimension
with a non-self-adjoint operator. The large volume of processed information and the
complexity of computations necessitate the use of computational clusters, which include video
adapters to increase the performance of the computing system and the speed of information processing.
The aim of the research is to develop a solution for a module that implements the algorithm
of the system of linear algebraic equations (SLAE) by the modified alternative triangular
iterative method (MATM) (self-adjoint and non-self-adjoint case) using NVIDIA CUDA technology.
A method for decomposition of the computational domain in a three-dimensional case is described.
A graph model of a parallel pipeline computational process is proposed, focused on the
GPU (Graphics Processing Unit). To determine the two-dimensional configuration of flows in the
computational unit, when performing one step of one step, the MATM is minimal. The studies have
shown that the choice of the method of decomposition of the computational domain in the form of
parallelepipeds must be performed taking into account the architecture of the video adapter. The
developed algorithm and software module make it possible to more effectively use the computational
resources of the GPU used to solve computationally laborious problems of hydrophysics.
References
General Meshes. Springer International Publishing, 2020, 186 p.
2. Dymnikov V.P., Tyrtyshnikov E.E., Lykosov V.N., Zalesnyy V.B. Matematicheskoe
modelirovanie klimata, dinamiki atmosfery i okeana: K 95-letiyu G.I. Marchuka i 40-letiyu
IVM RAN [Mathematical modeling of climate, atmospheric and ocean dynamics: To the 95th
anniversary of G.I. Marchuk and the 40th anniversary of the IVM RAS], Izvestiya RAN. Fizika
atmosfery i okeana [News of the Russian Academy of Sciences. Physics of the atmosphere and
ocean], 2020, Vol. 56, No. 3, pp. 251-254.
3. Chetverushkin B.N., Mingalev I.V., Chechetkin V.M., Orlov K.G., Fedotova E.A., Mingalev
V.S., Mingalev O.V. Modeli obshchey tsirkulyatsii atmosfery Zemli. Dostizheniya i napravleniya
razvitiya [Models of the general circulation of the Earth's atmosphere. Achievements
and development goals], Matematicheskoe modelirovanie [Mathematical modeling],
2020, Vol. 32, No. 11, pp. 29-46.
4. Goloviznin V.M., Chetverushkin B.N. Algoritmy novogo pokoleniya v vychislitel'noy
gidrodinamike [New generation algorithms in computational fluid dynamics], Zhurnal
vychislitel'noy matematiki i matematicheskoy fiziki [Journal of Computational Mathematics
and Mathematical Physics], 2018, Vol. 58, No. 8, pp. 20-29.
5. Matishov G.G., Gargopa Yu.M., Berdnikov S.V., Dzhenyuk S.L. Zakonomernosti
ekosistemnykh protsessov v Azovskom more [Regularities of ecosystem processes in the Sea
of Azov]. Moscow: Nauka, 2006, 304 p.
6. Bonaduce A., Staneva J., Grayek S., Bidlot J.-R., Breivik O. Sea-state contributions to sea-level
variability in the European Seas, Ocean Dynamics, 2020, 70 (12), pp. 1547-1569. DOI:
10.1007/s10236-020-01404-1.
7. Marchesiello P., Mc.Williams J.C., Shchepetkin A. Open boundary conditions for long-term
integration of regional oceanic models, Oceanic Modelling Journal. Netherlands: Elsevier BV,
2001, Vol. 3, No. 1-2, pp. 1-20. DOI: 10.1016/s1463-5003(00)00013-5.
8. Androsov A.A., Vol'tsinger N.E. Prolivy mirovogo okeana. Obshchiy podkhod k
modelirovaniyu [Straits of the world ocean. General approach to modeling]. Moscow: Nauka,
2005, 172 p.
9. Nieuwstadt F., Westerweel J., Boersma B.J. Turbulence. Introduction to Theory and Applications
of Turbulent Flows. Springer, 2016, 288 p.
10. Voevodin V.V., Voevodin Vl.V. Parallel'nye vychisleniya [Parallel computing]. Saint
Petersburg: BKhV-Peterburg, 2002, 608 p.
11. Xue W., Roy C.J. Multi-GPU performance optimization of a computational fluid dynamics
code using OpenACC, Concurrency and Computation Practice and Experience, 2020, 33 (4).
DOI: 10.1002/cpe.6036.
12. Xue W., Jackson C.W., Xue W., Roy C.J. Multi-CPU/GPU Parallelization, Optimization and
Machine Learning based autotuning of Structured Grid CFD Codes, AIAA Aerospace Sciences
Meeting, 2018, pp. 0362. DOI: 10.2514/6.2018-0362.
13. Nagatake T., Kunugi T. Application of GPU to Computational Multiphase Fluid Dynamics,
IOP Conference Series: Materials Science and Engineering, 2010, Vol. 10 (1), pp. 012024.
DOI: 10.1088/1757-899X/10/1/012024.
14. Munk D.J., Kipouros T., Vio G.A. Multi-physics bi-directional evolutionary topology optimization
on GPU-architecture, Engineering with Computers, 2019, Vol. 35 (4), pp. 1059-1079.
DOI: 10.1007/s00366-018-0651-1.
15. Sukhinov A.I., Atayan A.M., Belova Yu.V., Litvinov V.N., Nikitina A.V., Chistyakov A.E.
Obrabotka dannykh naturnykh izmereniy ekspeditsionnykh issledovaniy dlya
matematicheskogo modelirovaniya gidrodinamicheskikh protsessov Azovskogo morya [Processing
of data of field measurements of expeditionary research for mathematical modeling of
hydrodynamic processes of the Sea of Azov ], Vychislitel'naya mekhanika sploshnykh sred
[Computational mechanics of continuous media], 2020, Vol. 13, No. 2, pp. 161-174.
16. Sukhinov A.I., Chistyakov A.E., Shishenya A.V., Timofeeva E.F. Predictive Modeling of
Coastal Hydrophysical Processes in Multiple-Processor Systems Based on Explicit Schemes,
Mathematical Models and Computer Simulations, 2018, 10 (5), pp. 648-658. DOI:
10.1134/S2070048218050125.
17. Konovalov A.N. Metod skoreyshego spuska s adaptivnym poperemennotreugol'nym
pereobuslovlivatelem [The method of rapid descent with an adaptive alternating triangular reconditionalist
], Differentsial'nye uravneniya [Differential equations], 2004, Vol. 40, No. 7,
pp. 953-963.
18. Sukhinov A.I., Chistyakov A.E., Litvinov V.N., Nikitina A.V., Belova Yu.V., Filina A.A.
Vychislitel'nye aspekty matematicheskogo modelirovaniya gidrobiologicheskikh protsessov v
melkovodnom vodoeme [Computational aspects of mathematical modeling of hydrobiological
processes in a shallow reservoir], Vychislitel'nye metody i programmirovanie [Computational
methods and programming], 2020, Vol. 21, No. 4, pp. 452-469. DOI: https://doi.org/10.26089/
NumMet.v21r436.
19. Samarskiy A.A., Vabishchevich P.N. Chislennye metody resheniya zadach konvektsii-diffuzii
[Numerical methods for solving convection-diffusion problems. Stereotype]. Mosxow:
Knizhnyy d m «LIBROKOM», 2015, 248 p.
20. Oyarzun G., Borrell R., Gorobets A., Oliva A. MPI-CUDA sparse matrix–vector multiplication
for the conjugate gradient method with an approximate inverse preconditioner, Computers and
Fluids, 2014, Vol. 92, pp. 244-252. DOI: 10.1016/j.compfluid.2013.10.035.
21. Zheng Liang, Gerya Taras, Knepley Matthew, Yuen David, Zhang Huai, Shi Yaolin. Implementation
of a multigrid solver on a GPU for Stokes equations with strongly variable viscosity
based on Matlab and CUDA, International Journal of High Performance Computing Applications,
2014, 28 (1), pp. 50-60. DOI: 10.1007/978-3-642-16405-7_21.
