A Set of Methodological Test Programs for the Numerical Research of High-Performance Computing Systems Parameters
https://doi.org/10.15514/ISPRAS-2024-36(2)-9
Abstract
An integral part of the process of creating high-performance computing systems designed to solve problems of numerical modeling of various physical processes is to check their compliance with the characteristics stated during their design. At the same time, there is a problem of evaluating the performance of computing systems on synthetic tests, which are significantly primitive in mathematical complexity to real applied problems. The article considers a set of test programs developed by the authors, which allows to more accurately assess the real performance of computing systems.
About the Authors
Alexey Olegovich IGNATYEVRussian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 1998. Research interests: design of supercomputer systems, parallel numerical simulation programs development, operating systems development, methods and means of information security.
Sergey Yurievich MOKSHIN
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2016. Research interests: design of supercomputer systems, development of functional subsystems for high performance supercomputing systems, operating systems development, methods and means for protecting information.
Alexander Victorovich ERSHOV
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2020. Research interests: Difference methods of solution integrate-differential equations, parallel numerical simulation programs development.
Artem Vladimirovich KARPEEV
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2015. Research interests: numerical methods for solving the equation of continuum mechanics, parallel technologies for modern high-performance computing systems.
Rim Fanavievich MUKHAMADIEV
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2012. Research interests: development of parallel numerical simulation programs.
Elena Mikhailovna ROMANOVA
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2015. Research interests: particle transport equations, development of algorithms for solving the particle transport equation for modern computing platforms.
Denis Alexandrovich USHAKOV
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2015. Research interests: diffusion equations, development of algorithms for solving the diffusion equation for modern computing platforms.
Vadim Olegovich ANISOV
Russian Federation
The scientific assistant of Russian Federal Nuclear Center E. I. Zababakhin «All-Russian Scientific Research Institute of Technical Physics» since 2018. Research interests: diffusion equations, development of algorithms for solving the diffusion equation for modern computing platforms.
References
1. . Игнатьев А.О., Мокшин С.Ю. Типовая архитектура высокопроизводительной вычислительной системы для решения задач численного моделирования, Препринт РФЯЦ-ВНИИТФ № 265, Снежинск, 2020 г., 21 с.
2. . TOP-500. Available at: https://www.top500.org/ accessed 07.05.2024.
3. . HPC Linpack. Available at: https://netlib.sandia.gov/benchmark/hpl/, accessed 07.05.2024.
4. . HPCG. Available at: https://www.hpcg-benchmark.org/, accessed 07.05.2024.
5. . NAS Parallel Benchmarks. Available at: https://www.nas.nasa.gov/software/npb.html, accessed 07.05.2024.
6. . Gropp W., Lusk E., Skjellum A. Using MPI: Portable parallel programming with the message-passing interface. – Second edition. – The MIT Press, 1999
7. . OpenMP. Available at: https://www.openmp.org/, accessed 07.05.2024.
8. . CUDA Toolkit. Available at: https://developer.nvidia.com/cuda-toolkit/, accessed 07.05.2024.
9. . ASC Purple. Available at: https://asc.llnl.gov/computers/historic-decommissioned-machines/purple, accessed 07.05.2024.
10. . High Performance Computing benchmarks. Available at: https://openbenchmarking.org/suite/pts/hpc, accessed 07.05.2024.
11. . Алексеев А.В., Беляев С.П., Бочков А.И. и др. Методические прикладные тесты РФЯЦ-ВНИИЭФ для численного исследования параметров высокопроизводительных вычислительных систем. ВАНТ, сер. Математическое моделирование физических процессов. 2020. Вып. 2., с. 86-100.
12. . ГОСТ Р 57700.18-2019. Высокопроизводительные вычислительные системы. Требования к тестовым программам приемочных испытаний.
13. . Carlson B.G. A method of characteristics and other improvements in solutions methods for the transport equation. Nuclear Science and Engineering. 1976, №3, p. 408-425.
14. . Басс Л.П., Волощенко А.М., Гермогенова Т.А. Методы дискретных ординат в задачах о переносе излучения. Сборник, ИПМ им. М.В. Келдыша АН СССР, 1986.
15. . Владимиров В.С. Математические задачи односкоростной теории переноса частиц. Труды МИАН СССР, 1961.
16. . Корн Г., Корн Т. Справочник по математике для научных работников и инженеров – М.: Наука, 1970.
17. . Лебедев С.Н., Писарев В.Н., Романова Е.М. и др. Параллельные вычисления при решении стационарного уравнения переноса. Международный семинар “Супервычисления и математическое моделирование”, Тезисы, Саров, 2000.
18. . Самарский А.А. Теория разностных схем. – 3-е изд., испр. – М.: Наука, 1989. – 616 с.
19. . Гаджиев А.Д., Писарев В.Н., Рыкованов В.В., Шестаков А.А. Методика и программа для решения двумерного уравнения теплопроводности. ВАНТ. Методики и программы численного решения задач математической физики, 1985, вып. 1, с. 53-65.
20. . Писарев В.Н. О параметрическом семействе схем «РОМБ» для нелинейного уравнения теплопроводности. ВАНТ. Методики и программы численного решения задач математической физики, 1986, вып. 3, с. 44-52.
21. . Модестов Д.Г. Оценка временной постоянной при учете запаздывающих нейтронов. ВАНТ, 2022, вып.1, с 17-26.
22. . Седов Л.И. Методы подобия и размерности в механике. – 8-ое, переработанное изд. – М.: Наука, 1977.
23. . Gentry R.A., Martin R.E., Daly B.J. An eulerian differencing method for unsteady compressible flow problems // Journal of Computational Physics. – 1966. – vol. 1.
24. . Зельдович Я.Б., Райзер Ю.П. Физика ударных волн и высокотемпературных гидродинамических явлений. М. Наука, 1966г, 688стр.
Review
For citations:
IGNATYEV A.O., MOKSHIN S.Yu., ERSHOV A.V., KARPEEV A.V., MUKHAMADIEV R.F., ROMANOVA E.M., USHAKOV D.A., ANISOV V.O. A Set of Methodological Test Programs for the Numerical Research of High-Performance Computing Systems Parameters. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2024;36(2):109-126. (In Russ.) https://doi.org/10.15514/ISPRAS-2024-36(2)-9