Development of Supercomputer Technologies at the Institute of Mathematical Modelling and Keldysh Institute of Applied Mathematics of Russian Academy of Sciences
https://doi.org/10.23947/2587-8999-2024-8-1-12-28
Abstract
A review scientific work in the field of supercomputer technologies at the Institute of Mathematical Modelling and Keldysh Institute of Applied Mathematics is presented. Progress in supercomputer technologies, programming tools and technique (such as hyperbolization, load balancing, fault tolerance, adaptive mesh refinement, rational mesh decomposition) and several supercomputer applications are presented.
Keywords
supercomputers,
supercomputing resource center,
hyperbolic equations,
hyperbolization,
load balancing,
fault tolerance,
dynamically adaptive computational grids,
rational mesh decomposition
Supporting agencies: We would like to express our sincere gratitude to S.V. Polyakov the head of Supercomputing Resource Center at the Keldysh Institute for the valuable discussions and materials provided, as well as to the team of the Center for the opportunity to use the facilities of the Center for carrying out numerous computations mentioned in this publication.
About the Authors
M. V. Yakobovskiy
Keldysh Institute of Applied Mathematics of Russian Academy of Sciences
Russian Federation
Russian Federation
Mikhail V. Yakobovskiy, Deputy Director for Scientific Work
https://www.mathnet.ru/php/person.phtml?option_lang=rus&personid=22428
https://www.scopus.com/authid/detail.uri?authorId=57063571100
4, Miusskaya Sq., Moscow, 125047
M. A. Kornilina
Keldysh Institute of Applied Mathematics of Russian Academy of Sciences
Russian Federation
Russian Federation
Marina A. Kornilina, Researcher
option_lang=rus&personid=32912
https://www.scopus.com/authid/detail.uri?authorId=55117411600
4, Miusskaya Sq., Moscow, 125047
References
1. Top500 Supercomputer Sites. (In Russ.). URL: http://top500.org (accessed: 21.02.2024).
2. Elizarova T.G., Chetverushkin B.N. Application of multiprocessor transputer systems to solve problems of mathematical physics. Mathematical Modeling. 1992;4(11):75–100. (In Russ.).
3. Galushkin A.I., Tochenov V.A. Transputer systems ― the beginning of the formation of mass parallel computers in Russia. Neurocomputers: development, application. 2005;3:17–28. (In Russ.).
4. Transputer Systems – the Beginning of the Formation of Mass Parallel Computers in Russia. Collection of abstracts of reports NSCF’2014. Pereslavl-Zalessky: IPS named after A.K. Aylamazyan RAS; 2014. 45 p. (In Russ.). URL: https://2014.nscf.ru/TesisAll/0_PostMoore_Plenar/01_008_GalushkinAI.pdf (accessed: 21.02.2024).
5. Hybrid Computing Cluster K-100. (In Russ.). URL: http://www.kiam.ru/MVS/resourses/k100.html (accessed: 21.02.2024).
6. Davydov A.A., Latsis A.O., Lutsky A.E., Smolyanov Yu.P., Chetverushkin B.N., Shilnikov E.V. Multiprocessor computing system of hybrid architecture “MVS-Express”. Reports of the Academy of Sciences. 2010;434(4):459–463. (In Russ.).
7. Center for Collective Use of IPM named after M.V. Keldysh RAS. (In Russ.). URL: http://ckp.kiam.ru (accessed: 26.02.2024).
8. Chetverushkin B.N., Churbanova N.G. On the application of the principle of geometric parallelism for the (α−β)-iterative algorithm. Mathematical Modeling. 1991;3(3):123–129. (In Russ.).
9. Abalakin I.V., Chetverushkin. B.N. Kinetic consistent difference schemes as a model for describing gas dynamic flows. Mathematical Modeling. 1996;8(8):17–36. (In Russ.).
10. Chetverushkin B.N. Kinetically-consistent schemes in gas dynamics: a new model of viscous gas, algorithms, parallel implementation, applications. Moscow: Moscow State University Publishing House; 1999. 232 p. (In Russ.).
11. Chetverushkin B.N. Hyperbolic quasi-gas dynamics system. Mathematical Modeling. 2018;30(2):81–98. (In Russ.).
12. Chetverushkin B.N., D’Aschenzo N., Saveliev A.V., Saveliev V.I. Kinetic model and equations of magnetic gas dynamics. Computational Mathematics and Mathematical Physics. 2018;58(5):716–725. (In Russ.).
13. Saveliev V.I., Chetverushkin B.N. Modeling problems of magnetohydrodynamics on high-performance computing systems. Mathematical Modeling. 2020;32(12):3–13. (In Russ.).
14. Hubble Space Telescope Snapshot. (In Russ.). URL: https://apod.nasa.gov/apod/ap000706.html (accessed: 26.02.2024).
15. Cappello F. Fault Tolerance in Petascale/ Exascale Systems: Current Knowledge, Challenges and Research Opportunities. International Journal of High Performance Computing Applications. 2009;23(3):212–226.
16. Chetverushkin B.N., Yakobovsky M.V. Computational algorithms and fault tolerance of hyperexascale computing systems. Reports of the Academy of Sciences. 2017;472(1):1–5. (In Russ.).
17. Chetverushkin B.N., Yakobovsky M.V. Computational algorithms and architecture of high-performance systems. Preprints of the M.V. Keldysh Institute of Applied Mathematics. 2018;52:12. (In Russ.).
18. Chetverushkin B.N., Yakobovsky M.V., Kornilina M.A., Semenova A.V. Numerical Algorithms for HPC Systems and Fault Tolerance Communications. Computer and Information Science. 2019;1063:34. (In Russ.). https://doi.org/10.1007/978-3-030-28163-2_3
19. Kornilina M.A., Yakobovsky M.V. Modeling the evolution of complex nonlinear systems on multiprocessor computing complexes. Journal of Physical Chemistry. 1995;69(8):1545–1548. (In Russ.).
20. Dorodnitsyn L.V., Kornilina M.A., Chetverushkin B.N., Yakobovsky M.V. Modeling gas flows in the presence of chemically active components. Journal of Physical Chemistry. 1997;71(12):2275–2281. (In Russ.).
21. Samarsky A.A. Theory of Difference Schemes. Moscow: Nauka; 1989. 616 p.
22. Yakobovsky M.V. Processing grid data on distributed computing systems. Issues of atomic science and technology. Series Mathematical modeling of physical processes. 2004;2:40–53. (In Russ.).
23. Yakobovsky M.V. Incremental graph decomposition algorithm. Bulletin of Lobachevsky University of Nizhny Novgorod. Series Mathematical Modeling and Optimal Control. 2005;1(28):243–250. (In Russ.).
24. Fiedler M. A property of eigenvectors of nonnegative symmetric matrices and its application to graph theory. Czechoslovak Mathematical Journal. 1975;25(100):619–633. URL: http://www3.math.tu-berlin.de/Vorlesungen/SS14/MatricesGraphsPDEs/paper_for_students/CZMathJ-25-1975-Fiedler%20copy.pdf (accessed: 27.02.2024).
25. Hendrickson B., Leland R. A Multilevel Algorithm for Partitioning Graphs. Supercomputing ‘95 Proceedings. San Diego, CA, 1995. URL: http://www.leonidzhukov.net/hse/2016/networks/papers/MultilevelAlgorithmPartitioningGraphs.pdf (accessed: 27.02.2024).
26. Karypis G. Family of Graph and Hypergraph Partitioning Software URL: http://glaros.dtc.umn.edu/gkhome/views/metis/ (accessed: 27.02.2024).
27. Pothen A., Simon H.D., and Kang-Pu P.L. Partitioning Sparse Matrices with Eigenvectors of Graphs. Report RNR-89-009, July 1989. URL: http://snap.stanford.edu/class/cs224w-readings/Pothen89Partition.pdf (accessed: 27.02.2024).
28. Sukhinov A.A. Mathematical Modeling of Impurity Transport Processes in Liquids and Porous Media. Candidate of Physical and Mathematical Sciences dissertation. Moscow, 2009. 24 p. (In Russ.).
29. Sukhinov A.A. Reconstruction of an Ecological Disaster in the Azov Sea Based on Mathematical Models. Mathematical Modeling. 2008;20(6):15–22. (In Russ.).
30. Kornilina M.A., Yakobovsky M.V. Overhead Costs Assessment for Calculations on Locally Refined Grids. Preprints of the M.V. Keldysh Institute of Applied Mathematics. 2022;102:36 p. (In Russ.). https://doi.org/10.20948/prepr-2022-102
31. Grigoriev S.K., Zakharov D.A., Kornilina M.A., Yakobovsky M.V. Dynamic Load Balancing Using Adaptive Locally Refined Grids. Mathematical Modeling. 2023;35(12):69–88. (In Russ.). https://doi.org/10.20948/mm-2023-12-05
Review
For citations:
Yakobovskiy M.V., Kornilina M.A. Development of Supercomputer Technologies at the Institute of Mathematical Modelling and Keldysh Institute of Applied Mathematics of Russian Academy of Sciences. Computational Mathematics and Information Technologies. 2024;8(1):12-28. https://doi.org/10.23947/2587-8999-2024-8-1-12-28
Views:
380
Email this article 