A.A. Petrov1, A.A. Ermakov2, I.V. Maslov3

1–3 Bunin Yelets State University (Yelets, Russia)
1xeal91@yandex.ru, 2artem_ermakov_00@list.ru, 3maslov0906@gmail.com

The paper is devoted to a comparative analysis of the possibilities of application and evaluation of performance of parallel computing technologies on local equipment for solving tasks in artificial intelligence and nonlinear dynamics. The relevance of the research is associated with the active growth of data volumes used for constructing information models in scientific research, which makes the problem of optimal choice of software-hardware tools for parallel programming particularly important. A comparative analysis of popular parallel computing libraries was conducted from the perspective of their performance, ease of use, and scalability across different platforms. Computational experiments were carried out aimed at evaluating the performance of various software-hardware solutions in tasks related to artificial intelligence and nonlinear dynamics, taking into account matrix operations usage. The results of this study can be applied when searching for optimal parameters of software-hardware complexes for mathematical modeling of nonlinear systems.

Petrov A.A., Ermakov A.A., Maslov I.V. Application of parallel computing technologies on local hardware for solving artificial intelligence and nonlinear dynamics problems. Nonlinear World. 2025. V. 23. № 4. P. 30–35. DOI: https:// doi.org/10. 18127/j20700970-202504-03 (In Russian)

1. Al-Shafei A., Zareipour, H., Cao Y. High-Performance and Parallel Computing Techniques Review: Applications, Challenges and Potentials to Support Net-Zero Transition of Future Grids. Energies. 2022. V. 15. № 22. 8668. https://doi.org/10.3390/ en15228668.
2. Markus S., Martin M., Eddelbuettel D., Hao Y., Tierney L., Mansmann U. State of the Art in Parallel Computing with R. Journal of Statistical Software. 2009. V. 31. P. 1–27. https://doi.org/10.18637/jss.v031.i01.
3. Shukur H., Zeebaree S., Ahmed A., Zebari R., Ahmed O., Shams B., Tahir A., M.Sadeeq M. A State of Art Survey for Concurrent Computation and Clustering of Parallel Computing for Distributed Systems. Journal of Applied Science and Technology Trends. 2020. V. 01. P. 148–154. https://doi.org/10.38094/jastt1466.
4. Anirad K., Gandzhu S., Kazam M. Iskusstvennyj intellekt i komp'yuternoe zrenie: Real'nye proekty na Python, Keras i TensorFlow. M.: Naukovaya biblioteka. 2023 (In Russian).
5. Kabachnik D. Iskusstvennyj intellekt v promyshlennyh granichnyh vychisleniyah. STA. 2022. № 2. S. 36–37.
6. Petrov A.A., Chernomordov S.V., Kabanov M.A. Analiz instrumental'nogo obespecheniya geterogennyh vychislenij s primeneniem nejrosetevogo modelirovaniya. Nejrokomp'yutery: razrabotka, primenenie. 2023. T. 25. № 2. S. 21–29. DOI: 10.18127/j19998554-202202-02. EDN ZMTDDX (In Russian).
7. Chernomordov S.V., Druzhinina O.V., Masina O.N., Petrov A.A. Primenenie metodov mashinnogo obucheniya v zadachah nejrosetevogo modelirovaniya upravlyaemyh tekhnicheskih system. Nejrokomp'yutery: razrabotka, primenenie. 2022. T. 24. № 1. S. 25–35 (In Russian).
8. Druzhinina O.V., Korepanov E.R., Belousov V.V., Masina O.N., Petrov A.A. Opyt razrabotki metodov i sredstv nejrosetevogo modelirovaniya nelinejnyh sistem na baze otechestvennoj vychislitel'noj platformy «El'brus 801-PC». Nelinejnyj mir. 2020. T. 18. № 2. S. 5–17 (In Russian).
9. Strongin R.G., Gergel' V.P., Grishagin V.A., Barkalov K.A. Parallel'nye vychisleniya v zadachah global'noj optimizacii. M.: Izd-vo Mosk. gos. un-ta im. M.V. Lomonosova. 2013 (In Russian).
10. Boreskov A.V., Harlamov A.A., Markovskij N.D. Parallel'nye vychisleniya na GPU. Arhitektura i programmnaya model' CUDA. M.: MGU imeni M.V. Lomonosova. 2015 (In Russian).
11. Ciccozzi F. et al. A comprehensive exploration of languages for parallel computing. ACM Computing Surveys (CSUR). 2022. V. 55. № 2. P. 1–39.