Computational Technology Resources - CCC

F. Hajdu¹, C. Hajdu² and L. Környei³

¹Department of Machine Design, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary
²Department of Informatics, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary
³Department of Mathematics and Computational Sciences, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary

F. Hajdu, C. Hajdu, L. Környei, "Parallelization of Global Sensitivity Study of Nonlinear Systems Using Komondor HPC", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Eighth International Conference on Parallel, Distributed, GPU and Cloud Computing for Engineering", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 12, Paper 2.5, 2025,

Keywords: parallelization, sensitivity study, nonlinear system, numerical simulation, HPC, Sobol’s method.

Abstract

One way to develop a new nonlinear system model is to use its sensitivity study. The global sensitivity study of nonlinear systems requires a lot of parameter combinations and, therefore a lot of calculations. As the differential equations are independent of each parameter combination the process can be accelerated with parallelization. The aim of the research was to parallelize Sobol’s sensitivity study of nonlinear systems with a Duffing-type vibration system as an example. Using Komondor HPC different configurations were tested from which the OpenMP with 128 threads gave the best results. With the OpenMP configuration a strong and weak scalability study was performed, and it was shown that the problem is well scalable. Using job arrays 16 computers were utilized and a 1397.95-fold speedup could be achieved. With the results it will be possible to carry out the sensitivity analysis of difficult nonlinear systems effectively.

download the full-text of this paper (PDF, 13 pages, 527 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 12 PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, GPU AND CLOUD COMPUTING FOR ENGINEERING Edited by: P. Iványi, J. Kruis and B.H.V. Topping Paper 2.5 Parallelization of Global Sensitivity Study of Nonlinear Systems Using Komondor HPC F. Hajdu¹, C. Hajdu² and L. Környei³ ¹Department of Machine Design, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary ²Department of Informatics, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary ³Department of Mathematics and Computational Sciences, Faculty of Mechanical Engineering, Informatics and Electrical Engineering, Széchenyi István University, Győr, Hungary Full Bibliographic Reference for this paper F. Hajdu, C. Hajdu, L. Környei, "Parallelization of Global Sensitivity Study of Nonlinear Systems Using Komondor HPC", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Eighth International Conference on Parallel, Distributed, GPU and Cloud Computing for Engineering", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 12, Paper 2.5, 2025, Keywords: parallelization, sensitivity study, nonlinear system, numerical simulation, HPC, Sobol’s method. Abstract One way to develop a new nonlinear system model is to use its sensitivity study. The global sensitivity study of nonlinear systems requires a lot of parameter combinations and, therefore a lot of calculations. As the differential equations are independent of each parameter combination the process can be accelerated with parallelization. The aim of the research was to parallelize Sobol’s sensitivity study of nonlinear systems with a Duffing-type vibration system as an example. Using Komondor HPC different configurations were tested from which the OpenMP with 128 threads gave the best results. With the OpenMP configuration a strong and weak scalability study was performed, and it was shown that the problem is well scalable. Using job arrays 16 computers were utilized and a 1397.95-fold speedup could be achieved. With the results it will be possible to carry out the sensitivity analysis of difficult nonlinear systems effectively. download the full-text of this paper (PDF, 13 pages, 527 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions