DefinePK

DefinePK hosts the largest index of Pakistani journals, research articles, news headlines, and videos. It also offers chapter-level book search.

Parallelizability of tasks using ANSYS HPC on Supercomputer to solve large tasks of Civil Engineering - A case study


Article Information

Title: Parallelizability of tasks using ANSYS HPC on Supercomputer to solve large tasks of Civil Engineering - A case study

Authors: Zdenka Neuwirthova, Radim Cajka

Journal: ARPN Journal of Engineering and Applied Sciences

HEC Recognition History
Category From To
Y 2023-07-01 2024-09-30
Y 2022-07-01 2023-06-30
Y 2021-07-01 2022-06-30
X 2020-07-01 2021-06-30

Publisher: Khyber Medical College, Peshawar

Country: Pakistan

Year: 2020

Volume: 15

Issue: 5

Language: English

Categories

Abstract

Supercomputers are widely used across variety of industries. Civil Engineering methods of numerical modelling are conservative and new technologies are used only occasionally. As numerical models become more complex with more detailed material models and a more extensive construction model, the limits of standard workstations are often reached. When a large task with many unknowns and nonlinearities needs to be solved, the demands on hardware requirements are high. In such cases supercomputers are convenient because they allow to solve larger tasks then a standard workstation Low computational time is another advantage of a supercomputer. Finally, the supercomputer doesn't need a huge investment from the user, because only computing time is rented. Appropriate resource allocation is essential to maximize the computational potential of a supercomputer. The task in this article is aimed on a soil-foundation interaction. The Ansys HPC commercial software was used for the calculations on the Anselm cluster in the National Supercomputing Center IT4 Innovations in VSB-Technical University of Ostrava. The optimal method of resources allocation was observed and discussed with regard to task size and a calculation time. The article consists of two examples. First example was computed on 16 nodes with the 16 Ansys HPC licenses with various methods of recourses allocation. The second example evaluated using all available HPC licenses in the National Supercomputing Center IT4Innovations. The maximal possible task size was evaluated in all used methods.


Paper summary is not available for this article yet.

Loading PDF...

Loading Statistics...