Acceleration of integration in parametric integral equations system using CUDA. (May 2015)

• Record Type: Journal Article
• Authors: Kużelewski, Andrzej; Zieniuk, Eugeniusz; Kapturczak, Marta
• Is Part Of: Computers & structures. Volume 152(2015)
• Page Start: 113
• Page End: 124
• Publication Date: 2015-05
• Subjects: PIES -- GPU -- CUDA -- Multithreaded computing -- Boundary value problems; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2015.02.019
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: A parallel approach of calculating integrals in GPU-accelerated PIES is proposed. Processing power of GPU is used to accelerate calculating integrals. Computing time of GPU-accelerated PIES was examined. The accuracy of the solutions obtained using GPU-accelerated PIES was examined. The effectiveness of GPU-accelerated PIES was proved. Abstract: Methods of solving boundary value problems are expected to achieve high accuracy results in shortest possible time of calculations. In the previous papers the authors solved boundary value problems with high accuracy using parametric integral equations system. However, time of calculations was unsatisfactory. The most time consuming part of the method is calculation of integrals. In this paper the authors present approach to accelerate numerical integration in PIES by nVidia CUDA. The speed of integration increased up to 250 times whereas high accuracy of solutions was maintained. Examples included in this paper concern solving three-dimensional elasticity problems.