GPU software and architecture comparison for numerical simulation of partial deferential equations

• Autorzy: May Jon B. , Pera Donato

Identyfikatory

DOI

10.17466/tq2018/22.1/c

• Języki publikacji: EN

Abstrakty

EN

This paper will show a comparison between the Kepler, Maxwell and Pascal GPU architectures using CUDA-Fortran, with and without dynamic calls, to efficiently solve partial differential equations. The target is to show the possibility of using affordable hardware, such astheGTX670,GTX970 andGTX1080 NVIDIA GPUs, which are commonly found in personal and portable computers, for scientific applications. For simplicity we consider a standard wave equation where we use a second order finite difference method for the spatial and time discretizations to obtain the numerical solution. We found that, as we increase the spatialre solution of the domain we also increase the performance difference between the GPU and the Central Processing Unit (CPU).

Słowa kluczowe

EN

GPGPU; PDE

• Wydawca: Politechnika Gdańska
• Czasopismo: TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
• Rocznik: 2018
• Tom: Vol. 22. No 1
• Strony: 85--100
• Opis fizyczny: Bibliogr. 14 poz., rys., tab.

Twórcy

autor

May Jon B.

• Dipartimento di Ingegneria e Scienze dell’Informazione e Matematica, Università degli Studi dell’Aquila, via Vetoio (snc), Località Coppito, L’Aquila 67010, Italy

autor

Pera Donato

• Dipartimento di Ingegneria e Scienze dell’Informazione e Matematica, Università degli Studi dell’Aquila, via Vetoio (snc), Località Coppito, L’Aquila 67010, Italy

Bibliografia

• [1] Pera D 2013 Parallel numerical simulations of anisotropic and heterogeneous diffusionequations with GPGPU, PhD Thesis
• [2] Roniotis A, Marias K, Sakkalis V, Tsibidis G D and Zervakis M 2009 31st AnnualInternational Conference of the IEEE EMBS
• [3] Rubio F, Hanzich M, Farrès A, Puente de la J and Cela J M 2014 Computers andGeosciences70181
• [4] Weickert J 1998 Anisotropic Diffusion in Image Processing,ECMISeries, Teubner-Verlag
• [5] https://www.pgroup.com/resources/cudafortran.html
• [6] CUDA Programming Manual NVIDIA 2010
• [7] Kirk D and Hwu W-M 2010 Programming Massively Parallel Processors: A Hands-onApproachNVIDIA
• [8] Sanders J and Kandrot E 2010 CUDAby example An Introduction to General Purpose GPU Programming, Addison-Wesley
• [9] Kupferschmid M 2010 Classical Fortran: Programming for Engineering and Scientific Applications, Second Edition,CRC Press
• [10] White Paper NVIDIA Kepler GK110, http://www.nvidia.com
• [11] White Paper NVIDIA Maxwell GTX980, http://www.nvidia.com
• [12] White Paper NVIDIA Tesla P100, http://www.nvidia.com
• [13] https://en.wikipedia.org/wiki/Kepler
• [14] Eaton J W, Bateman D, Hauberg S and Wehbring R 2015 GNUOctave version 4.0.0manual: a high-level interactive language for numerical computations, Create Space Independent Publishing Platform

Uwagi

PL

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).