Publication detail

Transient heat transfer problem with phase changes solved by means of GPGPU with CUDA and MATLAB

KLIMEŠ, L. ŠTĚTINA, J.

Original Title

Transient heat transfer problem with phase changes solved by means of GPGPU with CUDA and MATLAB

English Title

Transient heat transfer problem with phase changes solved by means of GPGPU with CUDA and MATLAB

Type

conference paper

Language

en

Original Abstract

The paper presents the use of GPGPU via CUDA and MATLAB with GPU NVIDIA Tesla in a computationally demanding transient heat transfer problem with phase changes, which is often the case solved in various engineering applications. The comparison of computing performance between parallel GPU and state-of-the-art CPU numerical models was performed and results indicate that GPGPU is a very efficient approach for solving computational problems and allows for a significant speed-up of computations.

English abstract

The paper presents the use of GPGPU via CUDA and MATLAB with GPU NVIDIA Tesla in a computationally demanding transient heat transfer problem with phase changes, which is often the case solved in various engineering applications. The comparison of computing performance between parallel GPU and state-of-the-art CPU numerical models was performed and results indicate that GPGPU is a very efficient approach for solving computational problems and allows for a significant speed-up of computations.

Keywords

GPGPU, parallel computing, GPU, MATLAB, CUDA, transient heat transfer

RIV year

2013

Released

26.06.2013

Publisher

VUT v Brně

Location

Brno, Česká republika

ISBN

978-80-214-4755-4

Book

Proceedings of 19th International Conference on Soft Computing MENDEL 2013

Pages from

397

Pages to

402

Pages count

6

Documents

BibTex


@inproceedings{BUT100357,
  author="Lubomír {Klimeš} and Josef {Štětina}",
  title="Transient heat transfer problem with phase changes solved by means of GPGPU with CUDA and MATLAB",
  annote="The paper presents the use of GPGPU via CUDA and MATLAB with GPU NVIDIA Tesla in a computationally demanding transient heat transfer problem with phase changes, which is often the case solved in various engineering applications. The comparison of computing performance between parallel GPU and state-of-the-art CPU numerical models was performed and results indicate that GPGPU is a very efficient approach for solving computational problems and allows for a significant speed-up of computations.",
  address="VUT v Brně",
  booktitle="Proceedings of 19th International Conference on Soft Computing MENDEL 2013",
  chapter="100357",
  howpublished="print",
  institution="VUT v Brně",
  year="2013",
  month="june",
  pages="397--402",
  publisher="VUT v Brně",
  type="conference paper"
}