Publication detail

Many Core Acceleration of the Boundary Element Method

MERTA, M. ZAPLETAL, J. JAROŠ, J.

Original Title

Many Core Acceleration of the Boundary Element Method

Type

conference paper

Language

English

Original Abstract

The paper presents the boundary element method accelerated by the Intel Xeon Phi coprocessors. An overview of the boundary element method for the 3D Laplace equation is given followed by the discretization and its parallelization using OpenMP and the offload features of the Xeon Phi coprocessor are discussed. The results of numerical experiments for both single- and double-layer boundary integral operators are presented. In most cases the accelerated code significantly outperforms the original code running solely on Intel Xeon processors.

Keywords

Boundary Element Method, Intel Xeon Phi, MIC, Many Integrated Cores, Acceration

Authors

MERTA, M.; ZAPLETAL, J.; JAROŠ, J.

Released

3. 6. 2016

Publisher

Springer International Publishing

Location

Basel

ISBN

978-3-319-40360-1

Book

Proceedings of High Performance Computing in Science and Engineering

Edition

Lecture Notes in Computer Science

Pages from

116

Pages to

125

Pages count

10

URL

BibTex

@inproceedings{BUT130911,
  author="Michal {Merta} and Jan {Zapletal} and Jiří {Jaroš}",
  title="Many Core Acceleration of the Boundary Element Method",
  booktitle="Proceedings of High Performance Computing in Science and Engineering",
  year="2016",
  series="Lecture Notes in Computer Science",
  volume="9611",
  pages="116--125",
  publisher="Springer International Publishing",
  address="Basel",
  doi="10.1007/978-3-319-40361-8\{_}8",
  isbn="978-3-319-40360-1",
  url="http://link.springer.com/chapter/10.1007%2F978-3-319-40361-8_8"
}