A simple spectral algorithm for solving large-scale Poisson equation in 2D

X. Thibert-Plante, D. A. Yuen, A. P. Vincent

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

We show that it is possible with easy-to-program algorithms to reach spatial resolutions of the order of 108 grid points for computing the electric potential on 2D periodic lattices, such as the Si(111)7×7 surface. We have used a spectral Fourier technique and parallelized FFTs with OPEN_MP on SGI machines. This method can be easily extended to 3D.

Original languageEnglish (US)
Pages (from-to)89-97
Number of pages9
JournalComputer Physics Communications
Volume154
Issue number2
DOIs
StatePublished - Aug 1 2003

Bibliographical note

Funding Information:
We thank Paul Woodward for access to the Power Wall at LCSE, Renata Wentzcovitch for encouraging remarks, Shuxia, Zhang for technical assistance at MSI, Don Truhlar for helpful comments and Michel Beland from RQCHP for advice on OPEN_MP. Simulations have been made at the Minnesota Supercomputing Institute at Minneapolis and at CERCA at Montreal. This research has been financed by CRSNG Canada (Alain Vincent) and NSF and DOE (David Yuen). Xavier Thibert-Plante is a fellow of CRSNG Canada.

Keywords

  • Crystal lattice
  • FFT-parallelization
  • OPEN_MP
  • Poisson equation

Fingerprint

Dive into the research topics of 'A simple spectral algorithm for solving large-scale Poisson equation in 2D'. Together they form a unique fingerprint.

Cite this