Skip to main content
Journal cover image

RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions

Publication ,  Journal Article
Zhang, B; Huang, J; Pitsianis, NP; Sun, X
Published in: Communications in Computational Physics
August 1, 2016

We present RECFMM, a program representation and implementation of a recursive scheme for parallelizing the adaptive fast multipole method (FMM) on shared-memory computers. It achieves remarkable high performance while maintaining mathematical clarity and flexibility. The parallelization scheme signifies the recursion feature that is intrinsic to the FMM but was not well exploited. The program modules of RECFMM constitute a map between numerical computation components and advanced architecture mechanisms. The mathematical structure is preserved and exploited, not obscured nor compromised, by parallel rendition of the recursion scheme. Modern software system - CILK in particular, which provides graph-theoretic optimal scheduling in adaptation to the dynamics in parallel execution - is employed. RECFMM supports multiple algorithm variants that mark the major advances with low-frequency interaction kernels, and includes the asymmetrical version where the source particle ensemble is not necessarily the same as the target particle ensemble. We demonstrate parallel performance with Coulomb and screened Coulomb interactions.

Duke Scholars

Published In

Communications in Computational Physics

DOI

EISSN

1991-7120

ISSN

1815-2406

Publication Date

August 1, 2016

Volume

20

Issue

2

Start / End Page

534 / 550

Related Subject Headings

  • Applied Mathematics
  • 4601 Applied computing
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Zhang, B., Huang, J., Pitsianis, N. P., & Sun, X. (2016). RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions. Communications in Computational Physics, 20(2), 534–550. https://doi.org/10.4208/cicp.230216.140416sw
Zhang, B., J. Huang, N. P. Pitsianis, and X. Sun. “RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions.” Communications in Computational Physics 20, no. 2 (August 1, 2016): 534–50. https://doi.org/10.4208/cicp.230216.140416sw.
Zhang B, Huang J, Pitsianis NP, Sun X. RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions. Communications in Computational Physics. 2016 Aug 1;20(2):534–50.
Zhang, B., et al. “RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions.” Communications in Computational Physics, vol. 20, no. 2, Aug. 2016, pp. 534–50. Scopus, doi:10.4208/cicp.230216.140416sw.
Zhang B, Huang J, Pitsianis NP, Sun X. RECFMM: Recursive Parallelization of the Adaptive Fast Multipole Method for Coulomb and Screened Coulomb Interactions. Communications in Computational Physics. 2016 Aug 1;20(2):534–550.
Journal cover image

Published In

Communications in Computational Physics

DOI

EISSN

1991-7120

ISSN

1815-2406

Publication Date

August 1, 2016

Volume

20

Issue

2

Start / End Page

534 / 550

Related Subject Headings

  • Applied Mathematics
  • 4601 Applied computing