Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Efficient implementation of the fast multipole method.

Elias Rudberg1, Paweł Sałek

  • 1Department of Chemistry, University of Warwick, Coventry CV4 7AL, United Kingdom.

The Journal of Chemical Physics
|September 13, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

On-the-Fly Computation of Frontal Orbitals in Density Matrix Expansions.

Journal of chemical theory and computation·2017
Same author

Parameterless Stopping Criteria for Recursive Density Matrix Expansions.

Journal of chemical theory and computation·2016
Same author

Canonical density matrix perturbation theory.

Physical review. E, Statistical, nonlinear, and soft matter physics·2016
Same author

Heisenberg Exchange in Dinuclear Manganese Complexes:  A Density Functional Theory Study.

Journal of chemical theory and computation·2015
Same author

Gauge-Origin Independent Formulation and Implementation of Magneto-Optical Activity within Atomic-Orbital-Density Based Hartree-Fock and Kohn-Sham Response Theories.

Journal of chemical theory and computation·2015
Same author

Automatic Selection of Integral Thresholds by Extrapolation in Coulomb and Exchange Matrix Constructions.

Journal of chemical theory and computation·2015

This study introduces computational techniques to speed up the fast multipole method (FMM) for quantum chemistry calculations. Dynamically truncated expansions offer a tenfold increase in computational efficiency for multipole interactions.

Area of Science:

  • Computational Chemistry
  • Quantum Mechanics
  • Scientific Computing

Background:

  • The fast multipole method (FMM) is crucial for evaluating Coulomb matrix elements in Hartree-Fock and density functional theories.
  • Computational efficiency is a significant bottleneck in large-scale quantum mechanical calculations.
  • Existing FMM implementations require substantial computational resources.

Purpose of the Study:

  • To develop novel computational techniques for reducing the computational effort of the continuous fast multipole method.
  • To improve the efficiency of calculating Coulomb matrix elements in quantum chemistry.
  • To present an optimized FMM implementation for large-scale electronic structure calculations.

Main Methods:

  • Proposed a new definition for the extent of Gaussian charge distributions.

Related Experiment Videos

  • Introduced a novel method for partitioning charge distributions into branches.
  • Developed a new approach for estimating errors from multipole expansion truncation.
  • Implemented dynamically truncated multipole expansions.

    • Main Results:

    • Achieved a tenfold speedup in multipole interaction computations using dynamically truncated expansions compared to fixed orders.
    • Demonstrated the effectiveness of the proposed techniques through benchmark calculations on 3D systems.
    • Validated the usefulness of the implemented fast multipole method.

    Conclusions:

    • The developed computational techniques significantly enhance the efficiency of the fast multipole method.
    • Dynamically truncated multipole expansions are a key factor in achieving substantial speedups.
    • The optimized FMM implementation is a valuable tool for advanced quantum chemistry and materials science research.