CRPC-TR91186

Title:    Hierarchical Tree-Structures as Adaptive Meshes

Authors:  David J. Edelsohn
	  Northeast Parallel Architectures Center
	  Syracuse University
	  Syracuse, NY 13244

Date:     November 1991


Of the authors listed above, please indicate which are:

  Minority authors:

  Female authors:

  Student authors: David J. Edelsohn


Keywords (list up to 8):

  Adaptive Hierarchical Tree Structures, Adaptive Mesh Refinement, Particle
  (N-body) Simulations, Multipole Methods, Multiscale (Multilevel /
  Multigrid) Methods, Local Corrections, Operator Locality

Abstract:

  New adaptive mesh refinement algorithms provide an opportunity to utilize
  the same hierarchical tree-structures developed for multipole-based
  particle simulations in grid-based simulations of both continuum and
  particle problems.  Representing both a multipole method simulation and an
  adaptive mesh simulation with this same structure provides a natural
  formalism with which to unite these two classes of solvers.  This paper
  discusses how both methods exploit the same basic principle of locality
  evident in many systems, such as those governed by Poisson's Equation, and
  introduces issues and potential problems to be addressed in an
  implementation.  The locality of the systems and the resulting algorithms
  provide important benefits for implementations on massively parallel
  computers.


Publication History:

  Submitted to:

  Published in:

	Northeast Parallel Architectures Center Technical Report SCCS-193.

	International Journal of Modern Physics C, 4(5), p.909, 1993.

	Parallel Computing Works! (Geoffrey Fox, Roy Williams, and Paul
	Messina, editors), Morgan Kaufmann, 1993.