Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A parallel adaptive fast multipole method

Conference ·
OSTI ID:46206
; ; ;  [1]
  1. Stanford Univ., CA (United States). Computer Systems Lab.

The authors present parallel versions of a representative N-body application that uses Greengard and Rokhlin`s adaptive Fast Multipole Method (FMM). While parallel implementations of the uniform FMM are straightforward and have been developed on different architectures, the adaptive version complicates the task of obtaining effective parallel performance owing to the nonuniform and dynamically changing nature of the problem domains to which it is applied. They propose and evaluate two techniques for providing load balancing and data locality, both of which take advantage of key insights into the method and its typical applications. Using the better of these techniques, they demonstrate 45-fold speedups on galactic simulations on a 48-processor Stanford DASH machine, a state-of-the-art shared address space multiprocessor, even for relatively small problems. They also show good speedups on a 2-ring Kendall Square Research KSR-1. Finally, they summarize some key architectural implications of this important computational method.

OSTI ID:
46206
Report Number(s):
CONF-931115--
Country of Publication:
United States
Language:
English

Similar Records

A Massively Parallel Adaptive Fast Multipole Method on Heterogeneous Architectures
Journal Article · Sat Dec 31 23:00:00 EST 2011 · Communications of the ACM · OSTI ID:1039644

Hierarchical Particle Mesh: An FFT-accelerated Fast Multipole Method
Journal Article · Tue Jul 23 00:00:00 EDT 2019 · The Astrophysical Journal. Supplement Series (Online) · OSTI ID:1545079

Hierarchical N-body methods on shared address space multiprocessors
Conference · Thu Nov 30 23:00:00 EST 1995 · OSTI ID:125523