A Fast Parallel Treecode with GRAPE
- 25 June 2004
- journal article
- Published by Oxford University Press (OUP) in Publications of the Astronomical Society of Japan
- Vol. 56 (3) , 521-531
Abstract
In this paper, we describe an implementation of the Barnes–Hut treecode on a cluster of PCs (Beowulf-class machines) equipped with GRAPE hardware. Our implementation is designed to achieve good performance on clusters of GRAPE with a relatively slow network, such as the standard Gigabit Ethernet. With 12 processors, the speedup over single-processor code reaches around 10. Our code is available on the web.Keywords
This publication has 17 references indexed in Scilit:
- GOTPM: a parallel hybrid particle-mesh treecodeNew Astronomy, 2004
- Pseudoparticle Multipole Method: A Simple Method to Implement a High-Accuracy Tree CodeThe Astrophysical Journal, 2001
- GRAPE-5: A Special-Purpose Computer for N-Body SimulationsPublications of the Astronomical Society of Japan, 2000
- Performance and accuracy of a GRAPE-3 system for collisionless N-body simulationsMonthly Notices of the Royal Astronomical Society, 1998
- A parallel tree codeNew Astronomy, 1996
- Discreteness Noise versus Force Errors in N-Body SimulationsThe Astrophysical Journal, 1993
- A modified tree code: Don't laugh; It runsJournal of Computational Physics, 1990
- Are some N-body algorithms intrinsically less collisional than others?The Astrophysical Journal, 1990
- Error analysis of a tree codeThe Astrophysical Journal Supplement Series, 1989
- A hierarchical O(N log N) force-calculation algorithmNature, 1986