Simulating N-Body Systems on the Grid Using Dedicated Hardware

Authors:
Derek Groen;Simon Portegies Zwart;Steve Mcmillan;Jun Makino
Affiliations:
Section Computational Science, University of Amsterdam, Amsterdam, The Netherlands and Astronomical Institute "Anton Pannekoek", University of Amsterdam, Amsterdam, The Netherlands;Section Computational Science, University of Amsterdam, Amsterdam, The Netherlands and Astronomical Institute "Anton Pannekoek", University of Amsterdam, Amsterdam, The Netherlands;Drexel University, Philadelphia, United States;National Astronomical Observatory, Mitaka, Japan
Venue:
ICCS '08 Proceedings of the 8th international conference on Computational Science, Part I
Year:
2008

Citing 5
Cited 0

Performance evaluation and tuning of GRAPE-6 - towards 40 "real" Tflops

Proceedings of the 2003 ACM/IEEE conference on Supercomputing
Grid Environment for Computational Astrophysics Driven by GRAPE-6 with HMCS-G and OmniRPC

IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 4 - Volume 05
The Anatomy of the Grid: Enabling Scalable Virtual Organizations

International Journal of High Performance Computing Applications
Performance analysis of direct N-body algorithms for astrophysical simulations on distributed systems

Parallel Computing
Introducing grid speedup Γ: a scalability metric for parallel applications on the grid

EGC'05 Proceedings of the 2005 European conference on Advances in Grid Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present performance measurements of direct gravitational N-body simulation on the grid, with and without specialized (GRAPE-6) hardware. Our intercontinental virtual organization consists of three sites, one in Tokyo, one in Philadelphia and one in Amsterdam. We run simulations with up to 196608 particles for a variety of topologies. In many cases, high performance simulations over the entire planet are dominated by network bandwidth rather than latency. Using a global grid of GRAPEs our calculation time remains dominated by communication over the entire range of N, which was limited due to the use of three sites. Increasing the number of particles will result in a more efficient execution and for we expect the computation time to overtake the communication time. We compare our results with a performance model and find that our results are in accordance with the predicted values.