SCYON Abstract

Received on November 6 2006

Performance Analysis of Direct N-Body Algorithms on Special-Purpose Supercomputers

AuthorsStefan Harfst (1), Alessia Gualandris (2), David Merritt (1), Rainer Spurzem (3), Simon Portegies Zwart (2), and Peter Berczik (3,4)
Affiliation
(1) Department of Physics and Astronomy, Rochester Institute of Technology, Rochester, NY 14623
(2) Astronomical Institute Anton Pannekoek and Section Computational Science, University of Amsterdam, The Netherlands
(3) Astronomisches Rechen-Institut, Zentrum für Astronomie, Universität Heidelberg, Heidelberg, Germany
(4) Main Astronomical Observatory, National Academy of Science, Kiev, Ukraine, 03680
Submitted toNew Astronomy
Contactharfst@astro.rit.edu
URLhttp://www.grapecluster.rit.edu
Links

Abstract

Direct-summation N-body algorithms compute the gravitational interaction between stars in an exact way and have a computational complexity of the order of N2. Performance can be greatly enhanced via the use of special-purpose accelerator boards like the GRAPE-6A. However the memory of the GRAPE boards is limited. Here, we present a performance analysis of direct N-body codes on two parallel supercomputers that incorporate special-purpose boards, allowing as many as four million particles to be integrated. Both computers employ high-speed, Infiniband interconnects to minimize communication overhead, which can otherwise become significant due to the small number of "active" particles at each time step. We find that the computation time scales well with processor number; for 2x106 particles, efficiencies greater than 60% and speeds in excess of ~3 TFlops are reached.