arXiv ScienceSearch

arXiv · 0709.4552

Distributed N-body Simulation on the Grid Using Dedicated Hardware

Abstract

We present performance measurements of direct gravitational N -body simulation on the grid, with and without specialized (GRAPE-6) hardware. Our inter-continental 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. With this 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. Based on these timings we construct and calibrate a model to predict the performance of our simulation on any grid infrastructure with or without GRAPE. We apply this model to predict the simulation performance on the Netherlands DAS-3 wide area computer. Equipping the DAS-3 with GRAPE-6Af hardware would achieve break-even between calculation and communication at a few million particles, resulting in a compute time of just over ten hours for 1 N -body time unit. Key words: high-performance computing, grid, N-body simulation, performance modelling

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Derek Groen, Simon Portegies Zwart, Steve McMillan, Jun Makino. 2007-11-05. Distributed N-body Simulation on the Grid Using Dedicated Hardware. https://doi.org/10.1016/j.newast.2007.11.004

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

On binary pulsars and the force of gravity

The energy-momentum budget of the astrophysical systems can be studied by the exact local conservation equation derived by Landau and Lifshitz. We show that a similar equation is valid for the Einstein-Cartan gravity. We reanalyze a binary pulsar system using the Landau-Lifshitz conservation equation and show that the orbital period change rate can be completely understood as a curvature backreaction process. Taking into account the detailed theoretical and observational research of relativistic binary pulsar systems, especially the system of Hulse and Taylor, we conclude that general relativity and astrophysical observations rule out the existence of gravitational radiation. We comment upon the LIGO GW events and their alternative explanation, as well as the recent pulsar timing arrays data.

astro-ph

Oscillation frequencies and mode lifetimes in alpha Centauri A

We analyse our recently-published velocity measurements of alpha Cen A (Butler et al. 2004). After adjusting the weights on a night-by-night basis in order to optimize the window function to minimize sidelobes, we extract 42 oscillation frequencies with l=0 to 3 and measure the large and small frequency separations. We give fitted relations to these frequencies that can be compared with theoretical models and conclude that the observed scatter about these fits is due to the finite lifetimes of the oscillation modes. We estimate the mode lifetimes to be 1-2 d, substantially shorter than in the Sun.

astro-ph

Hipparcos period-luminosity relations for Miras and semiregular variables

We present period-luminosity diagrams for nearby Miras and semiregulars, selecting stars with parallaxes better than 20 per cent and well-determined periods. Using K-band magnitudes, we find two well-defined P-L sequences, one corresponding to the standard Mira P-L relation and the second shifted to shorter periods by a factor of about 1.9. The second sequence only contains semiregular variables, while the Mira sequence contains both Miras and semiregulars. Several semiregular stars show double periods in agreement with both relations. The Whitelock evolutionary track is shown to fit the data, indicating that the semiregulars are Mira progenitors. The transition between the two sequences may correspond to a change in pulsation mode or to a change in the stellar structure. Large amplitude pulsations leading to classical Mira classification occur mainly near the tip of the local AGB luminosity function.

astro-ph