On the use of a heterogeneous MIMD-SIMD platform to simulate the dynamics of globular clusters with a central massive object

Abstract The dynamics of a large stellar (globular) cluster containing N =128,000 stars has been simulated using a direct summation ( O ( N 2 )) method and a heterogeneous platform. Preliminary simulations have been carried out on model systems with and without the presence, in their center of mass, of a black hole whose mass has been varied from 0.02 to 0.1 times the total mass of the cluster. These simulations followed the evolution of the globular cluster in order to describe its dynamics over an interval of time sufficiently large with respect to the internal crossing time. Computations have demonstrated that the platform heterogeneity, allowing a very efficient use of the computational resources, can be considered a key feature for sustaining large computational loads. Our results show that the massive object in the center of the cluster alters the surrounding star distribution very quickly; the following evolution is much slower, as it occurs via two-body collisional relaxation.

[1]  Vittorio Rosato,et al.  Heterogeneity as key feature of high performance computing: the PQE1 prototype , 2000, Proceedings 9th Heterogeneous Computing Workshop (HCW 2000) (Cat. No.PR00556).

[2]  R. Spurzem.  Direct N-body simulations , 1999, astro-ph/9906154.

[3]  Makoto Taiji,et al.  Scientific simulations with special purpose computers - the GRAPE systems , 1998 .

[4]  Toshikazu Ebisuzaki,et al.  GRAPE-4: A Massively Parallel Special-Purpose Computer for Collisional N-Body Simulations , 1997 .

[5]  Vittorio Rosato,et al.  Search of molecular ground state via genetic algorithm: Implementation on a hybrid SIMD-MIMD platform , 1997 .

[6]  Thomas Lippert,et al.  HYPER-SYSTOLIC PROCESSING ON APE100/QUADRICS: n2-LOOP COMPUTATIONS , 1995, hep-lat/9512020.

[7]  Henk W. J. Blöte,et al.  Cluster algorithm special purpose computer for the three-dimensional Ising model , 1995 .

[8]  I. Arsenin.  Architectural choices for the Columbia 0.8 Teraflops machine , 1994, hep-lat/9412093.

[9]  Raffaele Tripiccione,et al.  A HARDWARE IMPLEMENTATION OF THE APE100 ARCHITECTURE , 1993 .

[10]  Raffaele Tripiccione,et al.  THE SOFTWARE OF THE APE100 PROCESSOR , 1993 .

[11]  Nicolai Petkov,et al.  Systolic Parallel Processing , 1992 .

[12]  Wang,et al.  Nonuniversal critical dynamics in Monte Carlo simulations. , 1987, Physical review letters.

[13]  Piet Hut,et al.  A hierarchical O(N log N) force-calculation algorithm , 1986, Nature.

[14]  S. Shapiro,et al.  The distribution of stars around a massive black hole , 1976, Nature.

[15]  L. Verlet.  Computer "Experiments" on Classical Fluids. I. Thermodynamical Properties of Lennard-Jones Molecules , 1967 .

[16]  H. Plummer.  On the Problem of Distribution in Globular Star Clusters: (Plate 8.) , 1911 .

[17]  Klaus Schilling,et al.  Hyper-Systolic Processing on the Quadrics: Improving Inter-Processor Communication by Simulated Annealing , 1997 .

[18]  R W Hockney,et al.  Computer Simulation Using Particles , 1966 .