A Parallel Tree-SPH code for Galaxy Formation

Abstract

We describe a new implementation of a parallel Tree-SPH code with the aim to simulate Galaxy Formation and Evolution. The code has been parallelized using SHMEM, a Cray proprietary library to handle communications between the 256 processors of the Silicon Graphics T3E massively parallel supercomputer hosted by the Cineca Supercomputing Center (Bologna, Italy). The code combines the Smoothed Particle Hydrodynamics (SPH) method to solve hydro-dynamical equations with the popular Barnes and Hut (1986) tree-code to perform gravity calculation with a NlogN scaling, and it is based on the scalar Tree-SPH code developed by Carraro et al(1998)[MNRAS 297, 1021]. Parallelization is achieved distributing particles along processors according to a work-load criterion. Benchmarks, in terms of load-balance and scalability, of the code are analyzed and critically discussed against the adiabatic collapse of an isothermal gas sphere test using 20,000 particles on 8 processors. The code results balanced at more that 95% level. Increasing the number of processors, the load-balance slightly worsens. The deviation from perfect scalability at increasing number of processors is almost negligible up to 32 processors. Finally we present a simulation of the formation of an X-ray galaxy cluster in a flat cold dark matter cosmology, using 200,000 particles and 32 processors, and compare our results with Evrard (1988) P3M-SPH simulations. Additionaly we have incorporated radiative cooling, star formation, feed-back from SNae of type II and Ia, stellar winds and UV flux from massive stars, and an algorithm to follow the chemical enrichment of the inter-stellar medium. Simulations with some of these ingredients are also presented.