Are you ready to FLY in the universe? A multi-platform [formula omitted]-body tree code for parallel supercomputers

Abstract

In the last few years, cosmological simulations of structures and galaxies formations have assumed a fundamental role in the study of the origin, formation and evolution of the universe. These studies improved enormously with the use of supercomputers and parallel systems, allowing more accurate simulations, in comparison with traditional serial systems. The code we describe, called FLY, is a newly written code (using the tree N-body method), for three-dimensional self-gravitating collisionless systems evolution. FLY is a fully parallel code based on the tree Barnes–Hut algorithm and periodical boundary conditions are implemented by means of the Ewald summation technique. We use FLY to run simulations of the large scale structure of the universe and of cluster of galaxies, but it could be usefully adopted to run evolutions of systems based on a tree N-body algorithm. FLY is based on the one-side communication paradigm to share data among the processors, that access to remote private data avoiding any kind of synchronism. The code was originally developed on CRAY T3E system using the logically SHared MEMory access routines (SHMEM) but it runs also on SGI ORIGIN systems and on IBM SP by using the Low-Level Application Programming Interface routines (LAPI). This new code is the evolution of preliminary codes (WDSH-PT and WD99) for cosmological simulations we implemented in the last years, and it reaches very high performance in all systems where it has been well-tested. This performance allows us today to consider the code FLY among the most powerful parallel codes for tree N-body simulations. The performance that FLY reaches is discussed and reported, and a comparison with other similar codes is preliminary considered. The FLY version 1.1 is freely available on http://www.ct.astro.it/fly/ and it will be maintained and upgraded with new releases.