Studying the core-cusp problem in cold dark matter halos using N-body simulations on GPU clusters

Abstract

The discrepancy in the mass-density profile of dark matter halos between simulations and observations, the core-cusp problem, is a long-standing open question in the standard paradigm of cold dark matter cosmology. Here, we study the dynamical response of dark matter halos to oscillations of the galactic potential which are induced by a cycle of gas expansion and contraction in galaxies driven by supernova feedback. We developed a fast tree-code for PC clusters with GPU which displays high performance and high scalability. We perform large scale N-body simulations to follow the dynamical evolution of dark matter halos under the effect of oscillating gravitational potential. Furthermore, we compare the results of simulations with an analytical model of the resonance between particles and density waves to understand the physical mechanism of the cusp-core transition.