MilkyWay@home is a distributed supercomputer that uses more than 32,000 volunteered computers (over 1.7 PetaFLOPS) to optimize parameter fits to data. This platform has been used to find and fit tidal streams in the Milky Way halo and to reconstruct the progenitor of the Orphan Stream. The code development for this system includes an N-body integrator that uses a Barnes-Hut tree algorithm and can be used outside of the distributed computing environment on one or multiple CPUs. This integrator allows the user to integrate a set of bodies through a variety of multi-component Milky Way potentials, or no external potential. We are developing capabilities to include a bar potential and the gravity of the Large Magellanic Cloud (LMC). The application will create the initial configuration of bodies for multi-component (spherically symmetric) dwarf galaxies, including combinations of both NFW and Plummer sphere density profiles. Alternatively, the initial configuration of bodies can be imported by hand. We include options for capturing timesteps, visualizing simulations, and can calculate a likelihood score indicating how closely a simulated tidal stream matches the observed positions of stream stars in the sky. We have developed the 'mwahpy' python package that allows easy access and analysis of N-body simulation results. All of the code is publicly available on GitHub. We are making this N-body code more accessible for general scientific use, and are seeking input from potential users.
AAS/Division of Dynamical Astronomy Meeting
- Pub Date:
- August 2020