Swarm-NG library

1.0

Swarm-NG is a collection of code to help scientists and engineers harness the power of GPUs. In the early releases, Swarm-NG will focus on the integration of an ensemble of N-body systems evolving under Newtonian gravity. There are existing libraries to perform the force calculation for large-N systems on GPUs. Swarm-NG does not replicate these, but rather focuses on integrating an ensemble of many systems where N is small. This is of a partiular interest for astronomers who study the chaotic evolution of planetary systems. In the long term, we hope Swarm-NG will allow for the efficient parallel integration of user-defined systems of ordinary differential equations.

Authors:

Aaron Boley, Saleh Dindar, Eric B. Ford, Mario Juric, Young In Yeo, and others from the Swarm-NG development team

Table of contents

• Getting Started with Swarm-NG
• Using command-line interface
• **Using C++ library interface**
  • Beginner tutorial for using the API
  • Advanced tutorial for using GPU integrators
  • Tutorial on using the Ensemble data structure.
• **Using the plugin system to extend the library**
  • Tutorial for Making a Propagator
  • Tutorial for making a monitor (stopper/logger)
  • Tutorial for implementing an Integrator

• Release Notes

Licensing, Citing & Acknowledging Swarm-NG

Citing Swarm-NG

When writing papers which made use of Swarm-NG, we strongly encourage you to cite the following technical paper describing Swarm-NG and the Swarm-NG Website.

• Dindar, S., E.B. Ford, M. Juric, Y.I. Yeo, J. Gao, A.C. Boley, B. Nelson, J. Peters "Swarm-NG: A CUDA Library for Parallel n-body Integrations with focus on Simulations of Planetary Systems", to be submitted to New Astronomy. arXiv:1208.1157v1 [astro-ph.EP].

• Website: http://www.astro.ufl.edu/~eford/code/swarm/docs

Licensing Information

• Swarm-NG v1.0 may be licensed under the GPL v3, that is included in the Swarm-NG release (docs/gpl-3.0.txt).

Acknowledgements

Swarm-NG is the product of a collaboration between astrophysicists and computer scientists. The major contributors to the current Swarm-NG codebase are Saleh Dindar, Eric Ford and Mario Juric.

The Swarm-NG library includes code reused or adapted from:

• the nVidia CUDA SDK (http://www.nvidia.com/object/cuda_home_new.html),
• the GNU Science Library (http://www.gnu.org/software/gsl/),
• Sverre Aarseth’s hermit2 code (http://www.ast.cam.ac.uk/~sverre/web/pages/nbody.htm), and
• John Chamber’s Mercury code (http://www.arm.ac.uk/~jec/home.html), which builds on a long history of n-body integrators, including Hal Levison and Martin Duncan’s Swift code (http://www.boulder.swri.edu/~hal/swift.html).

We are happy to acknowledge financial support from:

• the University of Florida’s Research Opportunity Seed Fund,
• the National Science Foundation (grant CCF-0728797; http://www.nsf.gov/),
• NASA’s Hubble Fellowship program (http://www.stsci.edu/institute/org/spd/hubble-fellowship/hubble-fellow-overview), and
• especially NASA’s Applied Information Systems Research Program (grant NNX09AM41G; http://aisrp.nasa.gov/).

We acknowledge additional support from:

• the University of Florida Astronomy Department (http://astro.ufl.edu/),
• UF CISE Prof. Jorg Peters' SurfLab (http://www.cise.ufl.edu/research/SurfLab/)
• UF High Performance Computing Center (http://www.hpc.ufl.edu/),
• the Harvard-Smithsonian Center for Astrophysics (http://www.cfa.harvard.edu/) and the Institute for Theory and Computation (http://www.cfa.harvard.edu/itc/),
• the Institute for Advanced Study (http://sns.ias.edu/), and
• nVidia Corporation (http://www.nvidia.com).

Further Information

Additional documentation is avaliable from http://www.astro.ufl.edu/~eford/code/swarm/docs/

For a Google Group with mailing list archives see http://groups.google.com/group/swarm-ng

For more information see http://www.astro.ufl.edu/~eford/code/swarm/

Source code is hosted on http://github.com/AstroGPU/swarm .