HOOMD-Blue, General-Purpose Many-Body Dynamics On the GPU

Monday, October 17, 2011: 1:34 PM
Duluth (Hilton Minneapolis)
Joshua A. Anderson1, Carolyn L. Phillips2, Trung D. Nguyen1 and Sharon C. Glotzer3, (1)Chemical Engineering, University of Michigan, Ann Arbor, MI, (2)Applied Physics, University of Michigan, Ann Arbor, MI, (3)Chemical Engineering and Biomedical Engineering, University of Michigan

HOOMD-blue[1] is an open source code for performing molecular dynamics and related many-body dynamics simulations on graphics processing units (GPUs). Current GPUs are widely available, inexpensive, and capable of teraflops of floating-point computation and hundreds of gigabytes per second of memory bandwidth.  In typical benchmarks, HOOMD-blue on a current generation GPU is about 80-100 times faster than on a single CPU core.  HOOMD-blue has a growing international user and developer base.

HOOMD-blue functionality includes standard molecular dynamics code features and options, including NVT, NPT, NVE and Brownian dynamic integrators and diverse pair, bond, and angle potentials. Release 0.10.0 includes rigid body constraints [2], which are commonly used in a wide range of molecular modeling applications from the atomistic scale, modeling the bonds in molecules such as water, carbon dioxide, and benzene, to the colloidal scale, modeling macroscopic rods, plates and patchy nanoparticles.  HOOMD-blue also includes long-range electrostatics computed with the PPPM (particle-particle/particle-mesh) method [3].  Simulation methods such as dissipative particle dynamics[4], energy minimization[2], and constraint forces have also been added. 

In this talk, we present an overview of HOOMD-blue and its capabilities, discuss the GPU algorithms that drive it, and demonstrate the performance that it delivers.

1] Find HOOMD-blue online at: http://codeblue.umich.edu/hoomd-blue/

2] Nguyen TD, Phillips CL, Anderson JA, and Glotzer SC, “Rigid body constraints realized in massively-parallel molecular dynamics on graphics processing units,” submitted.

3] Contributed by Barr S, Mertmann P, and  Panagiotopoulos AZ.  

4] Phillips CL, Anderson JA, and Glotzer SC, “Pseudo-Random Number Generation for Brownian Dynamics and Dissipative Particle Dynamics Simulations on GPU Devices”, submitted.

Extended Abstract: File Not Uploaded