Strong scaling of general-purpose molecular dynamics simulations on GPUs
Authors:
Jens Glaser,
Trung Dac Nguyen,
Joshua A. Anderson,
Pak Lui,
Filippo Spiga,
Jaime A. Millan,
David C. Morse,
Sharon C. Glotzer
Abstract:
We describe a highly optimized implementation of MPI domain decomposition in a GPU-enabled, general-purpose molecular dynamics code, HOOMD-blue (Anderson and Glotzer, arXiv:1308.5587). Our approach is inspired by a traditional CPU-based code, LAMMPS (Plimpton, J. Comp. Phys. 117, 1995), but is implemented within a code that was designed for execution on GPUs from the start (Anderson et al., J. Com…
▽ More
We describe a highly optimized implementation of MPI domain decomposition in a GPU-enabled, general-purpose molecular dynamics code, HOOMD-blue (Anderson and Glotzer, arXiv:1308.5587). Our approach is inspired by a traditional CPU-based code, LAMMPS (Plimpton, J. Comp. Phys. 117, 1995), but is implemented within a code that was designed for execution on GPUs from the start (Anderson et al., J. Comp. Phys. 227, 2008). The software supports short-ranged pair force and bond force fields and achieves optimal GPU performance using an autotuning algorithm. We are able to demonstrate equivalent or superior scaling on up to 3,375 GPUs in Lennard-Jones and dissipative particle dynamics (DPD) simulations of up to 108 million particles. GPUDirect RDMA capabilities in recent GPU generations provide better performance in full double precision calculations. For a representative polymer physics application, HOOMD-blue 1.0 provides an effective GPU vs. CPU node speed-up of 12.5x.
△ Less
Submitted 10 December, 2014;
originally announced December 2014.
Facetted patchy particles through entropy-driven patterning of mixed ligand SAMS
Authors:
Aaron Santos,
Jaime Andres Millan,
Sharon C. Glotzer
Abstract:
We present a microscopic theory that describes the ordering of two distinct ligands on the surface of a faceted nanoparticle. The theory predicts that when one type of ligand is significantly bulkier than all others, the larger ligands preferentially align themselves along the edges and vertices of the nanoparticle. Monte Carlo simulations confirm these predictions. We show that the intrinsic conf…
▽ More
We present a microscopic theory that describes the ordering of two distinct ligands on the surface of a faceted nanoparticle. The theory predicts that when one type of ligand is significantly bulkier than all others, the larger ligands preferentially align themselves along the edges and vertices of the nanoparticle. Monte Carlo simulations confirm these predictions. We show that the intrinsic conformational entropy of the ligands stabilizes this novel edge-aligned phase.
△ Less
Submitted 12 December, 2011;
originally announced December 2011.
