Molecular Dynamics

Molecular dynamics applications are extremely amenable to the massively parallel architecture of NVIDIA’s GPUs. In the charts below, we highlight work done on VMD and also molecular dynamics software packages such as NAMD and HOOMD.

With the introduction of NVIDIA Tesla Bio Workbench, it provides bio-physicists and computational chemists the tools to push the boundaries of bio-chemical research, optimizing the scientific workflow and accelerating the pace of research. Learn more.

GPU TEST DRIVE

AMBER and NAMD 5x Faster
Run AMBER and NAMD 5x faster. Try it now on a free, remotely-hosted cluster and see how easy it is to reduce simulation time from days to hours.

Take a Free and Easy Test Drive Today


Generalized Born simulation in AMBER San Diego Supercomputing Center	Scaling of NAMD on a GPU-cluster Theoretical and Computational Bio-physics Group, UIUC

Key Molecular Dynamics ISVs and Applications Using CUDA

ISV	DESCRIPTION	GPU ADVANTAGE
AMBER 11	Simulation of mechanics force fields, implicit & explicit solvent on CUDA	10x speed-up: from minutes on CPUs to seconds on GPUs
NAMD	Designed for high-performance simulation of large molecular systems	Four GPUs outperform a cluster of 16 Quadcore CPUs
GROMACS	Simulation of biochemical molecules with complicated bond interactions	3-5x speed-up: accelerates PME and implicit solvent Generalized Born methods
VMD	Visualizing and analyzing large bio-molecular systems in 3-D graphics	20-100x speed-up: Making possible interactive, 3-D display of molecular systems
Schrodinger Core Hopping	Rapid screening of novel cores to improve drug properties	Faster, more accurate and innovative drug development
NWCHEM	Computational chemistry package designed for HPC clusters	Estimated Release Date: Q3 2010

Other Relevant Software Using CUDA

Technical Reports on Molecular Dynamics on CUDA

Presentations

GPU Technology Conference Sessions

Keynote: High-Throughput Science , Hanspeter Pfister, Harvard University
GPU Accelerated Molecular Dynamics with AMBER, The Scripps Research Institute and San Diego Supercomputer Center
GPU Accelerated Visualization and Analysis in VMD, University of Illinois at Champaign-Urbana
Computational Biophysics and Long Range Electrostatics on GPUs, NVIDIA
Volunteer Computing for GPUs: Petaflops for Free, UC Berkeley
Harnessing the GPU for Surgical Training and Preoperative Planning, CSIRO
GPU Accelerated Solvers for ODEs Describing Cardiac Membrane Equations, UC San Diego
Reconstructing the Brain: Extracting Neural Circuitry with CUDA and MPI, Harvard University
Unlocking Biologically-Inspired Computer Vision: a High-Throughput Approach, MIT
A Large Scale Simulation of Lattice QCD with a GPU Cluster, National Taiwan University

SC09 Presentations

Overview of NVIDIA CUDA Support in AMBER - Lessons Learned, Capabilities Gained, Ross Walker, University of California San Diego, San Diego Supercomputing Center
Harnessing GPU Speed to Accelerate LAMMPS Particle Simulation, Paul Crozier, Sandia National Laboratory
Accelerating Molecular Modeling Applications with GPU Computing, John Stone, Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana Champaign

CUDA-Acceleration in Related Verticals

See Also

MATLAB is a registered trademark of The MathWorks, Inc.
Jacket is a trademark of AccelerEyes