NVIDIA Home > Products > High Performance Computing > Tesla Bio Workbench > NAMD on Tesla GPUs

 

The Team at University of Illinois at Urbana-Champaign (UIUC) has been enabling CUDA-acceleration on NAMD since 2007. They have performed scaling experiments on the NCSA Tesla-based Lincoln cluster and demonstrated that 4 Tesla GPUs can outperform a cluster with 16 quad core CPUs.

Download and Installation

• Download NAMD 2.7 beta 2 Binaries (UIUC website)
• Release Notes

Benchmark Data

NAMD scales very well on a Tesla GPU cluster as demonstrated by these results.

Data courtesy of Theoretical and Computational Bio-physics Group, UIUC

Technical Papers

There are several papers on GPU acceleration in NAMD and VMD available at the UIUC website.
To learn more, you can also refer to the NVIDIA molecular dynamics page to learn more about accelerating molecular dynamics applications using CUDA.

Presentations

• Publications and presentations on GPU acceleration in VMD and NAMD by Klaus Schluten’s Group
• Discussion of VMD and NAMD from SC09: Accelerating Molecular Modeling Applications with GPU Computing - John Stone, Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana Champaign
• Discussion of VMD and NAMD from 2009 GPU Technology Conference: GPU Accelerated Visualization and Analysis in VMD - John Stone, Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana Champaign
• More publications on accelerating molecular dynamics applications using GPUs
• More publications on accelerating quantum chemistry applications using GPUs

Discussion Forums

• GPU computing communities on GPUcomputing.net

Interviews / Videos

• SC09 BLOG on NVIDIA nTersect: Studying the H1N1 Virus using NVIDIA GPUs
• SC08 Interview of James Phillips, Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana Champaign

The Tesla Bio Workbench applications can be deployed on GPU-based desktop personal supercomputers or in data center solutions. Built on the revolutionary massively parallel CUDA architecture, these solutions are designed to accelerate the pace of computational science.

RECOMMENDED HARDWARE CONFIGURATION

Desktop Workstation Configuration	Data Center Configuration
GPUs 4 Tesla C1060 Computing CPU and Main Memory 2.33 GHz x86 CPU >16 GB (4 GB main memory per Tesla C1060 GPU)	GPUs per node Either Tesla S1070 (with 4 GPUs in a 1U) or hybrid servers with M1060 GPUs CPU and Main Memory 2.33 GHz x86 CPU per server 16 GB per server

WORKSTATION SOLUTIONS TESLA PERSONAL SUPERCOMPUTER For personal supercomputing at your desk Learn more > Where to buy >		DATA CENTER SOLUTIONS TESLA GPU COMPUTING CLUSTERS For computing with large-scale installations Learn more > Where to buy >