Paper title:  NAMD ? a Parallel, Object-Oriented Molecular Dynamics Program
Journal: International Journal of High Performance Computing Applications (1996) vol. 10, no. 4, 251-268

Author(s): Nelson, M.T., Humphrey, W., Gursoy, A., Dalke, A., Kale, L.V., Skeel, R.D., and Schulten, K..
*Theoretical Biophysics Group, Beckman Institute, University of Illinois, Urbana, Illinois 61801

Summary:
The paper discusses the development, features, implementation and application of NAMD, a parallel molecular dynamics (MD) program, for simulating large biomolecular systems in parallel programs.

NAMD was designed to enhance the performance of molecular dynamics simulations to systems with a large number of atoms by making it scalable to parallel computers. It has features which are the same as that of typical MD programs but with additional elements that makes it easier to use and to be manipulated by the user for specific conditions. These include NVE ensemble dynamics, velocity rescaling, Langevin dynamics, vacuum and spherical boundary conditions, NVE and NPT ensemble dynamics, ability to fix atoms positions. It can also attain full electrostatic computation without incurring high computational cost through the use of Distributed Parallel Multipole Tree Algorithm (DPMTA). It is highly modular and implemented using C++ so users may modify the code easily according to their system preference. It uses spatial decomposition with multi-threaded, message-driven design for efficient scaling to multiple processors that is tolerant of communication latency. Aside from these, NAMD can be connected to a molecular graphics program to visually simulate the behavior of molecules and modify a currently running simulation.

The paper concludes with an example on using NAMD. The software was applied to a protein-water system of 30,000+ atoms.


Contribution and application:

NAMD is a powerful computational engine which may be used by ESEL to elucidate the fouling mechanism that happens in membranes during SWRO desalination process.



By: Hannah Ebro
hannah@gist.ac.kr