Adaptive Load-Balancing for Force-Decomposition Based 3-Body Molecular Dynamics Simulations in A Heterogeneous Distributed Environment with Variable Number of Processors

2007 ◽  
Author(s):  
J.V. Sumanth ◽  
David R. Swanson ◽  
Hong Jiang
Keyword(s):  
Molecular Dynamics ◽  
Load Balancing ◽  
Molecular Dynamics Simulations ◽  
Variable Number ◽  
Distributed Environment ◽  
Dynamics Simulations ◽  
Force Decomposition

Fracture of Nanophase Ceramics: A Molecular-Dynamics Study

1996 ◽  
Vol 457 ◽  
Author(s):  
Aiichiro Nakano ◽  
Rajiv K. Kalia ◽  
Andrey Omeltchenko ◽  
Kenji Tsuruta ◽  
Priya Vashishta
Keyword(s):  
Molecular Dynamics ◽  
Silicon Nitride ◽  
Load Balancing ◽  
Molecular Dynamics Simulations ◽  
Dynamic Fracture ◽  
Parallel Computers ◽  
Fracture Surfaces ◽  
Affine Structures ◽  
Dynamics Simulations ◽  
Multiscale Algorithms

ABSTRACTNew multiscale algorithms and a load-balancing scheme are combined for molecular-dynamics simulations of nanocluster-assembled ceramics on parallel computers. Million-atom simulations of the dynamic fracture in nanophase silicon nitride reveal anisotropie self-affine structures and crossover phenomena associated with fracture surfaces.

scholarly journals Implementation of the force decomposition machine for molecular dynamics simulations

2012 ◽  
Vol 38 ◽  
pp. 243-247 ◽  
Author(s):  
Urban Borštnik ◽  
Benjamin T. Miller ◽  
Bernard R. Brooks ◽  
Dušanka Janežič
Keyword(s):  
Molecular Dynamics ◽  
Molecular Dynamics Simulations ◽  
Dynamics Simulations ◽  
Force Decomposition

scholarly journals The distributed diagonal force decomposition method for parallelizing molecular dynamics simulations

2011 ◽  
Vol 32 (14) ◽  
pp. 3005-3013 ◽  
Author(s):  
Urban Borštnik ◽  
Benjamin T. Miller ◽  
Bernard R. Brooks ◽  
Dušanka Janežič
Keyword(s):  
Molecular Dynamics ◽  
Molecular Dynamics Simulations ◽  
Decomposition Method ◽  
Dynamics Simulations ◽  
Force Decomposition

MOLECULAR DYNAMICS SIMULATIONS OF GRANULAR MATERIALS ON THE INTEL iPSC/860

1992 ◽  
Vol 03 (06) ◽  
pp. 1281-1293 ◽  
Author(s):  
GERALD H. RISTOW
Keyword(s):  
Molecular Dynamics ◽  
Load Balancing ◽  
Molecular Dynamics Simulations ◽  
Granular Materials ◽  
Distributed Memory ◽  
Two Dimensions ◽  
Three Dimensions ◽  
Parallel Computer ◽  
Dynamics Simulations ◽  
Specific Implementation

In this paper we present an efficient algorithm to perform Molecular Dynamics simulations on a distributed memory parallel computer, the Intel iPSC/860. The proposed model describes the flow properties of granular materials in two dimensions. The specific implementation on a 32 node iPSC/860, especially the message passing and load balancing algorithms, are discussed in detail. Performance data are shown for different computers and varying node numbers of the iPSC/860. As a physical example we calculate some properties of the outflow behavior from a two-dimensional hopper and we discuss possible extensions of our model to three dimensions. Our simulations show that Molecular Dynamics simulations can be implemented quite efficiently on a distributed memory parallel computer if one assures load balancing and optimizes the internode communications.

scholarly journals Scalable Molecular Dynamics for Large Biomolecular Systems

2000 ◽  
Vol 8 (3) ◽  
pp. 195-207 ◽  
Author(s):  
Robert K. Brunner ◽  
James C. Phillips ◽  
Laxmikant V. Kalé
Keyword(s):  
Molecular Dynamics ◽  
Load Balancing ◽  
Molecular Dynamics Simulations ◽  
Production Quality ◽  
Biomolecular Systems ◽  
Spatial Decomposition ◽  
Decomposition Scheme ◽  
Object Based ◽  
Dynamics Simulations ◽  
Performance Results

We present an optimized parallelization scheme for molecular dynamics simulations of large biomolecular systems, implemented in the production-quality molecular dynamics program NAMD. With an object-based hybrid force and spatial decomposition scheme, and an aggressive measurement-based predictive load balancing framework, we have attained speeds and speedups that are much higher than any reported in literature so far. The paper first summarizes the broad methodology we are pursuing, and the basic parallelization scheme we used. It then describes the optimizations that were instrumental in increasing performance, and presents performance results on benchmark simulations.

Sign in / Sign up

Export Citation Format

Share Document