Molecular dynamics on a distributed-memory multiprocessor
Journal of Computational Chemistry
Journal of Computational Chemistry
Achieving scalable parallel molecular dynamics using dynamic spatial domain decomposition techniques
Journal of Parallel and Distributed Computing - Special issue on dynamic load balancing
NAMD2: greater scalability for parallel molecular dynamics
Journal of Computational Physics - Special issue on computational molecular biophysics
Global arrays: a portable "shared-memory" programming model for distributed memory computers
Proceedings of the 1994 ACM/IEEE conference on Supercomputing
IPPS '98 Proceedings of the 12th. International Parallel Processing Symposium on International Parallel Processing Symposium
Proceedings of the 6th ACM conference on Computing frontiers
Dynamic topology aware load balancing algorithms for molecular dynamics applications
Proceedings of the 23rd international conference on Supercomputing
Hi-index | 0.00 |
NWChem is a computational chemistry software suite developed for massively parallel computers in the W. R. Wiley Environmental Molecular Sciences Laboratory at the U.S. Department of Energy's Pacific Northwest National Laboratory. This software integrates a range of modules for computational chemistry applications, including classical molecular dynamics simulations and quantum mechanical calculations. This contribution provides details of the classical molecular dynamics module and focuses on issues related to load balancing on massively parallel computers, in particular the IBM SPTM and the Cray T3ETM as examples of distributed and shared memory massively parallel architectures. The implementation of the molecular dynamics module of NWChem is based on a domain decomposition of the chemical system, taking advantage of the distribution of data to reduce the memory requirements and the locality of intermolecular interactions to reduce the communication requirements. This approach results in a more complex implementation because of the requirement of periodic atomic reassignments and the need for sophisticated load-balancing techniques.