Dual-level parallelism for ab initio molecular dynamics: Reaching teraflop performance with the CPMD code

Authors:
Jürg Hutter;Alessandro Curioni
Affiliations:
Physical Chemistry Institute, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland;IBM Research, Zurich Research Laboratory, 8803 Rüschlikon, Switzerland
Venue:
Parallel Computing
Year:
2005

Citing 1
Cited 7

Computer simulations for organic light-emitting diodes

IBM Journal of Research and Development

An Evaluation of Marenostrum Performance

International Journal of High Performance Computing Applications
A Microbenchmark Suite for Mixed-Mode OpenMP/MPI

IWOMP '09 Proceedings of the 5th International Workshop on OpenMP: Evolving OpenMP in an Age of Extreme Parallelism
Hybrid Message-Passing and Shared-Memory Programming in a Molecular Dynamics Application On Multicore Clusters

International Journal of High Performance Computing Applications
Parallel discrete element simulation of poly-dispersed granular material

Advances in Engineering Software
Early experience with scientific applications on the blue gene/l supercomputer

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing
Hybrid parallelization of a large-scale heart model

Facing the Multicore-Challenge II
Performance metrics in a hybrid MPI-OpenMP based molecular dynamics simulation with short-range interactions

Journal of Parallel and Distributed Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We show teraflop performance of the fully featured ab initio molecular dynamics code CPMD on an IBM pSeries 690 cluster. A mixed distributed-memory, coarse-grained parallel approach using the MPI library and shared-memory, fine-grained parallelism using OpenMP directives is used to optimally map the algorithms on the available hardware. The top performance achieved is ~20% of the peak performance and an estimated parallel efficiency of ~45% on 1024 processors for a system of 1000 atoms. The main limiting factor of parallel efficiency was found to be the latency of the interconnect.