Parallel simulation of three–dimensional bursting with MPI and OpenMP

Authors:
S. Tabik;L. F. Romero;E. M. Garzón;J. I. Ramos
Affiliations:
Depto de Arquitectura de Computadores y Electrónica, Universidad de Almería, Spain;Room I-320, E.T.S. Ingenieros Industriales, Universidad de Málaga, 29080, Spain;Depto de Arquitectura de Computadores y Electrónica, Universidad de Almería, Spain;Room I-320, E.T.S. Ingenieros Industriales, Universidad de Málaga, 29080, Spain
Venue:
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part II
Year:
2006

Citing 7
Cited 0

Linearization methods for reaction-diffusion equations: multidimensional problems

Applied Mathematics and Computation
Mathematical physiology

Mathematical physiology
Distributed Shared Memory: Concepts and Systems

Distributed Shared Memory: Concepts and Systems
Parallel scheduling of the PCG method for banded matrices rising from FDM/FEM

Journal of Parallel and Distributed Computing
Performance Evaluation of the SGI Altix 3700

ICPP '05 Proceedings of the 2005 International Conference on Parallel Processing
Optimizing Sparse Matrix-Vector Product Computations Using Unroll and Jam

International Journal of High Performance Computing Applications
Performance comparison of MPI and OpenMP on shared memory multiprocessors: Research Articles

Concurrency and Computation: Practice & Experience

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work presents a mathematical model and its parallel implementation via two parallel paradigms for the simulation of three–dimensional bursting phenomena. The mathematical model consists of four nonlinearly coupled partial differential equations and includes fast and slow subsystems. The differential equations have been discretized by means of a linearly–implicit finite difference method in equally–spaced grids. The resulting system of equations at each time level has been solved by means of an optimized Preconditioned Conjugate Gradient (PCG) method. The proposed mathematical model has been implemented via: (1) a message passing paradigm based on the standard MPI and (2) a shared address space paradigm based on SPMD OpenMP. The two implementations have been evaluated on two current parallel architectures, i.e., a cluster of biprocessors Xeon and an SGI Altix 3700 Bx2 based on Itanium. It is shown that better performance and scalability are obtained on the second platform.