Abstract
In this paper we present methods for efficient parallelization of the solution of pressure Poisson equation arising in 3D CFD forest fire modeling. The solution procedure employs the Conjugate Gradient method with implicit Modified ILU (MILU) preconditioner. The basic idea for parallelizing recursive incomplete-decomposition algorithms is to use a direct nested twisted approach in combination with a staircase method. Parallelization of MILU-CG solver is implemented in OpenMP environment for Non-uniform memory (NuMA) computer systems. Performance results of the parallelized algorithm are presented and analyzed for different number of processors (up to 16).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Morvan, D., Dupuy, J.L.: Modeling of Fire Spread through a Forest Fuel Bed Using a Multiphase Formulation. Combust. Flame 127, 1981–1994 (2001)
Morvan, D., Dupuy, J.L.: Modeling the Propagation of a Wildfire through a Mediterranean Shrub Using a Multiphase Formulation. Combust. Flame 138, 199–210 (2004)
Accary, G., Bessonov, O., Fougère, D., Meradji, S., Morvan, D.: Optimized Parallel Approach for 3D Modelling of Forest Fire Behaviour. In: Malyshkin, V.E. (ed.) PaCT 2007. LNCS, vol. 4671, pp. 96–102. Springer, Heidelberg (2007)
Scarella, G., Accary, G., Meradji, S., Morvan, D., Bessonov, O.: Three-Dimensional Numerical Simulation of the Interaction between Natural Convection and Radiation in a Differentially Heated Cavity in the Low Mach Number Approximation. In: CHT 2008 International Symposium on Advances in Computational Heat Transfer, CHT-08-193, Begell House, Inc. (2008)
Shewchuk, J.R.: An Introduction to the Conjugate Gradient Method without the Agonizing Pain. School of Computer Science, Carnegie Mellon University, Pittsburgh (1994)
Ortega, J.M.: Introduction to Parallel and Vector Solution of Linear Systems. Plenum Press, New York (1988)
Saad, Y.: Iterative Methods for Sparse Linear Systems. PWS Publishing, Boston (2000)
van der Vorst, H.A.: Large Tridiagonal and Block Tridiagonal Linear Systems on Vector and Parallel Computers. Par. Comp. 5, 45–54 (1987)
Bastian, P., Horton, G.: Parallelization of Robust Multi-Grid Methods: ILU-Factorization and Frequency Decomposition Method. SIAM J. Stat. Comput. 12, 1457–1470 (1991)
Elizarova, T., Chetverushkin, B.: Implementation of Multiprocessor Transputer System for Computer Simulation of Computational Physics Problems (in Russian). Mathematical Modeling 4(11), 75–100 (1992)
Vuik, C., van Nooyen, R.R.P., Wesseling, P.: Parallelism in ILU-Preconditioned GMRES. Par. Comp. 24, 1927–1946 (1998)
Gustafsson, I.: A Class of First Order Factorization Methods. BIT 18, 142–156 (1978)
Dagum, L., Menon, R.: OpenMP: An Industry-Standard API for Shared-Memory Programming. IEEE Computational Science and Engineering 5(1), 46–55 (1998)
Benzi, M.: Preconditioning Techniques for Large Linear Systems: A Survey. J. Comput. Phys. 182, 418–477 (2002)
Stone, H.L.: Iterative Solution of Implicit Approximations of Multidimensional Partial Differential Equations. SIAM J. Numer. Anal. 5, 530–558 (1968)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Accary, G., Bessonov, O., Fougère, D., Gavrilov, K., Meradji, S., Morvan, D. (2009). Efficient Parallelization of the Preconditioned Conjugate Gradient Method. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2009. Lecture Notes in Computer Science, vol 5698. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-03275-2_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-03275-2_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-03274-5
Online ISBN: 978-3-642-03275-2
eBook Packages: Computer ScienceComputer Science (R0)