Iterative methods for solving large-scale problems of structural mechanics using multi-core computers

• Autorzy: Fialko, S. Y.
• Języki publikacji: EN

Abstrakty

EN

The paper studies the conjugate gradient method for solving systems of linear algebraic equations with symmetric sparse matrices that arise when the finite-element method is applied to the problems of structural mechanics. The main focus is on designing effective preconditioning and parallelizing the method for multi-core desktop computers. Preconditioning is based on the incomplete Cholesky “by value” factorization method and implemented based on the technique of sparse matrices, which allows increasing convergence considerably without a significant increase of the computer's resources. Parallelization is implemented for the incomplete factorization as well as for iterative process stages. The method is integrated into the SCAD software package (www.scadsoft.com). The paper includes a discussion of the results of calculations done with direct and iterative methods for large-scale design models of tall buildings, originally from the SCAD Soft1 problem collection.

Słowa kluczowe

PL

metoda elementów skończonych; macierze rzadkie; komputery wielordzeniowe; budynki wielopiętrowe

EN

finite element method; preconditioned conjugate gradient method; sparse matrices; multi-core computers; multistory buildings

• Czasopismo: Archives of Civil and Mechanical Engineering
• Rocznik: 2014
• Tom: Vol. 14, no. 1
• Strony: 190--203
• Opis fizyczny: Bibliogr. 23 poz., rys., tab., wykr.

Twórcy

autor

Fialko, S. Y.

• Department of Physics, Mathematics and Applied Computer Science, Cracow University of Technology, Kraków 31-155, Poland,

Bibliografia

• [1] P.R. Amestoy, I. S Duff., J. Y. L’Excellent, Multi frontal parallel distributed symmetric and un symmetric solvers, Computer Methods in Applied Mechanics and Engineering 184 (2000) 501–520.
• [2] BLAS-Basic Linear Algebra Subprograms, 2008. URL: http:// www.netlib.org/blas/ (accessed 29.04.12).
• [3] Eun-Jin Im, K. Yelick, R. Vuduc, Sparsity: optimization framework for sparse matrix kernels, International Journal of High Performance Computing Applications 18 (2004) 135–158.
• [4] A. George, J.W.H. Liu, Computer solution of Sparse Positive Definite Systems, Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 1981.
• [5] A. George, J.W.H. Liu, The evolution of the minimum degree ordering algorithm, SIAM Review 31 (1989) 1–19.
• [6] G.H. Golub, C.F. Van Loan, Matrix Computations, 3rd ed., John Hopkins University Press, 1996.
• [7] N.I.M. Gould, Y. Hu, J.A. Scott, A Numerical Evaluation of Sparse Direct Solvers for the Solution of Large Sparse, Symmetric Linear Systems of Equations. Technical Report RAL-TR-2005-005, Ruther ford Appleton Laboratory, 2005.
• [8] J.W. Demmel, Applied Numerical Linear Algebra, SIAM, Philadelphia, 1997.
• [9] IMS Ls Fortran Library Features, Fortran Library Documentation, 2012. http://www.roguewave.com/support/ product-documentation/imsl-numerical-libraries/fortran-library. aspx (accessed 29.04.12).
• [10] Intels Math Kernel Library Reference Manual, Document Number: 630813-029US,2012. http://www.intel.com/ software/products/mkl/docs/WebHelp/mkl.htm (accessed 29.04.12).
• [11] Intel (R) Math Kernel Library for Windows OS User's Guide, Document Number: 315930-013US. Intel (R) MKL10.3- Windows OS. Threaded Functions and Problems, 2012. http://software.intel.com/sites/products/documentation/ hpc/composerxe/en-us/mklxe/mkl_userguide_win/ MKL_UG_managing_performance/Threaded_Routines. htm#blas (accessed 29.04.12).
• [12] A. Jennings, Development of an ICCG algorithm for large sparse systems, in: D.J. Evans (Ed.), Preconditioned Methods. Theory and Applications, Gordon and Breach, Science Publishers, Inc., 1983, pp.425–438.
• [13] S. Fialko, PARFES: a method for solving finite element linear equations on multi-core computers, Advances in Engineering Software 40 (12) (2010) 1256–1265.
• [14] S. Fialko, The Block Substructure Multi frontal Method for Solution of Large Finite Element Equation SETS, Technical Transactions,1-NP/2009, issue8, (2009) 175–188 (in Polish).
• [15] S. Fialko, High-performance aggregation element-by-element iterative solver for large-scale complex shell structure problems, Archives of Civil Engineering XLV (2) (1999) 193–207.
• [16] S. Fialko, A sparsein complete Cholesky conjugat egradient method for finite element analysis of large-scale problems in structural mechanics, In: Proceedings of the CMM-2007-Computer Methods in Mechanics, Lodz-Spala, Poland, June 19–22, 2007, pp. 145–146.
• [17] S. Fialko, A parallel sparse direct finite element solver for desktop computers, In: Proceedings of the 19th International Conference on Computer Methods in Mechanics, Warsaw, Poland, 9–12 May 2011, pp. 183–184.
• [18] K. Malkowski, Ingyu Lee, P. Raghavan, M.J. Irwin, Conjugate gradient sparse solver: performance-power characteristics, In: Proceedings of the Parallel and Distributed Processing Symposium, 2006.
• [19] METIS-Serial Graph Partitioning and Fill-Reducing Matrix Ordering. 2012. http://glaros.dtc.umn.edu/gkhome/metis/ metis/overview (accessed 29.04.12).
• [20] M. Papadrakakis, Solving Large-Scale Problems in Mechanics, John Wiley & Sons Ltd., 1993.
• [21] A.V. Perelmuter, S.Y. Fialko, Problems of computational mechanics relateto finite-element analysis of structural constructions, International Journal for Computational Civil and Structural Engineering 1 (2) (2005) 72–86.
• [22] O. Schenk, K. Gartner, Two-level dynamics cheduling in PARDISO: improved scalability on shared memory multiprocessing systems, Parallel Computing 28 (2002) 187–197.
• [23] M. Suarjana, H. Kincho Law, A robust in complete factorization based on value and space constraints, International Journal for Numerical Methods in Engineering 38 (1995) 1703–1719.

Identyfikatory

DOI

10.1016/j.acme.2013.05.009