Analysis of Parallel Preconditioned Conjugate Gradient Algorithms

The conjugate gradient method is an iterative technique used to solve systems of linear equations. The paper analyzes the performance of parallel preconditioned conjugate gradient algorithms. First, a theoretical model is proposed for estimation of the complexity of PPCG method and a scalability analysis is done for three different data decomposition cases. Computational experiments are done on IBM SP4 computer and some results are presented. It is shown that theoretical predictions agree well with computational results.

[1]  Takumi Washio,et al.  Ordering strategies and related techniques to overcome the trade-off between parallelism and convergence in incomplete factorizations , 1999, Parallel Comput..

[2]  Roger W. Hockney,et al.  Performance parameters and benchmarking of supercomputers , 1991, Parallel Comput..

[3]  Magolu Monga-Made.  Ordering strategies for modified block incomplete factorizations , 1995 .

[4]  Raimondas Čiegis,et al.  On the Efficiency of Scheduling Algorithms for Parallel Gaussian Elimination with Communication Delays , 2000, PARA.

[5]  Henk A. van der Vorst,et al.  Developments and trends in the parallel solution of linear systems , 1999, Parallel Comput..

[6]  Evripidis Bampis,et al.  Scheduling Algorithms for Parallel Gaussian Elimination With Communication Costs , 1998, IEEE Trans. Parallel Distributed Syst..

[7]  Goran Lj. Djordjevic,et al.  A Heuristic for Scheduling Task Graphs with Communication Delays Onto Multiprocessors , 1996, Parallel Comput..

[8]  Graham F. Carey,et al.  PCG: a software package for the iterative solution of linear systems on scalar, vector and parallel computers , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.

[9]  SANGBACK MA.  Comparisons of the Parallel Preconditioners for Large Nonsymmetric Sparse Linear Systems on a Parallel Computer , 2004, Int. J. High Speed Comput..

[10]  L. Dixon,et al.  Parallel algorithms for global optimization , 1993 .

[11]  A. Samarskii.  The Theory of Difference Schemes , 2001 .

[12]  Panos M. Pardalos,et al.  Topics in parallel computing in mathematical programming , 1993, Applied discrete mathematics and theoretical computer science.

[13]  Panos M. Pardalos,et al.  Parallel algorithms for global optimization problems , 1996, Solving Combinatorial Optimization Problems in Parallel.

[14]  Vipin Kumar,et al.  Performance and Scalability of Preconditioned Conjugate Gradient Methods on Parallel Computers , 1995, IEEE Trans. Parallel Distributed Syst..

[15]  Henk A. van der Vorst,et al.  A generalized domain decomposition paradigm for parallel incomplete LU factorization preconditionings , 2001, Future Gener. Comput. Syst..

[16]  Ganesh Thiagarajan,et al.  Parallelization strategies for element-by-element preconditioned conjugate gradient solver using high-performance Fortran for unstructured finite-element applications on linux clusters , 2002 .

[17]  George Karypis,et al.  Introduction to Parallel Computing , 1994 .

[18]  Yousef Saad,et al.  PSPARSLIB Users Manual: A Portable Library of parallel Sparse Iterative Solvers , 1998 .

[19]  Hasan Dag,et al.  A parallel implementation of chebyshev preconditioned conjugate gradient method , 2003, Second International Symposium on Parallel and Distributed Computing, 2003. Proceedings..

[20]  Wei-Pai Tang,et al.  Ordering Methods for Preconditioned Conjugate Gradient Methods Applied to Unstructured Grid Problems , 1992, SIAM J. Matrix Anal. Appl..

[21]  Raimondas Čiegis,et al.  Parallel numerical modelling of short laser pulse compression , 2005 .

[22]  Mardochée Magolu monga Made.  Ordering Strategies for Modified Block Incomplete Factorizations , 1995, SIAM J. Sci. Comput..

[23]  Raimondas Ciegis,et al.  Realistic Performance Prediction Tool for the Parallel Block LU Factorization Algorithm , 2003, Informatica.

[24]  Gene H. Golub,et al.  Matrix computations , 1983 .