Issues of parallel software development for the domain decomposition methods
Keywords:domain decomposition method, parallel algorithm, scalability, data structures, numerical experiment
Various aspects of parallel software development for the domain decomposition methods are considered: the application of MPI programming technology for cluster systems, the choice points in the design of parallel programs for the domain decomposition methods, the need to implement a matrix action without its explicit representation, the work with index sets in the software implementation of restriction and continuation operators as well as in the data exchange between subdomains. On a series of numerical experiments for a model problem, the questions of the best choice of the configuration of launching an executable program on a cluster are studied to minimize the computation time and a strategy for performing such experiments is proposed.
- D. S. Butyugin, Y. L. Guryeva, V. P. Il’in, et al., “Parallel Algebraic Solvers Library Krylov,” Vestn. South Ural Univ. Ser. Vychisl. Mat. Inf. 2 (3), 92-105 (2013).
- Y. Saad, Iterative Methods for Sparse Linear Systems (SIAM, Philadelphia, 2003; Mosk. Gos. Univ., Moscow, 2013).
- Y. Saad, “A Flexible Inner-Outer Preconditioned GMRES Algorithm,” SIAM J. Sci. Comput. 14 (2), 461-469 (1993).
- Y. L. Gurieva, V. P. Il’in, and D. V. Perevozkin, “Algebraic-Geometric and Information Structures of Domain Decomposition Methods,” Vychisl. Metody Programm. 17, 132-146 (2016).
- MPI Forum.
https://www.mpi-forum.org . Cited May 22, 2018.
- Message Passing Interface (MPI).
http://parallel.ru/vvv/mpi.html . Cited May 22, 2018.
- S. Pissanetzky, Sparse Matrix Technology (Academic, London, 1984; Mir, Moscow, 1988).
- X.-C. Cai, C. Farhat, and M. Sarkis, “A Minimum Overlap Restricted Additive Schwarz Preconditioner and Applications in 3D Flow Simulations,” Contemp. Math. 218, 479-485 (1998).
- X.-C. Cai and M. Sarkis, “A Restricted Additive Schwarz Preconditioner for General Sparse Linear Systems,” SIAM J. Sci. Comput. 21 (2), 792-797 (1999).
- Intel Parallel Programming Professional (Introduction).
- V. D. Korneev, Parallel Programming with MPI (Inst. Comput. Math. Math. Geophys., Novosibirsk, 2002) [in Russian].
- Biconjugate Gradient Method.
http://en.wikipedia.org/wiki/Biconjugate_gradient_method . Cited May 22, 2018.
- L. Giraud and R. S. Tuminaro, Algebraic Domain Decomposition Preconditioners , Technical Report ENSEEIHT-IRIT RT/APO/06/07 (Université Paul Sabatier, Toulouse, 2006).
- MUMPS: MUltifrontal Massively Parallel sparse direct Solver.
http://mumps.enseeiht.fr . Cited May 22, 2018.
- Siberian Supercomputing Center.
http://www2.sscc.ru/Default.htm . Cited May 22, 2018.
- Intel Trace Analyzer and Collector.
https://software.intel.com/en-us/intel-trace-analyzer . Cited May 22, 2018.
- Intel Math Kernel Library.
https://software.intel.com/en-us/mkl . Cited May 22, 2018.
- S. A. Lebedev, I. B. Meerov, A. V. Sysoev, et al., “Optimization and Application of the MUMPS Package for Solving the Three-Dimensional Stationary Strength Problems on Cluster Systems,” in Proc. Int. Supercomputer Conf., Novorossiysk, Russia, September 23-28, 2013 (Mosk. Gos. Univ., Moscow, 2013), pp. 233-237.