Abstract
To leverage the computational capability of modern supercomputers, existing algorithms need to be reformulated in a manner that allows for many concurrent operations. In this paper, we outline a framework that reformulates classical Schwarz waveform relaxation so that successive waveform iterates can be computed in a parallel pipeline fashion after an initial start-up cost. The communication costs for various implementations are discussed, and numerical scaling results are presented.
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
References
D. Bennequin, M.J. Gander, L. Halpern, A homographic best approximation problem with application to optimized Schwarz waveform relaxation. Math. Comput. 78(265), 185–223 (2009) [ISSN 0025-5718]
M. Bjørhus, On domain decomposition, subdomain iteration and waveform relaxation. Ph.D. thesis, University of Trondheim, Norway, 1995
A. Christlieb, C. Macdonald, B. Ong, Parallel high-order integrators. SIAM J. Sci. Comput. 32(2), 818–835 (2010)
Y. Courvoisier, M.J. Gander, Time domain Maxwell equations solved with Schwarz waveform relaxation methods, in Domain Decomposition Methods in Science and Engineering XX (Springer, Heidelberg, 2013), pp. 263–270
N. Feamster, H. Balakrishnan, J. Rexford, A. Shaikh, J. van der Merwe, The case for separating routing from routers, in Proceedings of the ACM SIGCOMM Workshop on Future Directions in Network Architecture (FDNA ’04) (ACM, New York, 2004), pp. 5–12. [ISBN 1-58113-942-X]
M.J. Gander, A.M. Stuart, Space-time continuous analysis of waveform relaxation for the heat equation. SIAM J. Sci. Comput. 19(6), 2014–2031 (1998) [ISSN 1064-8275]
M.J. Gander, L. Halpern, F. Nataf et al., Optimal convergence for overlapping and non-overlapping Schwarz waveform relaxation, in The Eleventh International Conference on Domain Decomposition Methods, Citeseer, 1999, ed. by C.H. Lai, P. Bjørstad, M. Cross, O. Widlund (1999), pp. 27–36
E. Giladi, H.B. Keller, Space-time domain decomposition for parabolic problems. Numer. Math. 93(2), 279–313 (2002) [ISSN 0029-599X]
C. Japhet, P. Omnes, Optimized Schwarz waveform relaxation for porous media applications, in Domain Decomposition Methods in Science and Engineering XX (Springer, Heidelberg, 2013), pp. 585–592
F. Lemarié, M. Patrick, L. Debreu, E. Blayo, Sensitivity of an Ocean-Atmosphere Coupled Model to the Coupling Method: Study of Tropical Cyclone Erica (2014). http://hal.inria.fr/hal-00872496
V. Tipparaju, W. Gropp, H. Ritzdorf, R. Thakur, J.L. Traff, Investigating high performance RMA interfaces for the MPI-3 standard, in International Conference on Parallel Processing, 2009, September 2009, pp. 293–300
S.G. Vandewalle, E.F. Van de Velde, Space-time concurrent multigrid waveform relaxation. Ann. Numer. Math. 1(1–4), 347–360 (1994) [ISSN 1021-2655]
Acknowledgement
This work was supported in part by Michigan State University through computational resources provided by the Institute for Cyber-Enabled Research and AFOSR Grant FA9550-12-1-0455. This work also used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number OCI-1053575.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Ong, B., High, S., Kwok, F. (2016). Pipeline Schwarz Waveform Relaxation. In: Dickopf, T., Gander, M., Halpern, L., Krause, R., Pavarino, L. (eds) Domain Decomposition Methods in Science and Engineering XXII. Lecture Notes in Computational Science and Engineering, vol 104. Springer, Cham. https://doi.org/10.1007/978-3-319-18827-0_36
Download citation
DOI: https://doi.org/10.1007/978-3-319-18827-0_36
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-18826-3
Online ISBN: 978-3-319-18827-0
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)