Abstract
The efficient utilization of parallel computational capabilities of modern hardware architecture is a must in large scale industrial applications. In this paper we focus on the parallelization of algebraic multigrid (AMG) in general and identify the respective challenges imposed on any hierarchical iterative linear solver. Moreover, we summarize the strategies employed in the parallel implementation of our SAMG library to cope with these issues and present some performance indicators of SAMG in real-world industrial applications.
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
Similar content being viewed by others
Notes
- 1.
SAMG is in fact used very successfully by our customers with more than 3000 cores.
- 2.
Note, however, that due to customer demand parallel SAMG can be run in a (low efficiency) mode which comes very close to a serial algorithm.
References
G.E. Moore, Cramming more components onto integrated circuits. Electronics 8, 114–117 (1965)
K. Stüben, Europort-D: commercial benefits of using parallel technology, in Parallel Computing: Fundamentals, Applications and New Directions, ed. by E. D’Hollander, F. Peters, G. Joubert, U. Trottenberg, R. Völpel. Advances in Parallel Computing, vol. 12 (North-Holland, Amsterdam, 1998), pp. 61–78
K. Stüben, J.W. Ruge, T. Clees, S. Gries, Algebraic multigrid — from academia to industry, in Scientific Computing and Angorithms in Industrial Simulation — Projects and Products of Fraunhofer SCAI, ed. by M. Griebel, A. Schüller, M.A. Schweitzer (Springer, New York, 2017)
U.M. Yang, Parallel algebraic multigrid methods — high performance preconditioners, in Numerical Solution of Partial Differential Equations on Parallel Computers, ed. by A.M. Bruaset, A. Tveito (Springer, Berlin, Heidelberg, 2006), pp. 209–236
MPI Forum. http://mpi-forum.org. Accessed 31 Mar 2017
STREAM: Sustainable Memory Bandwidth in High Performance Computers. https://www.cs.virginia.edu/stream/. Accessed 31 Mar 2017
The OpenMP API Specification for Parallel Programming. http://www.openmp.org. Accessed 31 Mar 2017
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this chapter
Cite this chapter
Plum, HJ. et al. (2017). Parallel Algebraic Multigrid. In: Griebel, M., Schüller, A., Schweitzer, M. (eds) Scientific Computing and Algorithms in Industrial Simulations. Springer, Cham. https://doi.org/10.1007/978-3-319-62458-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-62458-7_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-62457-0
Online ISBN: 978-3-319-62458-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)