Abstract
This paper presents an experimental comparison of dynamic partitioning techniques for blockwise parallel structured adaptive mesh refinement applications. A new partitioning technique, G-MISP, is described. Policies for the automatic selection of partitioner based on application and system state are outlined. Adaptive methods for the numerical solution to partial differential equations yield highly advantageous ratios for cost/accuracy compared to methods based upon static uniform approximations. Distributed implementations offer the potential for accurate solution of physically realistic models of important applications. They also lead to interesting challenges in dynamic resource allocation, e.g. dynamic load balancing. The results show that G-MISP is preferable for communication dominated cases where the block graph has high granularity. Recommendations for appropriate partitioning techniques, given application and system state, are given. It was found that our classification model needs to be extended to accurately capture the behavior of the cases studied.
This research was supported by the Swedish Foundation for Strategic Research via the program of Industrial Computational Mathematics.
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
Preview
Unable to display preview. Download preview PDF.
References
Stephen T. Barnard and Horst D. Simon. A fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems. Technical report, NAS Systems Division, NASA Ames Research Center, 1994.
M. J. Berger and P. Colella. Local adaptive mesh refinement for shock hydrodynamics. Journal of Computational Physics, 82, 1989.
Marsha J. Berger and Joseph Oliger. Adaptive mesh refinement for hyperbolic partial differential equations. Journal of Computational Physics, 53, 1983.
L. Ferm and P. Lötstedt. Blockwise adaptive grids with multigrid acceleration for compressible flow. AIAA J., 37: 121–123, 1999.
G. Karypis, K. Schloegel, and V. Kumar. PARMETIS— parallel graph partitioning and sparse matrix ordering library, version 2.0. Univ. of Minnesota, Minneapolis, MN, 1998.
P. Lötstedt and S. Söderberg. Parallel solution of hyperbolic PDEs with space-time adaptivity. In D. Hänel R. Vilsmeier, F. Benkhaldour, editor, Finite Volumes for Complex Applications II, pages 769–776, Paris, 1999. Hermes Science.
Chao-Wei Ou and Sanjay Ranka. Parallel remapping algorithms for adaptive problems. Technical report, Center for Research on Parallel Computation, Rice University, 1994.
M. Parashar, et al. A common data management infrastructure for adaptive algorithms for PDE solutions. Technical Paper at Supercomputing’ 97, 1997.
John R. Pilkington and Scott B. Baden. Partitioning with spacefilling curves. Technical Report CS94-34, CSE, UCSD, March 1994.
K. G. Powell, et al. A solution-adaptive upwind scheme for ideal magnetohydrodynamics. J. Comput. Phys., 154: 284–309, 1999.
Jarmo Rantakokko. Data Partitioning Methods and Parallel Block-Oriented PDE Solvers. PhD thesis, Uppsala University, 1998.
Hans Sagan. Space-filling curves. Springer-Verlag, 1994.
H. Samet. The design and analysis of spatial data structure. Addison-Wesley, Reading, Massachusetts, 1989.
K. Schloegel, et al. A performance study of diffusive vs. remapped load balancing schemes. Technical Report 98-018, Dept. of Computer Science, Univ. of Minnesota, Minneapolis, MN, 1998.
Mausumi Shee, Samip Bhavsar, and Manish Parashar. Characterizing the performance of dynamic distribution and load-balancing techniques for adaptive grid hierarchies. 1999.
Johan Steensland. http://www.tdb.uu.se/~johans/.. Work in progress, 1999.
Johan Steensland. Vampire homepage, http://www.caip.rutgers.edu/~johans/vampire/. 2000.
Johan Steensland, Sumir Chandra, Manish Parashar, and Michael Thuné. Characterization of domain-based partitioners for parallel samr applications. Submitted to PDCS2000, 2000.
Erlendur Steinthorsson and David Modiano. Advanced methodology for simulation of complex ows using structured grid systems. ICOMP, 28, 1995.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Steensland, J., Söderberg, S., Thuné, M. (2001). A Comparison of Partitioning Schemes for Blockwise Parallel SAMR Algorithms. In: Sørevik, T., Manne, F., Gebremedhin, A.H., Moe, R. (eds) Applied Parallel Computing. New Paradigms for HPC in Industry and Academia. PARA 2000. Lecture Notes in Computer Science, vol 1947. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-70734-4_20
Download citation
DOI: https://doi.org/10.1007/3-540-70734-4_20
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41729-3
Online ISBN: 978-3-540-70734-9
eBook Packages: Springer Book Archive