Abstract
Facing the problem of implementing an efficient solver for partial differential equations, we are, in general, confronted with a certain quandary between numerical efficiency and efficiency in the usage of hardware resources: Modern numerical methods require the handling of hierarchical multilevel data on adaptively refined data structures, which are mostly represented by trees. On the other hand, as data access is one of the most important bottlenecks in high performance computing, we would wish to process data linearly with a high locality in time and space to be able to exploit the capability of cache hierarchies. In this paper, we show an approach based on space-filling curves as an odering mechanism for the cells of space-tree grids, with the help of which we can transform our (inherently highly non-local) data respresentation by trees to a few linearly processed data sets. As a consequence, we reach extremely high cache hit-rates above 99, 9%. In addition, the used methods make both parallelization and multigrid algorithms on adaptive grids with hierarchical data very straightforward and efficient.
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
Aftosmis, M.J., Berger, M.J., Adomavivius, G.: A Parallel Multilevel Method for adaptively Refined Cartesian Grids with Embedded Boundaries. AIAA Paper, 2000.
Bader, M., Bungartz, H.-J., Prank, A., Mundani, R.-P.: Space tree structures for PDE software. In: P.M.A. Sloot, et. al. (editors). Computational Science-ICCS 2002, International Conference, Amsterdam. Proceedings, Part III, Lecture Notes in Computer Science 2331, pages 662–671, Springer, 2002.
Blanke, C: Kontinuitätserhaltende Finite-Elemente-Diskretisierung der Navier-Stokes-Gleichungen. Diplomarbeit, Institut für Informatik, TU München, 2004.
Braess: Finite Elements. Theory, Fast Solvers and Applications in Solid Mechanics. Cambridge University Press, 2001.
Chatterjee, S., Sen, S.: Chache-Efficient Matrix Transposition. In: Proceedings of HPCA-6, pages 195–205, Toulouse, France, January 2000.
Chatterjee, S., Lebeck, A.R., Patnala, P.K., Thottethodi, M.: Recursive array layouts and fast parallel matrix multiplication. In: Proceedings of Eleventh Annual ACM Symposium on Parallel Algorithms and Architectures, pages 222–231, Saint-Malo, France, 1999.
W. Clarke. Key-based parallel adaptive refinement for FEM. Bachelor thesis, Australian National Univ., Dept. of Engineering, 1996.
Demaine, E.D.: Cache-Oblivious Algorithms and Data Structures. In: Lecture Notes from the EEF Summer School on Massive Data Sets, Lecture Notes in Computer Science, BRICS, University of Aarhus, Denmark, June 27–July 1, 2002, to appear.
Ferziger, J.H., Peri”c, M.: Computational Methods for Fluid Dynamics. Springer-Verlag, Berlin Heidelberg, 1999.
Frank, A.: Organisationsprinzipien zur Integration von geometrischer Modellierung, numerischer Simulation und Visualisierung. Doctoral thesis, Institut für Informatik, TU München, 2000.
Frigo, M., Leierson, C.E., Prokop, H., Ramchandran, S.: Cach-oblivious algorithms. In: Proceedings of the 40th Annual Sympoisium on Foundations of Computer Science, pages 285–297, New York, October 1999.
Girault, V., Raviart, P.: Finite Element Methods for Navier-Stokes Equations. Springer-Verlag, Berlin Heidelberg, 1986.
Gresho, P.M., Sani, R.L., Engelman, M.S.: Incompressible Flow and the Finite Element Method-Advection-Diffusion and Isothermal Laminar Flow. John Wiley & Sons Ltd, Chichester, 1998.
Griebel, M., Dornseifer, T., Neunhoeffer, T.: Numerische Simulation in der Str”omungsmechanik. Vieweg, 1995.
Griebel, M.: Multilevelverfahren als Iterationsmethoden über Erzeugendensystemen. Habilitationsschrift, TU München, 1993.
Griebel, M., Knapek, S., Zumbusch, G., Caglar, A.: Numerische Simulation in der Moleküldynamik. Numerik, Algorithmen, Parallelisierung, Anwendungen, Springer, Berlin, Heidelberg, 2004.
Griebel, M., Zumbusch, G.W.: Parallel multigrid in an adaptive PDE solver based on hashing and space-filling curves. Parallel Computing, 25, 827–843, 1999.
Griebel, M., Zumbusch, G.W.: Hash based adaptive parallel multilevel methods with space-filling curves. In: H. Rollnik and D. Wolf, editors, NIC Series, 9, 479–492, Germany, 2002. Forschungszentrum Jülich.
Günther, F.: Eine cache-optimale Implementierung der Finite-Elemente-Methode. Doctoral thesis, Institut für Informatik, TU München, 2004.
Günther, F., Krahnke, A., Langlotz, M., Mehl, M., Pögl, M., Zenger, Ch.: On the Parallelization of a Cache-Optimal Iterative Solver for PDEs Based on Hierarchical Data Structures and Space-Filling Curves. Conference Proceedings EuroPVMMPI 2004, Budapest, September 2004, LNCS, Springer, to appear.
Günther, F., Mehl, M., Pögl, M., Zenger, Ch.: A cache-aware algorithm for PDEs on hierarchical data structures. Conference Proceedings PARA’ 04, Kopenhagen, June 2004, LNCS, Springer, submitted.
Günther, F., Mehl, M., Pögl, M., Zenger, Ch.: A cache-aware algorithm for PDEs on hierarchical data structures based on space-filling curves. SIAM Journal on Scientific Computing, in review.
Oden, J.T., Para, A., Feng, Y.: Domain decomposition for adaptive hp finite element methods. In: D.E. Keyes and J. Xu, editors, Domain decomposition methods in scientific and engineering computing, proceedings of the 7th int. conf. on domain decomposition, vol. 180 of Contemp. Math., pages 203–214, 1994, Pennsylvania State Universitiy.
Patra, A.K., Long, J., Laszloff, A.: Efficient Parallel Adaptive Finite Element Methods Using Self-Scheduling Data and Computations. HiPC, pages 359–363, 1999.
Pögl, M.: Entwicklung eines cache-optimalen 3D Finite-Element-Verfahrens für große Probleme. Doctoral thesis, Institut für Informatik, TU München, 2004.
Prokop, H.: Cache-Oblivious Algorithms. Master Thesis, Massachusetts Institute of Technology, 1999.
S. Roberts, S. Klyanasundaram, M. Cardew-Hall, and W. Clarke. A key based parallel adaptive refinement technique for finite element methods. In: B.J. Noye, M.D. Teubner, and A.W. Gill, editors, Proc. Computational Techniques and Apüplications: CTAC’ 97, pages 577–584, World Scientific, Singapore, 1998.
Sagan, H.: Space-Filling Curves. Springer-Verlag, New York, 1994.
Stevens, R.J., Lehar, A.F., Preston, F.H.: Manipulation and Presentation of Multidimensional Image Data Using the Peano Scan. IEEE Trans. Pattern An. and Machine Intelligence, Vol PAMI-5, pages 520–526, 1983.
Velho, L., de Miranda Gomes, J.: Digital Halftoning with Space-Filling Curves. Computer Graphics, 25, 81–90, 1991.
Verstappen, R.W.C.P., Veldman, A.E.P.: Symmetry-Preserving Discretization of Turbulent Channel Flow. In: M. Breuer, F. Durst, C. Zenger (editors): High Performance Scientific and Engineering Computing, Springer-Verlag, Berlin Heidelberg, 2001.
Zumbusch, G.W.: Adaptive Parallel Multilevel Methods for Partial Differential Equations. Habilitationsschrift, Universität Bonn, 2001.
Zumbusch, G.W.: On the quality of space-filling curve induced partitions. Z. Angew. Math. Mech., 81, 25–28, 2001. Suppl. 1, also as report SFB 256, University Bonn, no. 674, 2000.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Langlotz, M., Mehl, M., Weinzierl, T., Zenger, C. (2005). SkvG: Cache-Optimal Parallel Solution of PDEs on High Performance Computers Using Space-Trees and Space-Filling Curves. In: Bode, A., Durst, F. (eds) High Performance Computing in Science and Engineering, Garching 2004. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-28555-5_7
Download citation
DOI: https://doi.org/10.1007/3-540-28555-5_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-26145-2
Online ISBN: 978-3-540-28555-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)