Abstract
A new and simple method is proposed to choose the best data configuration in terms of processing phase time according to previous probing of edge-based matrix-vector products for codes using iterative solvers in unstructured grid problems. This method is realized as a suite of routines named EdgePack, acting during both pre-solution and solution phase, based on data locality optimization techniques and variations of matrix-vector product algorithm. Results have been demonstrating the great flexibility and simplicity of this method, which is suitable for distributed memory platforms in which different data configurations can coexist.
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
Burgess, D.A., Giles, M.B.: Renumbering unstructured grids to improve the performance of codes on hierarchical memory machines. Advances in Engineering Software 28, 189–201 (1997)
Carey, G.F., Swift, S., McLey, R.T.: Maximizing sparse matrix-vector product performance on RISC based MIMD computers. Journal of Parallel and Distributed Computing 37, 146–158 (1996)
Douglas, C.C., et al.: Cache Optimization for Structured and Unstructured Grid Multigrid. Electronic Transactions on Numerical Analysis 10, 21–40 (2000)
Gropp, W.D., et al.: Performance Modeling and Tuning of an Unstructured Mesh CFD Application. In: Proceedings of SC 2000, Dallas, Texas, United States, IEEE Computer Society Press, Los Alamitos (2000)
Löhner, R.: Renumbering Strategies for unstructured-grid solvers operating on shared-memory, cache-based parallel machines. Computer Methods in Applied Mechanics and Engineering 163, 95–109 (1998)
Oliker, L., et al.: Evaluation of cache-based superscalar and cacheless vector architectures for scientific computations. In: Proceedings of the 18th Annual International Conference on Supercomputing, Malo, France (2004)
Oliker, L., et al.: Parallel Conjugate Gradient: Effects of Ordering Strategies, Programming Paradigms, and Architectural Platforms. IEEE Transactions on Parallel and Distributed Systems 11(9), 931–940 (2000)
Oliker, L., et al.: Ordering Unstructured Meshes for Sparse Matrix Computations on Leading Parallel Systems. In: Rolim, J.D.P. (ed.) IPDPS-WS 2000. LNCS, vol. 1800, pp. 497–503. Springer, Heidelberg (2000)
Oliker, L., et al.: Effects of Ordering Strategies and Programming Paradigms on Sparse Matrix Computations. SIAM Review 44(3), 373–393 (2002)
Pinar, A., Heath, M.T.: Improving Performance of Sparse Matrix-Vector Multiplication, Conference on High Performance Networking and Computing. In: Proceedings of the 1999 ACM/IEEE Conference on Supercomputing (CDROM), Portland, Oregon, United States, IEEE Computer Society Press, Los Alamitos (1999)
Ribeiro, F.L.B., Coutinho, A.L.G.A.: Comparison between element, edge and compressed storage schemes for iterative solutions in finite element analyses. International Journal for Numerical Methods in Engineering 63(4), 569–588 (2005)
Vuduc, R., et al.: Performance Optimizations and Bounds for Sparse Matrix-Vector Multiply. In: Conference on High Performance Networking and Computing, Proceedings of the 2002 ACM/IEEE Conference on Supercomputing, Baltimore, Maryland, pp. 1–35. IEEE Computer Society Press, Los Alamitos (2002)
Peraire, J., Peiro, J., Morgan, K.: Multigrid solution of the 3d-compressible Euler equations on unstructured grids. Int. J. Num. Meth. Engrg. 36(6), 1029–1044 (1993)
Luo, H., Baum, J.D., Löhner, R.: Edge-based finite element scheme for the Euler equations. AIAA Journal 32(6), 1183–1190 (1994)
Coutinho, A.L.G.A., et al.: Edge-based finite element techniques for non-linear solid mechanics problems. Int. J. for Num. Meth. in Engrg. 50(9), 2053–2068 (2001)
Catabriga, L., Coutinho, A.L.G.A.: Implicit SUPG solution of Euler equations using edge-based data structures. Computer Methods in Applied Mechanics and Engineering 32, 3477–3490 (2002)
Martins, M.A.D., Alves, J.L.D., Coutinho, A.L.G.A.: Parallel Edged-Based Finite Techniques for Nonlinear Solid Mechanics. In: Palma, J.M.L.M., Dongarra, J.J., Hernández, V. (eds.) VECPAR 2000. LNCS, vol. 1981, pp. 506–518. Springer, Heidelberg (2001)
Elias, R.N., Martins, M.A.D., Coutinho, A.L.G.A.: Parallel Edge-Based Inexact Newton Solution of Steady Incompressible 3D Navier-Stokes Equations. In: Cunha, J.C., Medeiros, P.D. (eds.) Euro-Par 2005. LNCS, vol. 3648, pp. 1237–1245. Springer, Heidelberg (2005)
Coutinho, A.L.G.A., et al.: Performance comparison of data reordering algorithms for sparse matrix-vector multiplication in edge-based unstructured grid computations. Int. J. Num. Meth. Engng, accepted
Karypis, G., Kumar, V.: Metis 4.0: Unstructured Graph Partitioning and Sparse Matrix Ordering System. Technical report, Department of Computer Science, University of Minnesota, Minneapolis (1998), http://www.users.cs.umn.edu/~karypis/metis
Löhner, R., Galle, M.: Minimization of indirect addressing for edge-based field solvers. Communications in Numerical Methods in Engineering 18, 335–343 (2002)
Löhner, R.: Some useful renumbering strategies for unstructured grids. International Journal for Numerical Methods in Engineering 36, 3259–3270 (1993)
Sydenstricker, R.M., et al.: Edge-Based Interface Elements for Solution of Three-Dimensional Geomechanical Problems. In: Palma, J.M.L.M., et al. (eds.) VECPAR 2002. LNCS, vol. 2565, pp. 53–64. Springer, Heidelberg (2003)
Cuthill, E., McKee, J.: Reducing the bandwidth of sparse symmetric matrices. In: Proc. ACM Nat. Conf., pp. 157–172. ACM Press, New York (1969)
Barth, T.J.: Numerical Aspects of Computing Viscous High Reynolds Number Flows on Unstructured Meshes. In: AIAA, 29th Aerospace Sciences Meeting, AIAA 91-0721, Reno, Nevada, January 7-10, 1991, pp. 91–721 (1991)
Löhner, R.: Edges, Stars, Superedges and Chains. Comp. Meth. Appl. Mech. Eng. 111, 255–263 (1994)
Kalro, V., Tezduyar, T.E.: Parallel 3D Computation of Unsteady Flows around Circular Cylinders. Parallel Computing 23, 1235–1248 (1997)
Williamson, C.H.K.: Defining a Universal and Continuous Strouhal-Reynolds Number Relationship for the Laminar Vortex Shedding of a Circular Cylinder. Phys. Fluids 31, 2742–2744 (1988)
Baranyi, L.: Computation of Unsteady Momentum and Heat Transfer from a Fixed Circular Cylinder in Laminar Flow. Journal of Computational and Applied Mechanics 4(1), 13–25 (2003)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer Berlin Heidelberg
About this paper
Cite this paper
Martins, M., Elias, R., Coutinho, A. (2007). EdgePack: A Parallel Vertex and Node Reordering Package for Optimizing Edge-Based Computations in Unstructured Grids. In: Daydé, M., Palma, J.M.L.M., Coutinho, Á.L.G.A., Pacitti, E., Lopes, J.C. (eds) High Performance Computing for Computational Science - VECPAR 2006. VECPAR 2006. Lecture Notes in Computer Science, vol 4395. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-71351-7_23
Download citation
DOI: https://doi.org/10.1007/978-3-540-71351-7_23
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-71350-0
Online ISBN: 978-3-540-71351-7
eBook Packages: Computer ScienceComputer Science (R0)