A Comparison of Partitioning Schemes for Blockwise Parallel SAMR Algorithms

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.

[1]  Manish Parashar,et al.  Characterizing the Performance of Dynamic Distribution and Load-Balancing Techniques for Adaptive Grid Hierarchies , 1999 .

[2]  Manish Parashar,et al.  Characterization of domain-based partitioners for parallel SAMR applications , 2000 .

[3]  J.C. Browne,et al.  A Common Data Management Infrastructure for Adaptive Algorithms for PDE Solutions , 1997, ACM/IEEE SC 1997 Conference (SC'97).

[4]  Horst D. Simon,et al.  Fast multilevel implementation of recursive spectral bisection for partitioning unstructured problems , 1994, Concurr. Pract. Exp..

[5]  H. Sagan.  Space-filling curves , 1994 .

[6]  Sanjay Ranka,et al.  Parallel remapping algorithms for adaptive problems , 1995, Proceedings Frontiers '95. The Fifth Symposium on the Frontiers of Massively Parallel Computation.

[7]  Jarmo Rantakokko.  Data Partitioning Methods and Parallel Block-Oriented PDE Solvers , 1998 .

[8]  Erlendur Steinthorsson,et al.  Advanced methodology for simulation of complex flows using structured grid systems , 1995 .

[9]  Vipin Kumar,et al.  A Parallel Algorithm for Multilevel Graph Partitioning and Sparse Matrix Ordering , 1998, J. Parallel Distributed Comput..

[10]  Per Lötstedt,et al.  Parallel solution of hyperbolic PDEs with space-time adaptivity , 1999 .

[11]  P. Colella,et al.  Local adaptive mesh refinement for shock hydrodynamics , 1989 .

[12]  Hanan Samet,et al.  The Design and Analysis of Spatial Data Structures , 1989 .

[13]  P. Roe,et al.  A Solution-Adaptive Upwind Scheme for Ideal Magnetohydrodynamics , 1999 .

[14]  M. Berger,et al.  Adaptive mesh refinement for hyperbolic partial differential equations , 1982 .

[15]  Per Lötstedt,et al.  Blockwise adaptive grids with multigrid acceleration for compressible flow , 1999 .