Dynamic load balancing for large-scale adaptive finite element computation

For the parallel computation of partial differential equations, one key is the grid partitioning. It requires that each process owns the same amount of computations, and also, the partitioning quality should be proper to reduce the communications among processes. When calculating the partial differential equations using adaptive finite element methods, the grid and the basis functions adjust in each iteration, which introduce load balancing issues. The grid should be redistributed dynamically. This paper studies dynamic load balancing algorithms and the implementation on the adaptive finite element platform PHG. The numerical experiments show that algorithms studied in this paper have good partitioning quality, and they are efficient.

[1]  Mark T. Jones,et al.  Computational Results for Parallel Unstructured Mesh Computations , 1994 .

[2]  Bruce Hendrickson,et al.  A Multi-Level Algorithm For Partitioning Graphs , 1995, Proceedings of the IEEE/ACM SC95 Conference.

[3]  Robert D. Falgout,et al.  hypre: A Library of High Performance Preconditioners , 2002, International Conference on Computational Science.

[4]  Joachim Schöberl,et al.  NETGEN An advancing front 2D/3D-mesh generator based on abstract rules , 1997 .

[5]  Zhangxin Chen,et al.  Accelerating iterative linear solvers using multiple graphical processing units , 2015, Int. J. Comput. Math..

[6]  Joachim Sch.  NETGEN An advancing front 2D/3D-mesh generator based on abstract rules , 1997 .

[7]  Hui Liu,et al.  GPU-based Parallel Reservoir Simulation for Large-scale Simulation Problems , 2012 .

[8]  Vipin Kumar,et al.  A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs , 1998, SIAM J. Sci. Comput..

[9]  George Karypis,et al.  Parmetis parallel graph partitioning and sparse matrix ordering library , 1997 .

[10]  Zhangxin John Chen,et al.  Parallel Preconditioners for Reservoir Simulation on GPU , 2012 .

[11]  Zhangxin Chen,et al.  Reservoir Simulation on NVIDIA Tesla GPUs , 2012 .

[12]  J. D. Teresco,et al.  A Comparison of Zoltan Dynamic Load Balancers for Adaptive Computation ∗ , 2022 .

[13]  P. Campbell,et al.  Dynamic Octree Load Balancing Using Space-Filling Curves ∗ , 2003 .

[14]  Hui Liu,et al.  ACCELERATING PRECONDITIONED ITERATIVE LINEAR SOLVERS ON GPU , 2014 .

[15]  Karen Dragon Devine,et al.  Parallel adaptive hp -refinement techniques for conservation laws , 1996 .

[16]  Song Yu,et al.  SPARSE MATRIX-VECTOR MULTIPLICATION ON NVIDIA GPU , 2012 .

[17]  Shahid H. Bokhari,et al.  A Partitioning Strategy for Nonuniform Problems on Multiprocessors , 1987, IEEE Transactions on Computers.

[18]  George Cybenko,et al.  Dynamic Load Balancing for Distributed Memory Multiprocessors , 1989, J. Parallel Distributed Comput..

[19]  Song Yu,et al.  DEVELOPMENT OF A RESTRICTED ADDITIVE SCHWARZ PRECONDITIONER FOR SPARSE LINEAR SYSTEMS ON NVIDIA GPU , 2014 .

[20]  Horst D. Simon,et al.  Partitioning of unstructured problems for parallel processing , 1991 .

[21]  Bruce Hendrickson,et al.  Tinkertoy parallel programming: a case study with Zoltan , 2005, Int. J. Comput. Sci. Eng..

[22]  Zhangxin Chen,et al.  Parallel construction of Hamiltonian paths for conforming tetrahedral meshes , 2013, Int. J. Comput. Math..

[23]  Y. F. Hu,et al.  Load Balancing for Unstructured Mesh Applications , 1999, Scalable Comput. Pract. Exp..

[24]  Alex Pothen,et al.  PARTITIONING SPARSE MATRICES WITH EIGENVECTORS OF GRAPHS* , 1990 .

[25]  Zhang,et al.  A Parallel Algorithm for Adaptive Local Refinement of Tetrahedral Meshes Using Bisection , 2009 .

[26]  Bruce Hendrickson,et al.  Dynamic load balancing in computational mechanics , 2000 .

[27]  Curt Jones,et al.  A Heuristic for Reducing Fill-In in Sparse Matrix Factorization , 1993, PPSC.

[28]  Karen D. Devinea,et al.  New Challenges in Dynamic Load Balancing , 2004 .

[29]  Zhangxin Chen,et al.  GPU-Based Parallel Reservoir Simulators , 2014 .

[30]  J. Tinsley Oden,et al.  Problem decomposition for adaptive hp finite element methods , 1995 .

[31]  P. K. Jimack.  An overview of dynamic load-balancing for parallel adaptive computational mechanics codes , 2000 .

[32]  Hui Liu,et al.  Existence and construction of Hamiltonian paths and cycles on conforming tetrahedral meshes , 2011, Int. J. Comput. Math..

[33]  Joel H. Saltz,et al.  A Hypergraph-Based Workload Partitioning Strategy for Parallel Data Aggregation , 2001, PPSC.

[34]  Leonid Oliker,et al.  Efficient load balancing and data remapping for adaptive grid calculations , 1997, SPAA '97.

[35]  Ümit V. Çatalyürek,et al.  Decomposing Irregularly Sparse Matrices for Parallel Matrix-Vector Multiplication , 1996, IRREGULAR.