Adaptive Mesh Refinement for Large Scale Parallel Computing with DOLFIN

The design of a parallel implementation of the finite element software DOLFIN is described. To be able to handle large scale problems a new distributed mesh is developed. New distributed preand post processing routines are implemented in order to efficiently deal with the massive amount of data. New parallel mesh refinement algorithms are described and implemented. In order to obtain maximum parallel efficiency for adaptive computations, a dynamic load balancer is implemented. The entire implementation has been successfully tested on a massively parallel Blue Gene/L and a performance model is given to explain the experimental results.

[1]  Vipin Kumar,et al.  Parallel multilevel graph partitioning , 1996, Proceedings of International Conference on Parallel Processing.

[2]  George L.-T. Chiu,et al.  Overview of the Blue Gene/L system architecture , 2005, IBM J. Res. Dev..

[3]  Leonid Oliker,et al.  Experiments with Repartitioning and Load Balancing Adaptive Meshes , 1999 .

[4]  José E. Moreira,et al.  Blue Gene/L programming and operating environment , 2005, IBM J. Res. Dev..

[5]  A Performance Study of Diffusive vs . Remapped Load-Balancing Schemes , 1999 .

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

[7]  Chris Walshaw,et al.  Parallel mesh partitioning on distributed memory systems , 2002 .

[8]  John E. Savage,et al.  Parallel Refinement of Unstructured Meshes , 1999 .

[9]  Stephen T. Barnard.  PMRSB: Parallel Multilevel Recursive Spectral Bisection , 1995, SC.

[10]  Jarmo Rantakokko.  A Local Refinement Algorithm for Data Partitioning , 2000, PARA.

[11]  Vipin Kumar,et al.  Multilevel Graph Partitioning Schemes , 1995, ICPP.

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

[13]  Leonid Oliker,et al.  Parallel tetrahedral mesh adaptation with dynamic load balancing , 2013, Parallel Comput..

[14]  M. Rivara.  Selective refinement/derefinement algorithms for sequences of nested triangulations , 1989 .

[15]  Gary L. Miller,et al.  Geometric Mesh Partitioning: Implementation and Experiments , 1998, SIAM J. Sci. Comput..

[16]  William J. Schroeder,et al.  A Parallel Approach for Efficiently Visualizing Extremely Large, Time-Varying Datasets , 2000 .

[17]  Leonid Oliker,et al.  PLUM: Parallel Load Balancing for Unstructured Adaptive Meshes. Degree awarded by Colorado Univ. , 2013 .

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

[19]  Vipin Kumar,et al.  Multilevel Diffusion Schemes for Repartitioning of Adaptive Meshes , 1997, J. Parallel Distributed Comput..