Volume 9, Issue 2
Parallel Data Partitioning Strategy in Solving Large Scale Electromagnetic Scattering Problems

Y. Hu, W. Tong, X. Wang & X. Zhi

Int. J. Numer. Anal. Mod., 9 (2012), pp. 257-269

Published online: 2012-09

Preview Purchase PDF 1 3308
Export citation
  • Abstract
The multilevel fast multipole algorithm (MLFMA) has shown great efficiency in solving large scale electromagnetic scattering problems. However, when unknowns become up to tens of millions, it is not trivial to keep high performance because of the complicated structure and calculation of MLFMA. In order to get rid of the bottleneck caused by load balancing, a parallel data partitioning strategy is proposed based on the hierarchical structure of an oct-tree of MLFMA. We present our data partitioning strategy in the light of different layers' properties including the processing of three kinds of layers in the tree and a fine-grained decomposition. We also put forward a solution of a coexisting data correlating problem, using a transition layer. Meanwhile, with the purpose of minimizing communication time in distributed memory system, a redundant technique is applied in the distributed layer. Parallel efficiency analysis demonstrates that the computational cost in parallelization of MLFMA can be asymptotically cut, and a high parallel efficiency can be obtained in our implementation.
  • Keywords

Multilevel Fast Multipole Algorithm (MLFMA) Parallel Data Partitioning Strategy Hierarchical Structure Data Correlating Problem and Redundant Technique

  • AMS Subject Headings

35R35 49J40 60G4

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{IJNAM-9-257, author = {Y. Hu, W. Tong, X. Wang and X. Zhi}, title = {Parallel Data Partitioning Strategy in Solving Large Scale Electromagnetic Scattering Problems}, journal = {International Journal of Numerical Analysis and Modeling}, year = {2012}, volume = {9}, number = {2}, pages = {257--269}, abstract = {The multilevel fast multipole algorithm (MLFMA) has shown great efficiency in solving large scale electromagnetic scattering problems. However, when unknowns become up to tens of millions, it is not trivial to keep high performance because of the complicated structure and calculation of MLFMA. In order to get rid of the bottleneck caused by load balancing, a parallel data partitioning strategy is proposed based on the hierarchical structure of an oct-tree of MLFMA. We present our data partitioning strategy in the light of different layers' properties including the processing of three kinds of layers in the tree and a fine-grained decomposition. We also put forward a solution of a coexisting data correlating problem, using a transition layer. Meanwhile, with the purpose of minimizing communication time in distributed memory system, a redundant technique is applied in the distributed layer. Parallel efficiency analysis demonstrates that the computational cost in parallelization of MLFMA can be asymptotically cut, and a high parallel efficiency can be obtained in our implementation.}, issn = {2617-8710}, doi = {https://doi.org/}, url = {http://global-sci.org/intro/article_detail/ijnam/625.html} }
TY - JOUR T1 - Parallel Data Partitioning Strategy in Solving Large Scale Electromagnetic Scattering Problems AU - Y. Hu, W. Tong, X. Wang & X. Zhi JO - International Journal of Numerical Analysis and Modeling VL - 2 SP - 257 EP - 269 PY - 2012 DA - 2012/09 SN - 9 DO - http://doi.org/ UR - https://global-sci.org/intro/article_detail/ijnam/625.html KW - Multilevel Fast Multipole Algorithm (MLFMA) KW - Parallel Data Partitioning Strategy KW - Hierarchical Structure KW - Data Correlating Problem KW - and Redundant Technique AB - The multilevel fast multipole algorithm (MLFMA) has shown great efficiency in solving large scale electromagnetic scattering problems. However, when unknowns become up to tens of millions, it is not trivial to keep high performance because of the complicated structure and calculation of MLFMA. In order to get rid of the bottleneck caused by load balancing, a parallel data partitioning strategy is proposed based on the hierarchical structure of an oct-tree of MLFMA. We present our data partitioning strategy in the light of different layers' properties including the processing of three kinds of layers in the tree and a fine-grained decomposition. We also put forward a solution of a coexisting data correlating problem, using a transition layer. Meanwhile, with the purpose of minimizing communication time in distributed memory system, a redundant technique is applied in the distributed layer. Parallel efficiency analysis demonstrates that the computational cost in parallelization of MLFMA can be asymptotically cut, and a high parallel efficiency can be obtained in our implementation.
Y. Hu, W. Tong, X. Wang & X. Zhi. (1970). Parallel Data Partitioning Strategy in Solving Large Scale Electromagnetic Scattering Problems. International Journal of Numerical Analysis and Modeling. 9 (2). 257-269. doi:
Copy to clipboard
The citation has been copied to your clipboard