Parallel FMM-FFT solver for the analysis of hundreds of millions of unknowns

An efficient parallel implementation of the fast multipole method (FMM) combined with the fast fourier transform (FFT) is presented. The good scaling properties of the FMM-FFT, combined with a careful parallelization strategy, has shown to be very effective when using large parallel high performance supercomputers. For the case of very large problems, with hundreds of millions of unknowns, a nested scheme has been derived that further reduces the memory consumption. A challenging problem with more than 0.5 billion unknowns has been solved using this implementation, which demonstrates the ability of the algorithm to take advantage of the availability of supercomputers for the analysis of large, leading-edge electromagnetic problems.

[1]  L. Gurel,et al.  Efficient Parallelization of the Multilevel Fast Multipole Algorithm for the Solution of Large-Scale Scattering Problems , 2008, IEEE Transactions on Antennas and Propagation.

[2]  Xiao-Min Pan,et al.  A Sophisticated Parallel MLFMA for Scattering by Extremely Large Targets [EM Programmer's Notebook] , 2008, IEEE Antennas and Propagation Magazine.

[3]  R. Coifman,et al.  The fast multipole method for the wave equation: a pedestrian prescription , 1993, IEEE Antennas and Propagation Magazine.

[4]  J.L. Volakis,et al.  Massively Parallel Fast Multipole Method Solutions of Large Electromagnetic Scattering Problems , 2007, IEEE Transactions on Antennas and Propagation.

[5]  D. Wilton,et al.  Electromagnetic scattering by surfaces of arbitrary shape , 1980 .

[6]  L. Gurel,et al.  Hierarchical parallelization of the multilevel fast multipole algorithm for the efficient solution of large-scale scattering problems , 2008, 2008 IEEE Antennas and Propagation Society International Symposium.

[7]  L. Gurel,et al.  A Hierarchical Partitioning Strategy for an Efficient Parallelization of the Multilevel Fast Multipole Algorithm , 2009, IEEE Transactions on Antennas and Propagation.

[8]  Ozgur Ergul,et al.  Efficient Parallelization of Multilevel Fast Multipole Algorithm , 2006 .

[9]  Jiming Song,et al.  Fast Illinois solver code (FISC) , 1998 .

[10]  Jiming Song,et al.  Multilevel fast‐multipole algorithm for solving combined field integral equations of electromagnetic scattering , 1995 .

[11]  Weng Cho Chew,et al.  Scalable electromagnetic scattering computations , 2002, IEEE Antennas and Propagation Society International Symposium (IEEE Cat. No.02CH37313).

[12]  Ö. Ergül,et al.  Fast and accurate solutions of extremely large integral-equation problems discretised with tens of millions of unknowns , 2007 .

[13]  Ozgur Ergul,et al.  Parallel-MLFMA Solution of CFIE Discretized with Tens of Millions of Unknowns , 2007 .

[14]  J.M. Song,et al.  ScaleME: a portable scaleable multipole engine for electromagnetic and acoustic integral equation solvers , 1998, IEEE Antennas and Propagation Society International Symposium. 1998 Digest. Antennas: Gateways to the Global Network. Held in conjunction with: USNC/URSI National Radio Science Meeting (Cat. No.98CH36.

[15]  W.C. Chew,et al.  Solving large scale electromagnetic problems using a Linux cluster and parallel MLFMA , 1999, IEEE Antennas and Propagation Society International Symposium. 1999 Digest. Held in conjunction with: USNC/URSI National Radio Science Meeting (Cat. No.99CH37010).

[16]  S. Velamparambil,et al.  10 million unknowns: is it that big? [computational electromagnetics] , 2003, IEEE Antennas and Propagation Magazine.

[17]  Jiming Song,et al.  Monte Carlo simulation of electromagnetic scattering from two-dimensional random rough surfaces , 1997 .

[18]  G. Sylvand.  Performance of a parallel implementation of the FMM for electromagnetics applications , 2003 .

[19]  Y. Saad,et al.  GMRES: a generalized minimal residual algorithm for solving nonsymmetric linear systems , 1986 .

[20]  S. Velamparambil,et al.  Analysis and performance of a distributed memory multilevel fast multipole algorithm , 2005, IEEE Transactions on Antennas and Propagation.

[21]  Özgür Ergül,et al.  Solutions of Extremely Large Electromagnetics Problems Involving Tens of Millions of Unknowns Using Parallel MLFMA and Preconditioners , 2008 .