Parallel implementation of the steepest descent fast multipole method (SDFMM) on a Beowulf cluster for subsurface sensing applications

D. Jiang, W. Meleis, M. El-Shenawee, E. Mizan, M. Ashouei, C. Rappaport
2002 IEEE Microwave and Wireless Components Letters  
We present the parallel, MPI−based implementation of the SDFMM computer code using a thirty two−node Intel Pentium−based Beowulf cluster. The SDFMM is a fast algorithm that is a hybridization of the Method of Moment (MoM), the Fast Multipole Method (FMM) and the Steepest Descent Integration rule (SDP), which is used to solve large−scale linear systems of equations produced electromagnetic scattering problems. An overall speedup of 7.2 has been achieved on the 32−processor Beowulf cluster and a
more » ... ignificant reduced runtime is achieved on the 4−processor 667MHz Alpha workstation.
doi:10.1109/7260.975724 fatcat:s4hymlljevebxptwgpihgr2xhq