M2L optimization in FMBEM and its GPU implementation

Abstract

The translation from multipole moments to local moments (M2L) in the fast multipole boundary element method (FMBEM) costs too much time; we compare three methods of M2L optimization from the three following aspects: accuracy, efficiency and memory usage with an engineering numerical example, and then present a GPU parallel algorithm using CUDA for one of the front three methods which transfers child cell's coefficients to their father cell, meanwhile, improve the tree structure by redefining the whole cells in different levels which can avoid writing data conflict in the parallel strategy. Finally, we use the threedimensional elastic BEM problems of chassis parts to verify the algorithm, and the result shows that the accelerating effect of this method is significant.

A CAD/CAE incorporate software framework using a unified representation architecture

“…However, the BEM application has so far been limited to relatively small problems since the memory and computational complexity. In order to overcome the long response time defect, a software (fast multipole BEM) and hardware (GPU parallel computing) accelerated algorithm is used to speed up the analysis process which is illustrated in detail in the literature [46][47][48][49]. This block is also responsible for checking the accuracy, validating the performance.…”

“…The integration can be implemented in the same framework without external data exchange. The computation module adopts software accelerating algorithm (FMBEM) and hardware accelerating algorithm (GPU) to enlarge the scale of the solution and shorten the iterative time of solving [46][47][48][49][50][51][52]. The application scenario of the proposed integrated CAD/CAE framework system is as follows:…”