Revealing Feasibility of FMM on ASIC: Efficient Implementation of N-Body Problem on FPGA

Abstract

FPGAs have been improved significantly in terms of performance and capacity over the last 20 years. The scale of FPGA based design also sparked off the demands for high-level synthesis to handle complicated applications. A well known intricate application is the FMM (Fast Multipole Method) algorithm of N-body problem, which is so complicated that it was not implemented on FPGA as reported in literature. In this paper, we take high level modeling and design tools, i.e. Simulink and System Generator to implement major modules in FMM algorithm to solve the N-Body problem on FPGA. Besides the impressive performance speedup on FPGA, we improve the efficiency by merging the circuits for the common logic among modules in the algorithm. Our experience in efficiently implementing the FMM algorithm will be taken as a useful reference for researchers working on FPGA applications as well as high performance computing.