The document describes a hardware acceleration solution for N-body simulation using an FPGA. N-body simulation involves calculating the forces between a large number of bodies over multiple time steps. The proposed solution uses a semi-dataflow architecture with tiling to map the all-pairs force calculation to the FPGA in a highly parallel manner. Results show the FPGA implementation achieves speeds of 4400 million particle-pairs per second, outperforming a CPU and achieving performance competitive with state-of-the-art ASIC and GPU implementations.