Improving Multiple Precision Integer Multiplication on GPUs
Makoto Miyazaki, Susumu Matsumae · 2017
In this paper, we propose an efficient GPU algorithm for multiple precision integer multiplication (MPIM for short). We improve the existing MPIM algorithm that uses a data-structure called product digit table. Our proposed algorithm reduces those redundant computations and memory accesses executed on the original algorithm. We evaluate the performance of our algorithm by computer simulation.