Improving GPU Simulations of Spiking Neural P Systems

Abstract

In this work we present further extensions and improvements\n\t\t\t\t of a Spiking Neural P system (for short, SNP systems) simulator on graphics\n\t\t\t\t processing units (for short, GPUs). Using previous results on representing SNP\n\t\t\t\t system computations using linear algebra, we analyze and implement a compu-\n\t\t\t\t tation simulation algorithm on the GPU. A two-level parallelism is introduced\n\t\t\t\t for the computation simulations. We also present a set of benchmark SNP sys-\n\t\t\t\t tems to stress test the simulation and show the increased performance obtained\n\t\t\t\t using GPUs over conventional CPUs. For a 16 neuron benchmark SNP system\n\t\t\t\t with 65536 nondeterministic rule selection choices, we report a 2.31 speedup of\n\t\t\t\t the GPU-based simulations over CPU-based simulations.