Energy-efficient fuzzy control model for GPU-accelerated packetclassification

摘要

As a core component of many network infrastructures, packet classification requires matching packetheaders against a series of predefined rules. Its performance determines, to some extent, how fast packetscan be processed. There already exists many proposals, which optimize the throughput of packetclassification, but few of them take power consumption into account. To meet the requirements of greennetwork computing, this paper focuses on energy-efficient solutions that provide reasonable throughput aswell. Similar to recent advancements, the graphics processing unit (GPU) is adopted to accelerate rulematching. Then, inspired by the frequency-variable energy-consuming model for air conditioners, a fuzzy control–based energy efficiency optimizing model is proposed for GPU-accelerated packet classification.As demonstrated in the evaluation experiments, when the GPU is in the idle status, the proposed modelcan save 10 W. In running status, the fuzzy control–based energy efficiency optimizing model can avoidGPU shutdown issue caused by GPU self-protection mechanism when the GPU temperature rises to 95°C.Furthermore, by improving the resource configuration of GPU kernels according to the model, the overallenergy efficiency is enhanced by up to 15.5%, while simultaneously keeping throughput at the same level.