A type of energy saving strategy with adaptive link transmission rates and a sleep mode for edge computing networks

Abstract

As the communication industry develops, issues such as how to conserve energy, reduce environmental pollution, and realize more sustainable communication are becoming more pressing. In this paper, a type of energy saving strategy for edge computing networks is proposed. In this energy saving strategy, links will transfer to the sleep mode from the active mode as soon as there are no packets ready for transmission, i.e., the link will be idle once the output buffer is empty. In the active mode, the link transmission rate is adaptively adjusted based on the number of packets in the output buffer with two thresholds. In order to capture the digital nature of modern communication and numerically evaluate the system performance, we establish a discrete-time multiple-vacation queue with variable service rates to model the proposed energy saving strategy. By using the method of a matrix-geometric solution, we analyze the queueing model in the steady state, and obtain the formulas for the performance measures in terms of the energy saving rate of the system, the average latency of packets and the utilization of the edge server. Moreover, we provide numerical results with analysis and simulation to investigate the influence of the system parameters on the system performance. Finally, we construct a cost function and develop an intelligent searching algorithm to give the jointly optimal values for the two buffer thresholds and the lower transmission rate of packets.