Global Energy Optimization Strategy Based on Delay Constraints in Edge Computing Environment

Abstract

Edge Computing is one of the core technology of 5G networks. Edge computing deploys servers at the edge of the wireless access network, sinking cloud computing capabilities to the edge of the network, sharing the computing pressure of mobile users nearby, and improving the computing power of the entire network. Energy consumption is one of the important research issues of edge computing. At present, research on edge computing focuses on the energy consumption of terminal device, while little attention is paid to the energy consumption of edge servers. In this paper, considering the above two kinds of energy consumption, a global energy optimization strategy based on delay constraint in edge computing environment is proposed. Specifically, first, we use queuing theory to analyze the average delay of each terminal device and edge cloud processing computing tasks in the Internet of Things network, and the average delay of the entire system processing computing tasks. Secondly, we use the average delay as a constraint to establish a mathematical model for minimizing the total energy consumption of the device and the server. Then, we design a genetic algorithm-based offloading computation optimization algorithm to solve the above problems, so as to obtain the number of running servers in the edge cloud and the offload probability of IoT devices. Finally, the goal of minimizing the energy consumption of the overall system under the time delay constraint is achieved. The simulation experiment verifies the effectiveness of the energy optimization strategy.