Edge Computing Task Scheduling Based on Improved African Vulture Optimization Algorithm

With the growth of the Internet of Things (IoT), how edge computing is currently appropriate to handle computationally intensive tasks to ensure the quality of service requested by users is one of the key issues faced today. For the task requires CPU, memory, bandwidth and disk multiple types of resources on the computing node load factor and cost. In this paper, we propose an edge computing scheduling strategy based on Chaotic Selective Nonlinear Weighted African Vulture Optimization Algorithm (CSNW-AVOA). The algorithm firstly uses Tent chaotic mapping to initialize the population, secondly proposes selective nonlinear weights based on the algorithm properties, and finally uses Lévy flights in the development phase. The results of simulation experiments show that the performance of CSNW-AVOA is significantly improved compared to four mainstream scheduling algorithms, namely, AVOA, OAWOA, IGWO, and PSO, and its load balancing rate and cost consumption are both improved and reduced compared to AVOA. The algorithm effectively utilizes the resources of edge computing nodes and improves the resource utilization.