User business queuing optimization method based on edge computing

The invention relates to a user business queuing optimization method based on edge computing. The method comprises the following steps: firstly, classifying different kinds of business generated by auser according to a user identification number and quality of service requirements by using a packet classifier, and then placing the classified business in corresponding sub-cache management modules;secondly, establishing a batch arrival model, performing probability estimation on business arrival by using the batch arrival model; thirdly, obtaining service parameters of an edge computing node according to the batch arrival model, and then selecting an edge node according to the service parameters of the edge computing node; and finally setting the time threshold as T, and controlling the opening and closing of the edge computing node according to the time threshold T so as to save energy. Due to the large number of edge devices in the 5G environment and the large number of eruptive business volumes, according to