MEC-oriented distributed service scheduling method

Abstract

The invention discloses an MEC-oriented distributed service scheduling method, which comprises the following steps of: for any service in a service set, determining M MEC servers closest to the service according to a shortest route algorithm, adding the M MEC servers into an available MEC server set; for any service in the available MEC server set, determining, by using a polling mode, whether the jth computing resource on each server is idle or not at an i moment; if the jth computing resource is idle, allocating the computing resource to the service; and in the same way, performing the same process on the j+1th computing resource to the Jth computing resource, obtaining the resource allocated to the service on each MEC server, namely the size of task segmentation, determining the starting time and the ending time of the service, and calculating the completion time of all the services. According to the method, the service segmentation problem and the MEC resource allocation problem are fused together for joint optimization, the computing resources of the servers are fully utilized, the load between the servers is balanced, and the queuing waiting time of the terminal is reduced.

Description

technical field [0001] The invention relates to a service scheduling method, in particular to a distributed service scheduling method of MEC, and belongs to the technical field of network resource allocation. Background technique [0002] With the rapid development of the mobile Internet, the number of smart mobile devices is increasing rapidly, and the proportion of cloud data centers is increasing. By 2021, 94% of all workloads will be processed in cloud data centers, and the business volume carried by data centers in 2021 has doubled compared with 2016. This means that the amount of data to be carried by the mobile communication network will be dozens, hundreds, or even thousands of times that of the current one. If a single cloud computing model is still adopted, the speed and delay cannot be guaranteed at all. In this context, experts put forward the concept of MEC (Mobile Edge Computing, Mobile Edge Computing), which is a network architecture that provides services re...

AI Technical Summary

Problems solved by technology

However, there is still very little research on distributed MEC services, especially the joint optimization research on business segmentation and MEC resource allocation is still blank

Existing technologies either only consider distributed business segmentation, or only consider MEC resource allocation, and do not jointly optimize the two problems

Images