Dynamic resource allocation method in mobile edge computing environment and system thereof

Abstract

The invention relates to a dynamic resource allocation method in a mobile edge computing environment and a system thereof, which are characterized in that jobs are divided into CPU intensive jobs and I / O intensive jobs, resource nodes are divided into CPU main resources and I / O main resources, and the similarity of the tasks and the resource nodes, the priority of the tasks and the network transmission cost of the tasks and the resource nodes are acquired; according to the job resource matching model, a task number and a resource node number are obtained, if the task number is greater than the resource node number, a resource node obtained by subtracting the resource node number from the task number is added, otherwise, the job is matched with the resource node by using the weighted bipartite graph, and an optimal solution for matching the job with the resource node is obtained; according to the method, the matched computing resource nodes are allocated for the jobs, the resource occupancy rate is effectively reduced, the resource utilization rate is improved, the problem of flow blockage is effectively relieved, a large number of channel bandwidth resources are prevented from being occupied, the system performance is improved, and meanwhile the aspects of average job execution delay, job hit rate and the like are greatly improved.

Description

technical field [0001] The invention relates to a dynamic resource allocation method and system in a mobile edge computing environment. Background technique [0002] With the increasing popularity of 5G networks and the rapid development of Internet of Things technology, various forms of new applications have been promoted, such as AR, VR, autonomous driving, smart cities, telemedicine and other technical fields. The different types of computing data generated by these new application scenarios pose a huge challenge to the network center, because these application scenarios can bring higher bandwidth rates, lower system energy consumption and computing delays to mobile terminals, As a result, the computing tasks and different types of services that need to be processed by the network center grow exponentially, which brings a heavy burden to the core network. Therefore, with the increase of network traffic carried by various mobile terminals and IoT devices, the centralized ...

AI Technical Summary

Problems solved by technology

Aiming at the problem of computing resource allocation in mobile edge computing, some experts and scholars have devoted themselves to research on resource allocation in the MEC network environment, but most of the research is based on joint computing offloading, channel allocation, spectrum allocation, power allocation and other technologies for resource allocation. And the optimization goals are mostly unloading delay, system energy consumption or user income, etc., but there are few related studies on dynamic resource allocation under multi-user and multi-MEC servers

However, the traditional method will cause serious traffic congestion and occupy a large amount of channel bandwidth resources.

Images