SLA-oriented cloud edge collaborative resource arrangement and request scheduling intelligent optimization method

Abstract

The invention discloses an SLA-oriented cloud edge collaborative resource arrangement and request scheduling intelligent optimization method. The method comprises the following steps of S1, initializing training parameters of a neural network and edge node states in an edge cluster, S2, acquiring related data of the service request under the current time scale and the state of a resource unit in a resource channel, S3, acquiring resources pre-allocated to the resource unit by each edge node, S4, allocating corresponding resources to the resource units according to the pre-allocated resources, S5, solving each resource channel in parallel based on the problem of maximizing the overall throughput of the system to obtain a service arrangement set, S6, performing service arrangement according to the service orchestration set, and taking the throughput rate under the current edge cloud system as an award, S7, updating the neural network, and S8, performing iteration in sequence according to the method until training convergence. An intelligent strategy is provided for service arrangement and request assignment of the side cloud system, and the SLA of various user services is effectively ensured.

Description

Technical field [0001] The present invention belongs to the field of edge computing technology, and in particular, it is directed to a cloud-oriented cloud synergistic resource arranging and request scheduling intelligence optimization method. Background technique [0002] With the arrival of all things and the continuous development of the wireless network, the number of devices from the network edge and the data generated are rapidly increased. The centralized processing mode of the cloud computing model as the core will not be able to efficiently handle the data generated by the edge device, and we introduce edge computing technology based on cloud primary technology. Edge computing technology aims to utilize cloud computing capacity, without having a large communication delay in accessing the cloud, but to achieve all the potential of edge calculation, still need to assign limited edge cloud resources to competition requests to competition requests intelligent strategy . [0...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the unreasonable technical problem of complex edge clusters allocating limited resources among competing requests, the present invention proposes an SLA-oriented intelligent optimization method for cloud-edge collaborative resource arrangement and request scheduling, by using cloud-native, edge computing, and artificial intelligence technologies , the combination of deep reinforcement learning and traditional algorithms solves the problem of reasonably and efficiently allocating limited resources between competing requests in complex edge clusters, and can effectively guarantee the SLA of different services

Images