Collaborative storage scheduling method based on alternating direction multiplier method

Abstract

The invention discloses a collaborative storage scheduling method based on an alternating direction multiplier method, which comprises the steps of acquiring relevant information of nodes in the mobile edge cloud, and building an optimization model corresponding to the total utility of the mobile edge cloud; building a corresponding constraint formula based on a principle of node transmission; solving the optimization model according to a preset algorithm based on the alternating direction multiplier method to obtain an optimized storage scheduling scheme corresponding to the optimization model; and performing a storage scheduling task on the corresponding nodes in the mobile edge cloud by adopting the optimized storage scheduling scheme. According to the collaborative storage scheduling method based on the alternating direction multiplier method, a storage scheduling scheme is acquired after an iterative operation through considering and integrating heterogeneous information of the nodes in the mobile edge cloud, and the performance of cloud computation is improved; and the time delay of task execution and the total cost of system running can be reduced through building the optimization model of the total utility, local information of the nodes is utilized as far as possible, and the overall reliability of the system is ensured.

Description

technical field [0001] The invention relates to the technical field of data storage scheduling, in particular to a collaborative storage scheduling method based on an alternating direction multiplier method. Background technique [0002] Recently, mobile edge cloud has received increasing attention because it can provide better cloud services for emerging technologies such as big data, Internet of Things, and smart home. Mobile Edge Cloud (MEC, Mobile Edge Cloud) can be regarded as an open cloud platform that integrates network, computing, storage and core application capabilities. This new form of cloud computing is proposed to effectively expand the performance of cloud computing in the edge network environment, so as to solve one of the most important challenges in cloud computing services: how to deal with the increasing demand of mobile edge applications and the limited local resources of mobile edge devices. contradiction. In this form of mobile edge cloud, all nodes...

AI Technical Summary

Problems solved by technology

[0003] In fact, data storage operations in this environment still face many challenges

First, most cloud computing services today still use centralized algorithms, which is difficult to meet the most basic requirements of mobile edge cloud

Second, most devices or applications in the mobile edge cloud are highly latency-sensitive

At the same time, considering the large amount of stored data in the remote data center and the limited bandwidth between nodes, although users need timely services, the delay caused by long-distance network communication is inevitable, which will definitely affect the user experience.

Third, the existing non-centralized storage algorithms are not fully considered and do not adapt to the new features reflected in the development of mobile edge cloud

Therefore, in the process of implementing this application, the inventor found that the reliability, stability and energy utilization rate of mobile edge cloud storage scheduling in the prior art are not ideal

Images