Research On Data Placement In Cooperative Cloud-edge Environments

With the advancement of society and the continuous development of technology,various applications continue to appear and massive amounts of data increase.People put forward higher requirements for user experience.It forces people to find more effective ways to explore the intrinsic value of data and optimize the placement of data reasonably.The emergence of cloud computing and edge computing provide more possibilities for meeting social needs.However,cloud computing and edge computing have a wide variety of services and resources,and their computing and storage capabilities are also very different.Therefore,in the face of increasingly complex network environments and application requirements,the collaborative development of cloud computing and edge computing has become a trend.In addition,since most data in the real world has temporal and spatial attribute information,and some data characteristics also have relevance and variability in the space-time dimension.These characteristics have a significant impact on data placement,and existing studies lack consideration in this regard.Therefore,in the cloud-edge collaboration environment,there is an urgent need to obtain a reliable and effective data placement solution.Based on the above issues,this article has carried out the following research on how to model and calculate from the temporal and spatial attributes of data,how to analyze the correlation between data regions,and how to optimize data placement in the cloud-edge collaborative environment.(1)Temperature matrix-based data placement.This chapter considers the characteristics of the data itself and proposes a data temperature matrix based on the data temperature,and proposes a replica selection algorithm to meet user latency requests.Then,the cost matrix is obtained according to the data server mapping relationship obtained by the data copy matrix and the regional server matrix.Finally,the improved Hungarian algorithm is used to obtain the data placement strategy based on the data cost matrix.The experimental results show that the method proposed in this chapter can effectively reduce the cost and load rate under the premise of satisfying the user's access latency.(2)Research on the model and calculation of the regional value of data.Based on the above data temperature calculation model,this chapter further defines the regional value of data model.Because the data temperature lacks consideration of the mutual influence between regions,it also has a great influence on the data placement.Therefore,this chapter draws on the idea of PageRank algorithm and gives two judgment rules for data temperature and regional value.That is,the higher the temperature value,the greater the value contribution,but at the same time the farther the distance,the smaller the value contribution.Finally,a comparative analysis of the strategies that consider the data temperature and the regional value of data is carried out under the simulated data set and the real data set.It can be learned that the regional value and its placement strategy considering the mutual influence between regions can bring better results.(3)Optimized placement based on the regional value of data.Based on the above regional value of data research,this chapter proposes a method of data optimal placement for cost and user access latency in a cloud-edge collaborative environment.First,this chapter gives the definition of related problems and optimization goals,and then combines regional value calculations to propose a data optimization placement method based on replica selection.At the same time,the experiment compared multiple traditional data storage methods and computational intelligence algorithms.It is proved from multiple angles that the proposed method can ensure high efficiency and obtain better results in terms of cost.This paper proposes a strategy for data placement in a cloud-edge environment,takes into account the spatio-temporal characteristics of data,and discusses and analyzes the regional value characteristics of the data.A large number of experiments have verified that the method proposed in the article can effectively improve the effect of data placement in a cloud-edge fusion environment.