The Optimal Deployment of Edge Servers and Cloud Data Center
Liping Yang
School of Mathematics and Finance, Hunan University of Humanities, Science and Technology, Loudi Hunan, 417000, P.R. China.
Xudong Wu
School of Information, Hunan University of Humanities, Science and Technology, Loudi Hunan, 417000, P.R. China.
Ningxin Liu
School of Mathematics and Finance, Hunan University of Humanities, Science and Technology, Loudi Hunan, 417000, P.R. China.
Chengzhi Liu *
School of Mathematics and Finance, Hunan University of Humanities, Science and Technology, Loudi Hunan, 417000, P.R. China.
*Author to whom correspondence should be addressed.
Abstract
In edge computing, the signal of the 5th generation (5G) base station is often required to be covered by edge servers to guarantee a high transmission rate. And the smaller the total distance between the 5G base stations and the edge servers is, the faster the transmission rate is. Therefore, we need to study the optimal deployment of edge servers and the cloud data center. Firstly, the cluster analysis is used to cluster the 5G base stations and obtain the cluster centers, then the edge servers are deployed in the cluster centers. Secondly, the centroid method is used to determine the location of the cloud data center. By adopting the elbow and gap statistic methods, we also study the optimal number of edge servers to be deployed. Computer simulation results show that our methods provide better deployment locations for the edge servers and the cloud data center.
Keywords: Edge computing, deployment, edge server, cloud data center, centroid method, cluster analysis