Optimal Placement of Cloudlets Considering Electric Power Communication Network and Renewable Energy Resource

To overcome the likely unbearable service delay in the centralized cloud computing diagram, edge computing offers a way to shorten it by placing the small-scale cloud infrastructures, such as cloudlets in close proximity to end users. A well-designed cloudlet placement scheme can provide the proximal and economic service for the dynamic traffic loads and applications, which ask for stringent quality of service (QoS) requirements, particularly service delay. However, it is a nontrivial task to efficiently deploy cloudlets among geo-distributed communication network nodes. To tackle this cloudlet placement problem (CPP), an entropy-weight-based placement algorithm (EWPA) that considers both the electric power communication network (CN) characteristics and geo-distributed renewable energy resources (RE) is proposed. Case studies compare the proposed algorithm with an enumeration-based placement algorithm (EPA), random placement algorithm (RPA), as well as other benchmarks. Results show that the proposed algorithm can obtain a near optimal cloudlet placement scheme and outperforms these benchmarks in average service delay and total cost.