Optimizing Deployment Period of Service Replication in 5G Edge Network

Leveraging cloud infrastructure to network edge facilitates the development of the delay-sensitive applications which requires minimal response latency to ensure desired Quality of Experience (QoE). To ensure desired QoE, fast relocation of service instances is required supporting user mobility. Existing works focus on proactive migration of service instances to reduce overall deployment time. However, provisioning service replicas over multiple edge nodes results in increased resource consumption as well as overall deployment cost. In this paper, we have developed a deployment policy of service replicas which optimizes the deployment period to minimize the QoE degradation and deployment cost. In particular, we have formulated two Mixed Integer Linear Programming (MILP) problems which minimize the QoE degradation due to service migration and the deployment cost of replicas, incorporating the mobility pattern of the user. Performance evaluation results show the effectiveness of The proposed system compared to state-of-the-art works.