Resource-Aware and Profit Maximizing SFC Deployment in the Cloud-Edge Collaborative Network

The cloud-edge collaborative network that integrates the advantages of abundant resources of cloud computing and low latency of edge computing, achieves more efficient resource utilization and higher quality of service and has become a research hotspot. In Network function virtualization (NFV), services are realized by end-to-end service function chains (SFCs). The combination of cloud-edge collaborative network and NFV that allows VNFs in SFCs to be flexibly deployed on the cloud and the edge, leveraging the advantages of both. However, current research on the SFC deployment in the cloud-edge collaborative network still faces challenges. Hence, this paper studies the SFC deployment problem in the cloud-edge collaborative network and proposes a resource-aware profit maximizing SFC deployment (RA-PMSD) scheme under the latency, computing and bandwidth resources constraints. A resource-aware heuristic algorithm is proposed for the RA-PMSD scheme, which separately deploys SFCs according to the latency requirements. A dynamic threshold is set based on the edge average remaining resources in the edge network, which serves as a judgement condition for a VNF to select the node type (cloud or edge). Simulation results show that the proposed algorithm achieves higher total profit and acceptance rate and lower average latency than other algorithms.