SimV: A Self-extension Similarity Viterbi Algorithm for Service Function Chain Deployment for Multi-access Edge Computing

Advancements in Multi-access Edge Computing (MEC) and Network Function Virtualization (NFV) offer a lot of potential for providing flexible and low-latency network services at the edge. In NFV, traditional middleboxes are transformed into Virtual Network Functions (VNFs), which are then combined into Service Function Chains (SFCs) to deliver complex network services. However, the features of MEC hosts and VNFs, along with the heterogeneity of their resources, make it challenging to develop low-latency deployment solutions. In this paper, we formulate the problem of Service Function Chain Deployments in Edge networks to minimize Latency under Heterogeneous conditions (SDELH). To address this problem, we designed a self-extension Similarity Viterbi algorithm (SimV), which contains three sub-algorithms. Firstly, a basic Viterbi algorithm is employed to find the delay-optimized SFC deployment path. Secondly, a similarity algorithm is designed to assess suitable hosts for each VNF to streamline the matching process and reduce the time complexity of the Viterbi algorithm. Lastly, a self-extension method is devised to broaden the solution space for enhancing the SFC acceptance rate. Extensive simulations are conducted. Based on the comparison simulation results, SimV significantly reduces decision time for deployment, and outperforms the compared algorithms in terms of latency, number of accepted requests, and scalability, with average improvements of 20.37%, 6.04%, and 14.49%, respectively.