LAVP: A Latency-Aware Virtual Network Function Placement Strategy for Service Function Chain in Network Function Virtualization

Abstract

Nowadays, businesses have benefited greatly from Network Function Virtualization (NFV), while users have gained new experiences. By transforming network functions into Virtual Network Functions (VNFs) running on High-Volume Servers (HVSs) instead of using dedicated hardware devices, NFV enables the reduction of Capital Expenditure (CAPEX) and Operation Expenditure (OPEX). However, NFV is now facing numerous issues, including performance, stability, and security. In this article, we examine the influence of VNF placement on the end-to-end latency. To reduce latency, the Latency-Aware VNF Placement (LAVP) algorithm, a heuristic technique, is offered to deal with the VNF placement problem by finding the best path for the Service Function Chain (SFC) while ensuring VNF ordering. The VNFs can then be resourced and deployed on the nodes along the path. Our algorithm is implemented on two different topologies and compared with other algorithms, namely First Fit, Best Fit, Worst Fit, ASR, AvReStrategy and an ant colony algorithm. The experimental results show that our proposed solution can significantly reduce delays for the service chain with less execution time compared to other strategies.