Delay-aware and resource-efficient service function chain mapping in inter-datacenter elastic optical networks

作者：Min Zhu, Jiahua Gu, Tianyu Shen, Jianyu Zhang, Pingping Gu · 发表于：Journal of Optical Communications and Networking · 年份：2022 · DOI：10.1364/jocn.461263 · 被引用次数：12 · 研究领域：Software-Defined Networks and 5G、Advanced Optical Network Technologies、Advanced Photonic Communication Systems

Network function virtualization (NFV) is emerging as a promising paradigm for network architectures. By migrating network functions from dedicated hardware appliances to software instances running in a virtualized environment, NFV promises to offer a more flexible way to deploy and manage service function chains (SFCs). When deploying these SFCs to users, the network operators require not only the user’s demands (e.g., end-to-end delay) to be satisfied, but require the cost of SFC mapping to be minimized (e.g., resource consumption). To fulfill these two goals, in this paper, we have investigated how to realize the delay-aware and resource-efficient SFC mapping in inter-datacenter elastic optical networks. We first formulate an integer linear programming (ILP) model to solve the problem exactly. The main optimization goal in the ILP model is to jointly minimize resource consumption and end-to-end delay to achieve optimal virtual network function placement. Then, a delay-aware and load-balancing mapping algorithm (DALB-MA) is proposed to obtain a near-optimal solution in a reasonable amount of time. Finally, we evaluate the proposed ILP model and heuristic algorithms via extensive simulations. The results indicate that the proposed ILP model and the DALB-MA outperform the benchmarks in terms of block rate, average cost, number of CPUs used, maximum frequency slot index, and delay margin gain.