Dependency Aware and Delay Guaranteed SFC Placement in Multi-domain Networks
Yanming Liu, Chuangchuang Zhang, Yihao Li, Shuning Zhang, Hongyong Yang · 2023
As a new paradigm, Network Function Virtualization (NFV) enables flexible and customized service provisioning. NFV decouples network functions from dedicated hardware devices, and a set of Virtual Network Functions (VNFs) are chained in a predefined order as a Service Function Chain (SFC) to build customized network services. Considering the increase in network scale and dependency relationship between different VNFs, how to efficiently place the SFCs in multi-domain networks to meet Quality of Service (QoS) requirements is a serious challenge for NFV. In this paper, we focus on the SFC placement optimization problem in multi-domain networks taking in account VNF dependency relationships and QoS requirements. First, the SFC placement optimization problem in multi-domain networks is formulated as a multi-objective optimization model with the aim to maximize service acceptance ratio and minimize operational cost while guaranteeing delay requirement of SFC requests. Then we propose a Dependency Aware and delay Guaranteed heuristic SFC Placement algorithm, called DAGP, to automatically place the VNFs in multi-domain networks in an adaptive and cost efficient manner. Finally, evaluation results demonstrate that compared with comparison mechanisms, our proposed SFC placement algorithm can obtain higher service acceptance ratio and smaller operational cost.