مکانیابی زنجیره های تابع سرویس در رایانش مه با استفاده از بهینه سازی الگوریتم رقابت استعماری

Network Function Virtualization (NFV) is a new technology for consolidating network functions on high-volume storage, servers, and switches located anywhere in the network. Virtual Network Functions (VNF) are connected together to provide Service Function Chain (SFC) network service. According to the quality of service requirements and network characteristics, SFCs are provided by performing the two tasks of locating VNFs and embedding links in underlying networks. Reducing the cost of deployment is a desirable goal for all service providers in cloud-edge environments to increase their profits. However, increasing resource utilization in order to reduce deployment cost may lead to increased service delay, resulting in increased service level agreement violations and decreased user satisfaction. Dynamic location of SFCs by reusing VNF instances can lead to improved resource utilization and time saving. In this paper, dynamic SFC positioning algorithm with VNF reuse in cloud-fog ecosystem is proposed using Imperialist Competitive algorithm. The proposed algorithm is a planning model for the location of SFCs based on the reuse of initialized VNFs, which can create a compromise between service costs and service quality by considering the limitations. Considering the cost criterion, the proposed algorithm has a ۳۱% superiority over the SPFF algorithm and ۹% over the SSO method.

کلیدواژه ها:

Service function chain ، localization ، network function virtualization ، fog computing ، Imperialist Competitive Algorithm.