2016/11/10 by Abhishek Gupta, M. Farhan Habib, Uttam Mandal +3
Computer Science · Engineering · #Advanced Optical Network Technologies #Caching and Content Delivery #Chaining #Cloud computing #Computer network #Computer science #Distributed computing #Network service #Network topology #Networking hardware #Operating system #Provisioning #Service (business) #Software-Defined Networks and 5G #Virtual network #cs.NI
paper · pdf · doi:10.1016/j.comnet.2018.01.028
published as https://doi.org/10.1016/j.comnet.2018.01.028
arxiv created 2016/11/10 · openalex publication_date 2018/01/23 · arxiv updated 2018/04/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
Network functions (e.g., firewalls, load balancers, etc.) have been traditionally provided through proprietary hardware appliances. Often, hardware appliances need to be hardwired back to back to form a service chain providing chained network functions. Hardware appliances cannot be provisioned on demand since they are statically embedded in the network topology, making creation, insertion, modification, upgrade, and removal of service chains complex, and also slowing down service innovation. Hence, network operators are starting to deploy Virtual Network Functions (VNFs), which are virtualized over commodity hardware. VNFs can be deployed in Data Centers (DCs) or in Network Function Virtualization (NFV) capable network elements (nodes) such as routers and switches. NFV capable nodes and DCs together form a Network enabled Cloud (NeC) that helps to facilitate the dynamic service chaining required to support evolving network traffic and its service demands. In this study, we focus on the VNF service chain placement and traffic routing problem, and build a model for placing a VNF service chain while minimizing network resource consumption. Our results indicate that a NeC having a DC and NFV capable nodes can significantly reduce network-resource consumption.