Demand-Aware Network Function Placement

Network function virtualization is an emerging network resource utilization approach which decouples network functions from proprietary hardware and enables adaptive services to end-user requests. To accommodate the network function requests, network function instances are created and deployed at runtime. In this paper, we study a network virtualization scheme to orchestrate and manage networking and network function services. We propose an integrated design for network function instance allocation and end-to-end demand realization sharing the same physical substrate network and demonstrate that the corresponding network design problem is NP complete. A mixed-integer programming formulation is proposed first to find its optimal solution, followed by a two-player pure-strategy game model which captures the competition on physical resources between network function instance allocation and routing. We then design an algorithm based on iterative weakly dominated elimination in Game Theory. Computational results demonstrate the value of the integrated approach and its ability to allocate network function instances supporting end-to-end requests with limited physical resources in optical networks.