Virtual Network Mapping - An Optimization Problem

Network Virtualization is acclaimed to be a key component for the Future Internet by enabling the coexistence of heterogeneous (virtual) networks on the same physical infrastructure, providing the dynamic creation and support of different networks with different paradigms and mechanisms in the same physical network. A major challenge in the dynamic provision of virtual networks resides on the efficient embedding of virtual resources into physical ones. Since this problem is known to be NP-hard, previous research focused on designing heuristicbased algorithms; most of them do not consider a simultaneous optimization of the node and the link mapping, leading to non optimal solutions. This paper proposes an integer linear programming formulation to solve the Virtual Network embedding problem, as a simultaneous optimization of virtual nodes and links placement, providing the optimal boundary for each virtual network mapping. A link−node formulation is used and the multi-commodity flow constrain is applied. In addition, a heuristic algorithm for virtual network embedding is also proposed and compared against the optimal formulation. The performance of the Integer Linear Programming formulation and of the heuristic are evaluated by means of simulation. Simulation experiments show significant improvements of the virtual network acceptance ratio, in average additional 10% of the virtual network requests are accepted when using the proposed formulation, which corresponds, in average, to more 7 virtual networks accommodated on the physical network.