Network Slice Resource Estimation and Embedding in 5G Mobile Networks

Network Slicing is a key-feature of the fifth generation of mobile networks (5G), to overcome the challenge of increasingly diverging network requirements emerging from new use-cases like the IoT, autonomous driving and Industry 4.0. Network Slices are logical end-to-end networks sharing a common mobile network infrastructure. Due to the dynamic nature of Network Slices, there is a strong demand for fast Network Slice resource estimation and embedding. However, end-to-end mobile Network Slice embedding comes with special requirements, which makes it difficult to apply known Virtual Network Embedding (VNE) algorithms. For instance, low latency Network Slices may require deployment of their services and application close to the user, at the edge of the mobile network. Since the users of those services might be widely distributed and mobile, multiple instance of the same application will be required to be available on numerous edge clouds. This paper tackles the problem of Network Slice embedding with multiple application instantiation. A nearly optimal as well as an efficient heuristic method for solving the Network Slice embedding problem with multiple application instantiation are provided. The proposed Integer Linear Program (ILP) based model and algorithm provides a nearly optimal solution for the number of application instances as well as the best locations for deployment in the mobile network. In addition, a scalable heuristic for the optimal deployment of Network Slices in this scenario is provided. Evaluation of both algorithms proves their suitability for Network Slice embedding with multiple application instantiation and confirms the accuracy of the proposed heuristic.