Mobile networks have been traditionally composed of three segments: The Radio Access Network (RAN), the Core Network, and the Transport Network interconnecting the RAN with the Core. Developing 5G network architectures nowadays challenges the traditional split of RAN, Transport and Core by adding computing capabilities at the mobile network edge (RAN/Transport) and thus, blurring their boundaries. Such bolstering of the edge capabilities is motivated by the need in 5G to support a wide range of vertical industries, such as automotive, e-Health and media & entertainment, with very diverse and stringent requirements. In fact, from a network architecture perspective, 5G is being defined as the communications and IT substrate on top of which all sorts of diverse services will be offered, co-existing in an isolated and independent manner. The network will have to support simultaneously the requirements of vertical industries (e.g., automotive, media and entertainment, e-Health, industry 4.0) in addition to more conventional M(V)NO deployments. At the infrastructure level, two trends rule its design and deployment. First, the above context entails an increasing technical and cost pressure leading to infrastructure sharing, at least in some parts of the network. And second, the variety of scenarios in general will entail a heterogeneity of technologies not necessarily belonging to the same administrative domain, reinforcing the need of defining appropriate trading mechanisms among providers.
Furthermore, Multi-access Edge Computing (MEC) shifts the required mobile network functionality closer to the mobile edge. Deploying these edge capabilities is quite compelling for mobile operators as they can now handle a massive amount of local and raw traffic directly at the edge, thus keeping it away from their Transport and Core; hence saving capacity and cost. These edge capabilities also bring an opportunity to leverage on a rich set of context information available at the edge. In this way, context-awareness can be used at large to optimise end-to-end network performance as well as to offer new customised services to other providers including vertical industries. With the computing infrastructure now spanning from the distributed edge to distant data centres and central clouds, it becomes clear that Mobile Transport Networks shall transform from today’s rigid interconnection solutions into an SDN/NFV/MEC- based 5G Mobile Transport and Computing Platform (MTP) able of to meet the specific needs of vertical industries.
Therefore, the need for sharing the infrastructure, yet guaranteeing the requirements to the various tenants naturally leads to network slicing. But building such end-to-end slices in a multi-party multi-domain multi-technology framework poses several challenges that must be appropriately handled at the architectural and deployment points of view. These challenges are currently being addressed in several research projects (e.g., H2020 5G-Transformer, 5G-Exchange, 5G-Crosshaul) and standardization bodies. This workshop expects to become a meeting point between industry and academia to discuss about the present and future of network slicing control and management in 5G networks.