• Quality of service
• network slicing
• QoS
• OpenFlow
• ONOS
• Open Network Operating System
• 5G
• Software Defined Networking
• SDN

Due to the continuously increasing number of users and the convergence of many services on the global network infrastructure, Internet Service Providers (ISP) are still facing the problem of congestion control and management on their networks. Specifically, with the advent of 5G the services to be supported may feature very different requirements (e.g., in terms of required bandwidth and latency). In this context, the concept of slicing has emerged in conjunction with the Software Defined Networking (SDN) instruments to enable the logical segmentation of a single network into multiple logical networks on a per-service basis.
This thesis designs, develops and evaluates a new SDN application for the Open Network Operating System (ONOS) capable to perform network slicing providing differentiated QoS levels. Specifically, network slices will be dynamically configured by the SDN controller and for each of them it will be possible to provide connectivity isolation, performance isolation a prioritization of the traffic. For achieving these features the SDN controller uses several SDN and networking tools like meters and queues.
After the presentation of the state of the art of the various technology leveraged in this thesis, the design of the system is described in great detail, i.e. tools/components used/developed for realize the application. Then, the results obtained in an experimental environment are presented, focusing on the evaluation of the transmission rate and latency.