Self-reconfigurable software-defined networks based on automated planning

Abstract

Background. 5G networks come with the promise to serve a wide range of use cases, each of which features a set of diverse and stringent requirements. By leveraging the potential of Software Defined Networks and Networks Functions Virtualization, a common infrastructure is sliced into multiple logical networks according to the diversified service demands, enabling the implementation of 5G network slicing. As any emerging networking technology, 5G network slicing imposes new challenges to the already complex network management. For instance, reconfiguration operations must be split between the Infrastructure Provider (InP) and the tenants leasing network slices. However, since tenants are not networking professionals, a bespoke reconfiguration system is required allowing them to timely reconfigure network slices meeting end-user demands, with minimal dependance on InP. Aims. This thesis introduces an approach based on Automated Planning for achieving self-configuration in SDN. The main objective of the thesis is divided into three parts: (i) Design an approach based on automated planning for providing self-configuration in SDN, (ii) Present a reference implementation of the proposed approach, and (iii) Evaluate the proposed approach regarding time-response, time-planning, and network overhead.