A Solution for Goal-oriented Policy Refinement in NFV Management and Orchestration Systems.

In this thesis, we propose an automated policy refinement procedure for NFV-MANO systems based on planning and description logic reasoning. It is implemented in ATOM, a framework that enables NFV-MANO to record high-level goals and performs a fully automated policy refinement that derives enforceable rules to govern NFV-MANO behavioral choices. Besides, ATOM provides a semantic verification system to find inconsistencies among policies. Experiments have shown that our solution scales well considering realistic scenarios with different sizes.