Fault and performance management in multi-cloud virtual network services using AI: A tutorial and a case study

Carriers find Network Function Virtualization (NFV) and multi-cloud computing a potent combination for deploying their network services. The resulting virtual network services (VNS) offer great flexibility and cost advantages to them. However, vesting such services with a level of performance and availability akin to traditional networks has proved to be a difficult problem for academics and practitioners alike. There are a number of reasons for this complexity. The challenging nature of management of fault and performance issues of NFV and multi-cloud based VNSs is an important reason. Rule-based techniques that are used in the traditional physical networks do not work well in the virtual environments. Fortunately, machine and deep learning techniques of Artificial Intelligence (AI) are proving to be effective in this scenario. The main objective of this tutorial is to understand how AI-based techniques can help in fault detection and localization to take such services closer to the performance and availability of the traditional networks. A case study, based on our work in this area, has been included for a better understanding of the concepts.