NHAM: An NFV High Availability Architecture for Building Fault-Tolerant Stateful Virtual Functions and Services

Resumo

Despite the many advantages of Network Function Virtualization (NFV) technology, the dependability of virtual services must be carefully addressed so that NFV can meet the requirements of commercial carriers. In particular, it is essential to provide mechanisms to ensure their correct and continuous operation. In this work we propose NHAM: an NFV High Availability Module designed within the NFV-MANO (NFV Management and Orchestration) reference model. NHAM allows the creation and management of fault-tolerant virtual network services consisting of stateful VNFs (Virtualized Network Functions) and SFCs (Service Function Chains). The proposed architecture provides fault management, including a choice of recovery mechanisms that can be applied depending on the specific needs of each service. The solution is holistic in the sense that it does not require any modifications of the source code of VNFs/SFCs to make them fault-tolerant. The strategy is based on SFC buffer management coupled with VNF checkpoint/restore, providing high availability in a transparent way. A prototype was implemented and experimental results are presented.