Design and Implementation of SFC Monitoring and Alerting Based on YANG Data Model

Service Function Chaining (SFC) is an emerging technology that provides flexible service management and traffic control through an ordered set of service functions (SF). However, the variable deployment of service functions makes network architecture more complex and introduces uncertainty in the distribution of available resources. Additionally, the likelihood of service interruption rises with network configuration complexity. So, it is essential to monitor the SFC and alert the network operator when SFC problems are encountered. The SFC monitoring and alerting method based on the YANG data model is proposed. Each SF /SFF on the service function chaining is monitored, mainly for CPU utilization, memory utilization, bandwidth utilization, latency, average load and other SF metrics, and throughput and latency of SFF. Thresholds are set for each indicator. When a component exceeds the preset threshold, an abnormal information alarm will be triggered, and then the monitoring and alerting module will be modeled using the YANG data modeling language. Finally, the monitoring and alerting module in OpenDaylight (ODL) is implemented, and the experimental verification results are presented.