Intelligent awareness of delay-sensitive Internet traffic in Digital Twin Network

The increasingly complex network architecture and numerous emerging network services make it difficult for traditional "best effort " consumer Internet to support current Production Internet/5G requirements for differentiated SLA commitments, low latency, and deterministic bandwidth. To solve various difficulties and challenges faced by the development of ultra-large-scale networks, the Digital Twin Network has become a research hotspot. In this paper, based on the traffic modeling problem involved in the digital twin layer of digital twin network architecture, we proposed an intelligent awareness method for delay-sensitive network traffic. Through the real-time awareness of delay-sensitive traffic, intelligent routing of network services is realized and meets the requirements of scenarios, such as high-value traffic assurance and intelligence traffic scheduling and grooming for network congestion. We experimentally confirmed the scheme by traffic data collected in realistic network environments, and it has an awareness accuracy of more than 90%.