A Delay Optimization Service Function Chain Rescheduling Algorithm Based on Deep Reinforcement Learning

6G is the core technology of future mobile communication networks. It uses network functions virtualization architecture as the key foundation to promote the reconstruction of the telecommunication network architecture. With the surge of service traffic and the emergence of delay-sensitive applications, such as in the field of intelligent transportation, reducing end-to-end network delay is crucial to developing new network technologies. The existing low-latency methods for network functions virtualization have difficulty constructing the shortest matching path and lack flexibility in scheduling algorithms. Therefore, this study proposes a virtual network function deployment method based on Deep Reinforcement Learning for reducing transmission delay by finding the shortest matching path and a service function chain rescheduling mechanism based on network calculus theory to reduce the queuing delay and improve the flexibility of the scheduling algorithm. The research will enhance low-latency virtual network function deployment and service function chain scheduling, providing users end-to-end service quality assurance, improving adaptability to new 6G application scenarios, and promoting the rapid development of next-generation internet technologies.