内嵌式数据处理单元的服务功能链编排机制OA

In response to the demand for high flexibility and high-performance network services for large-scale and diver-sified business applications in cloud-edge-end scenarios,a service function chain orchestration mechanism on embedded data processing units was proposed.Firstly,an architecture of service function chain on embedded data processing units was proposed to address the issue of traditional service function chains being unable to balance high flexibility and high-speed processing.Secondly,in response to the heterogeneous resources requirement of service functions under this archi-tecture,a service function chain deployment problem was formulated by comprehensively incorporating resource occu-pancy,hardware accelerator utilization,end-to-end latency,and deployment response time.To solve this problem,a ser-vice function chain deployment method based on the genetic particle swarm optimization algorithm was developed,and the process of dividing sub chains and merging similar requests was designed to improve the speed of the proposed method.Simulation results show that,compared with other methods,the proposed method provides better service func-tion chain deployment decisions and has a faster decision-making speed.