A Programmable Computing Power Network Forwarding and Scheduling Method Based on SRv6

With the rapid development of science and technology, the demand for computing service resources in various industries is increasing. At the same time, catalyzed by new technologies such as 5G, AI, cloud computing, and big data, richer and more complex information data is produced. In order to cope with the changes in the current development of informatization, computing power networks emerged as the times require. Compared with traditional networks, computing power networks can more accurately select appropriate computing power service resources for users, and can provide the best routing path based on the status of the network in real time. For now, computing power networks still face three major issues: first, how to select the best computing resources for users, second, how to plan routing paths to meet user needs, and finally, how to achieve efficient and accurate forwarding. In response to the challenges faced, this paper proposes a programmable computing power network forwarding and scheduling method based on SRv6 technology. It first selects the best computing power service resources for users by using the user satisfaction function S. Secondly, the OSPF routing algorithm and network forwarding cost indicators are used to calculate the best routing and forwarding path for users. Finally, based on the selected best path, SRv6 segment routing technology is used to achieve accurate, efficient, and low-latency forwarding.