The 2021 China Mobile Global Partner Conference with the theme of “Numbers Are Everything, Smart Computing the Future” kicked off in Guangzhou. At the main forum this morning, Wu Hequan, academician of the Chinese Academy of Engineering, delivered a speech on the theme of “IPv6 Helps Build a Network Centered on Computing Power”.
In his speech, he pointed out: “The new era of IPv6 coincides with the parallel development of new generation information technologies such as the Internet of Things, big data, cloud computing, blockchain, artificial intelligence, and 5G. IPv6 has become a new generation of IT bearer platform and unified cloud. The bearer at the edge of the network will play an important role in cloud-network convergence and multi-cloud collaboration.”
Wu Hequan emphasized that at the same time, computing power networks pose many challenges to IPv6. IPv6 requires continuous innovation in deterministic wide area networks, variable-length IP addresses, and network security. IPv6 will develop in parallel with computing power networks.
Cloud business development requires a computing power network
According to data released by the Ministry of Industry and Information Technology, in 2020, telecom operators’ cloud computing business revenue will increase by 85.8% over the previous year. By the first three quarters of 2021, revenue from cloud computing services of telecom operators has increased by 94.8% during the same period, an amazing growth rate.
Wu Hequan said that cloud services not only require communication connections, but also high-reliability, low-cost, and flexible use case resources. In order to adapt to the development of cloud services, it is necessary to build a computing power network.
In order to realize the perception, interconnection and coordinated scheduling of ubiquitous computing and services, the computing power network architecture system can be logically divided into computing power service layer, computing power platform layer, computing power resource layer, computing power routing layer and network. Five functional modules of the resource layer. Based on the ubiquitous computing power resources of the network, the computing power platform layer completes the abstraction, modeling, control and management of computing power resources, and informs the computing power routing layer through the computing power notification module, and the computing power routing layer comprehensively considers users According to demand, network resource status and computing resource status, the service application is scheduled to the appropriate node to achieve optimal resource utilization and ensure the ultimate user experience.
According to reports, the computing power network contains many technologies, including delay-sensitive network technology, deterministic network technology, network function virtualization, as well as computing-first network, telecommunications trusted blockchain, IPv6 and IPv6-based segmentation Route SRv6 and so on.
IPv6 unified cloud network edge bearer
In Wu Hequan’s view, IPv6 and IPv6-based segmented routing SRv6 technology play an important role. APN6 is application-aware based on IPv6. The traditional IP header only contains basic information for routing, such as source and destination addresses. The network cannot identify the type of service and service level (SLA) requirements carried by the IP packet. Network planning usually adopts light-load mode. (30-40%), low resource utilization and application-insensitive network channels make it difficult for operators to provide differentiated services and refined management.
The further development of the IPv6 frame extension header contains more application information, which is transmitted through the device forwarding plane and can be processed by nodes along the way to reduce the interaction impact between controllers. In the future, the IPv6 extended header can be used to embed the carried data attributes to support the management of cross-border data circulation.
From the perspective of IPv6 flow monitoring, the traditional monitoring method adopts an indirect test method of sending simulated monitoring messages, which cannot guarantee that the simulated messages are consistent with the real service path, and the accuracy of service packet loss monitoring can only reach 10-3. The flow monitoring technology does not require external probes, and OAM is carried in user messages. The processing node collects data and cares according to the OAM instruction information in the message.
SRv6 supports cross-network intelligent provisioning and operation and maintenance
And SRv6 can provide network slicing to support specific VPN services. SRv6 uses the IPv6 extension header SRH to push into the explicit route, and completes the hop-by-hop transmission by continuously updating the destination address of the node in the path. SRv6 Policy has a Color attribute. The controller dyes different customers and business flows to mark the service level requirements of different applications. The head node matches the coloring of the service flow according to the service strategy issued by the controller, completes the network path planning in compliance with the service SLA, and supports deterministic services.
At present, the SRv6 header overhead is too large. The main reason is that the backhaul traffic of the base station needs to traverse the metropolitan area network and the backbone network. The end-to-end SR tunnel can reach 8 hops or more, and the SID corresponding to each hop is 128 bits in length. The 8-layer SID generates an SRH with a length of 128 Bytes. The average payload is only 256 Bytes, which is low in efficiency and large in delay. The length of the 8-layer SRv6 SID is equivalent to the depth of the 32-layer MPLS label, which exceeds the processing capacity of the network core. IPv6 in the SRv6 domain is usually assigned the same Prefix, and the carrier networkrouterThe number generally does not exceed 1000, indicating that the number of nodes and adjacent labels is limited, and the SRv6 header has the possibility of compression.
It is reported that Chinese companies have led the IETF to propose the G-SRv6 program. Using innovative technologies such as compressed redundant prefix and two-dimensional pointer positioning, the SRH list can include both encoded 128-bit SID and compressed G-SID, and the compressed G-SID is indicated through the control plane. G-SRv6 can be mixed with SRv6 complete SID scenarios Collaborative work. There is no need to modify the existing address plan and routing plan, and the existing network equipment can be supported by software upgrades, inheriting the SRv6 network programming capabilities.
For the needs of government-enterprise network outreach, low-latency dedicated line products and network slicing can be provided according to the type of application, as well as large-bandwidth pass-throughs that are delay-insensitive. It is also possible to select the corresponding WAN and the path to meet the cloud requirements based on the application information and the location of the cloud. Based on the information security indication in the application information, data traffic can be forwarded to the local cloud and public cloud to ensure that internal data does not leave the enterprise. Value-added services can be provided on demand, with firewalls and intrusion detection paths for Internet traffic, and WAN acceleration for video conferencing traffic.
Wu Hequan added that SRv6 supports cross-network intelligent provisioning and operation and maintenance. The intention engine is to translate business intentions into network language to simulate network design and planning. The analysis engine collects and analyzes user traffic and performance data through real-time telemetry. The execution engine converts network design and planning into specific network commands, and allows network equipment to automatically execute through standard interfaces. The intelligent engine, based on the analysis engine, provides risk prediction and handling suggestions through AI and continuously upgraded experience base, or directly feeds back to the intention engine to realize automatic network optimization.
Wu Hequan said that in the future, the computing power network will face many customer needs, and it is often necessary to provide targeted services to these customers, so that the intent engine can be opened to customers, and customers do not need to know the specific deployment of the operator’s network like programming slices in the cloud. Channels, across multiple networks, are like a single network, online orders, hundreds of parameters, day-level opening, agreed operation and maintenance, and tenant SLA visible.
He emphasized that in the era of cloud-network integration and multi-cloud collaboration, if there are no coordination measures, even if it is multi-cloud, the effect will be poor. Therefore, the network must not only quickly establish communication connections that meet customer bandwidth and delay requirements, but also provide customers with flexible scheduling capabilities for low-cost, high-reliability, and safe use of computing power services. The network architecture evolves to a network architecture centered on example services, truly achieving cloud-network integration.