Network detection and data drainage method and system based on hop-by-hop

Abstract

The invention provides a hop-by-hop network detection and data drainage method and system. The method comprises the following steps: S1, dividing a network into a plurality of network domains; s2, initiating path detection to obtain a detection result; s3, calculating an optimal path based on a detection result, and generating a forwarding strategy; step S4, based on the forwarding strategy, providing an interface to the outside for guiding generation of a forwarding path and a forwarding behavior; and S5, maintaining and detecting the link state. According to the method, a network link detection scheme is planned and implemented from the perspective of a software-defined network, and equipment manufacturers are guided to perform function research and development through a user-defined interface. By means of link analysis indexes with rich functions, the problem of low fault positioning efficiency in a multi-hop network is solved, hop-by-hop visualization of a link state is realized, a basis is provided for network traffic engineering, a forwarding strategy is generated, and a fine customized forwarding service is provided for each service.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and more specifically, to a hop-by-hop network detection and data diversion method and system. Background Technology

[0002] With the rapid development of internet technology, the scale of networks is constantly expanding, and the types of services are becoming increasingly complex. The transmission paths of data packets in the network are becoming more and more dynamic, making the need for network performance optimization, fault diagnosis, and path controllability increasingly urgent. Network path detection, as a core technology to ensure stable network operation, measures and evaluates data packet transmission paths using specific methods to optimize network performance and diagnose network problems.

[0003] Currently, common network path detection technologies include Traceroute and BFD. Traceroute sends data packets with different time-to-live (TTL) values ​​and relies on ICMP echo messages to record the routers along the path and latency information, helping administrators troubleshoot, optimize performance, and identify security risks. BFD, as a general link detection protocol, can detect link faults in milliseconds and, in conjunction with protocols such as MPLS and OSPF, enables rapid link detection and failover.

[0004] However, existing network link detection and diagnosis methods mostly focus on end-to-end connectivity, lacking effective intervention in intermediate node paths. For example, in BFD detection, the first and last nodes send packets independently, while intermediate nodes rely on routing tables, which may lead to inconsistencies between outgoing and return packet paths; while Traceroute can display the path, it cannot control or customize the packet transmission path.

[0005] With the evolution of network technology, path programming technologies, represented by SRv6, have emerged. SRv6 allows defining the path of a data packet by specifying a series of SIDs in the SRH header of the IPv6 packet. Each SID represents a node or link the packet traverses in the network. SRv6 allows service message senders to dynamically adjust paths during packet transmission without requiring specific network configuration. SRv6 network paths are programmable, services are programmable, and forwarding behavior is programmable, bringing application information into the network. This trend necessitates end-to-end hop-by-hop defined path detection technology to provide accurate data for network path programming.

[0006] A patent search revealed invention patent CN105656744A, which discloses a service chain path identification method, device, and service chain. The method includes: a first tunnel edge device receiving traffic data packets carrying path information from a tunnel starting device via a tunnel connection; the first tunnel edge device marking the traffic data packets with IP option identifiers and mapping the IP option identifiers to path information; replacing the path information in the traffic data packets with the IP option identifiers; sending the traffic data packets to a value-added service device; receiving returned traffic data packets from the value-added service device; identifying the path information of the returned traffic data packets based on the IP option identifiers and mappings; replacing the IP option identifiers in the returned traffic data packets with the path information; and sending the traffic data packets carrying the path information to the next target device according to the path information. This patent only addresses service chain path identification, lacks hop-by-hop detection, and is difficult to handle complex network environments.

[0007] In summary, given the problems of the existing technologies, researching a hop-by-hop network detection and data diversion method and system has become a critical task that urgently needs to be addressed. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the purpose of this invention is to provide a hop-by-hop network detection and data diversion method and system.

[0009] A hop-by-hop network detection and data diversion method provided by the present invention includes the following steps:

[0010] Step S1: Divide the network into multiple network domains. The network includes an IPv4 network, an IPv6 network, and an SDN network. The network nodes of the IPv4 network and the IPv6 network are deployed separately, and applications are deployed on the network nodes. The SDN network is centrally managed by the controller.

[0011] Step S2: If the network is an IPv4 network or an IPv6 network, then path probing is initiated at the edge nodes of each network domain. The path probing is executed by the application, the probing results are obtained, and the probing results are returned to the edge node that initiated the probing.

[0012] If the network is an SDN network, then the network nodes in each network domain will report their own node link information to the controller;

[0013] Step S3: If the network is an IPv4 network or an IPv6 network, the edge node summarizes the detection results, selects the optimal path, converts the optimal path into the corresponding forwarding policy, and sends the forwarding policy to the downstream nodes on the optimal path.

[0014] If the network is an SDN network, the controller generates a forwarding policy based on the node link information and distributes the forwarding policy to each relevant network node.

[0015] Step S4: If the network is an IPv4 network, the edge node converts the forwarding policy into an application layer forwarding table, and the network node performs data forwarding according to the application layer forwarding table.

[0016] If the network is an IPv6 network, the edge nodes perform data forwarding through the source routing extension header;

[0017] If the network is an SDN network, each network node executes the forwarding policy issued by the controller;

[0018] Step S5: The link status is detected through the heartbeat keep-alive mechanism. If the service ends or the link is abnormal, the adjustment mechanism and resource reclamation are triggered.

[0019] Preferably, the network is divided into network domains based on network type, physical distance, logical function, security requirements, or management boundaries.

[0020] Preferably, in step S1, edge nodes with dual-stack capabilities are deployed at the edge of the IPv4 network or IPv6 network.

[0021] Preferably, in step S2, the detected parameters include latency, bandwidth, and jitter.

[0022] Preferably, in step S2, in the IPv4 network, the edge node sends a path probe control message, the payload of which contains a list of probe nodes; the application initiates probes through the path probe control message, each intermediate node performs hop-by-hop detection based on the next-hop information in the list of probe nodes, measures latency, bandwidth and jitter parameters, and returns the probe results to the edge node;

[0023] In IPv6 networks, probe packets are generated using the built-in source routing function and sent to the target node through a specified transmission link. During the transmission of the probe packets, each intermediate node performs hop-by-hop detection based on the source routing information and records network parameters, and finally returns the probe results to the edge node.

[0024] In an SDN network, node link information includes bandwidth usage.

[0025] Preferably, in step S3, in an IPv4 or IPv6 network, the edge node aggregates the detection results from each probe node and selects the transmission path with the lowest latency as the optimal path based on the latency data in the detection results.

[0026] Preferably, in step S4, in an IPv4 network, user data packets are forwarded at the edge nodes of the network domain according to the forwarding policy;

[0027] In an IPv6 network, when a user data packet includes a source routing extension header, it is forwarded according to the routing policy in the source routing extension header. When a user data packet does not include a source routing extension header, the edge node encapsulates the forwarding policy into a newly generated source routing extension header, and downstream nodes perform data forwarding according to the source routing extension header of the user data packet.

[0028] Preferably, in step S5, in an IPv4 or IPv6 network, when an edge node detects the termination of a service or an abnormal link, it triggers BFD detection or a dynamic routing protocol to adjust the path and reclaim failed link resources.

[0029] In an SDN network, if the controller detects that a service has ended or a link is abnormal, it updates the forwarding policy and reclaims the resources of the failed link.

[0030] Preferably, the recovery of failed link resources in step S5 includes:

[0031] In IPv4 or IPv6 networks, edge nodes send route cancellation commands to nodes along the optimal path. The route cancellation command contains information about the routing table entries to be deleted. After receiving and parsing the route cancellation command, the nodes along the path remove the specified routing table entries and return an operation confirmation to the edge nodes.

[0032] This invention also provides a hop-by-hop network detection and data diversion system, comprising:

[0033] Module M1 divides the network into multiple network domains, including IPv4 networks, IPv6 networks, and SDN networks. Network nodes for IPv4 and IPv6 networks are deployed separately, with applications deployed on the network nodes. The SDN network is centrally managed by the controller.

[0034] Module M2, if the network is an IPv4 network or an IPv6 network, initiates path probing at the edge nodes of each network domain. The probing content includes latency, bandwidth and jitter. The path probing is executed by the application, the probing results are obtained, and the probing results are returned to the edge node that initiated the probing.

[0035] If the network is an SDN network, then the network nodes in each network domain will report their own node link information to the controller;

[0036] Module M3, if the network is an IPv4 network or an IPv6 network, the edge node summarizes the detection results, selects the optimal path, converts the optimal path into the corresponding forwarding policy, and sends the forwarding policy to the downstream nodes on the optimal path.

[0037] If the network is an SDN network, the controller generates a forwarding policy based on the node link information and distributes the forwarding policy to each relevant network node.

[0038] Module M4: If the network is an IPv4 network, the edge nodes will convert the forwarding policy into an application layer forwarding table, and the network nodes will perform data forwarding according to the application layer forwarding table.

[0039] If the network is an IPv6 network, the edge nodes perform data forwarding through the source routing extension header;

[0040] If the network is an SDN network, each network node executes the forwarding policy issued by the controller;

[0041] Module M5 detects link status through a heartbeat keep-alive mechanism. If it detects that the service has ended or the link is abnormal, it triggers an adjustment mechanism and reclaims resources.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] 1. This invention utilizes a software-defined network architecture and hop-by-hop detection mechanism, guiding equipment manufacturers in functional development through custom interfaces. Leveraging rich link analysis metrics, it addresses the low efficiency of fault location in multi-hop networks, achieving hop-by-hop visualization of link status, providing a basis for network traffic engineering, generating forwarding strategies, and offering refined, customized forwarding services for each service.

[0044] 2. This invention divides network domains based on multiple dimensions such as network type and security requirements, and designs differentiated detection mechanisms for IPv4 / IPv6 / SDN network characteristics to ensure full-scenario link detection capabilities in a dual-stack network environment.

[0045] 3. This invention combines application layer path probing with centralized management by the SDN controller. By probing multi-dimensional indicators such as application execution latency, bandwidth, and jitter, it solves the inefficiency problem of traditional network fault location relying on hop-by-hop troubleshooting.

[0046] 4. The edge nodes of this invention dynamically generate the optimal path based on the detection results, and implement hop-by-hop forwarding control through the application layer forwarding table or source routing extension header, providing a unified and fine-grained traffic scheduling capability for networks with different protocol stacks.

[0047] 5. The heartbeat keep-alive mechanism of this invention works in conjunction with BFD / dynamic routing protocol to achieve millisecond-level link fault detection and resource reclamation. Attached Figure Description

[0048] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0049] Figure 1 This is a flowchart of a hop-by-hop network detection and data diversion method according to an embodiment of the present invention;

[0050] Figure 2 Network segmentation is performed based on IP characteristics in this embodiment of the invention;

[0051] Figure 3 This is a schematic diagram of SDN network management in an embodiment of the present invention. Detailed Implementation

[0052] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0053] This invention provides a hop-by-hop network detection and data redirection method and system, focusing on path status detection in IP networks, suitable for large-scale networking scenarios. In this scenario, inter-node communication often needs to traverse multiple nodes or autonomous systems. Operations personnel plan forwarding paths for various service traffic based on traffic engineering techniques. This invention achieves the detection and optimization of such network paths through the following steps: Step S1, dividing the network into multiple network domains; Step S2, initiating path probing and obtaining the probing results; Step S3, calculating the optimal path based on the probing results and generating a forwarding strategy; Step S4, providing an interface based on the forwarding strategy to guide the generation of forwarding paths and forwarding behaviors; Step S5, maintaining and detecting link status. This invention plans and implements a network link detection scheme from the perspective of software-defined networking, guiding equipment manufacturers to develop functions through custom interfaces. Utilizing rich link analysis metrics, it solves the problem of low fault location efficiency in multi-hop networks, achieves hop-by-hop visualization of link status, provides a basis for network traffic engineering, generates forwarding strategies, and provides refined customized forwarding services for each service.

[0054] Example 1:

[0055] Figure 1 This is a flowchart of a hop-by-hop network detection and data diversion method according to an embodiment of the present invention.

[0056] like Figure 1 As shown, this embodiment provides a hop-by-hop network detection and data diversion method, including the following steps:

[0057] Step S1: Divide the network into multiple network domains. The network includes an IPv4 network, an IPv6 network, and an SDN network. Network nodes for the IPv4 network and the IPv6 network are deployed separately, and applications are deployed on the network nodes.

[0058] Specifically, the network is divided into network domains based on network type, physical distance, logical function, security requirements, or management boundaries. Each network domain may have different protocol characteristics. In this embodiment, it is divided into IPv4 networks and IPv6 networks according to IP type.

[0059] Figure 2 Network segmentation is performed on the IP features in this embodiment of the invention.

[0060] like Figure 2 As shown, circles represent network nodes, and characters inside the circles represent node numbers. Network domains include network domain A1, network domain A2, and network domain A3. Network domain A1 contains network nodes n2, n4, and n7; network domain A2 contains network nodes n3, n5, and n6; and network domain A3 contains network nodes n1 and n8, as well as network domains A1 and A2.

[0061] Furthermore, in step S1, edge nodes with dual-stack capabilities are deployed at the edge of the IPv4 or IPv6 network.

[0062] Figure 3 This is a schematic diagram of SDN network management in an embodiment of the present invention.

[0063] like Figure 3 As shown, the SDN network is centrally managed by the controller.

[0064] Step S2: If the network is an IPv4 network or an IPv6 network, path probing is initiated at the edge nodes of each network domain. The path probing is executed by the application, the probing results are obtained, and the probing results are returned to the edge node that initiated the probing. If the network is an SDN network, the network nodes of each network domain report their own node link information to the controller.

[0065] Specifically, in step S2, the detection content includes latency, bandwidth, and jitter.

[0066] More specifically, in step S2, in the IPv4 network, the edge node sends a path probe control message, the payload of which contains a list of probe nodes; the application initiates a probe through the path probe control message, each intermediate node performs hop-by-hop detection based on the next-hop information in the list of probe nodes, measures latency, bandwidth and jitter parameters, and returns the probe results to the edge node;

[0067] In this embodiment, the link generation process in the IPv6 network is similar to that in the IPv4 probe process. The core difference is that in the IPv6 network, probe packets are generated using the built-in source routing function and sent to the target node through a specified transmission link. During the transmission of the probe packets, each intermediate node performs hop-by-hop detection based on the source routing information and records network parameters, and finally returns the probe results to the edge node. In the SDN network, node link information includes bandwidth usage.

[0068] Step S3: If the network is an IPv4 network or an IPv6 network, the edge nodes summarize the detection results, select the optimal path, convert the optimal path into the corresponding forwarding policy, and send the forwarding policy to the downstream nodes on the optimal path; if the network is an SDN network, the controller generates a forwarding policy based on the node link information and sends the forwarding policy to each relevant network node.

[0069] In this embodiment, when the optimal path is: network node n1-network node n2-network node n7-network node n8, the forwarding strategy includes: network node n1 is configured to forward from interface 1 to network node n2, node n2 is configured to receive from interface 1 and forward from interface 2 to network node n7, network node n7 is configured to receive from interface 1 and forward from interface 2 to network node n8, and network node n8 is configured to receive data from interface 1.

[0070] Specifically, in step S3, in an IPv4 or IPv6 network, the edge node aggregates the detection results from each probe node and selects the transmission path with the lowest latency as the optimal path based on the latency data in the detection results.

[0071] In step S4, if the network is an IPv4 network, the edge node converts the forwarding policy into an application layer forwarding table, and the network node performs data forwarding according to the application layer forwarding table; if the network is an IPv6 network, the edge node performs data forwarding through the source routing extension header; if the network is an SDN network, each network node executes the forwarding policy issued by the controller.

[0072] Specifically, in step S4, in an IPv4 network, user data packets are forwarded at the edge nodes of the network domain according to the forwarding policy; in an IPv6 network, when a user data packet includes a source routing extension header, it is forwarded according to the routing policy in the source routing extension header; when a user data packet does not include a source routing extension header, the edge nodes encapsulate the forwarding policy into a newly generated source routing extension header, and downstream nodes perform data forwarding according to the source routing extension header of the user data packet.

[0073] Step S5: The link status is detected through the heartbeat keep-alive mechanism. If the service ends or the link is abnormal, the adjustment mechanism and resource reclamation are triggered.

[0074] Specifically, in step S5, in an IPv4 or IPv6 network, when an edge node detects the termination of a service or a link anomaly, it triggers BFD detection or a dynamic routing protocol to adjust the path and reclaim failed link resources; in an SDN network, if the controller detects the termination of a service or a link anomaly, it updates the forwarding policy and reclaims failed link resources.

[0075] Furthermore, step S5, which involves reclaiming failed link resources, includes:

[0076] In IPv4 or IPv6 networks, edge nodes send route cancellation commands to nodes along the optimal path. The route cancellation command contains information about the routing table entries to be deleted. After receiving and parsing the route cancellation command, the nodes along the path remove the specified routing table entries and return an operation confirmation to the edge nodes.

[0077] In this embodiment of the IPv4 network, a link probe signaling control message is set up for link probe. A list of link probe nodes is defined in the packet payload field. Each network node performs link detection according to the next-hop information in the link probe node list. The detection content includes latency, bandwidth, and jitter. Each node returns the detection results to the head node, i.e., the node that initiated the detection request. After synthesizing the calculation results of multiple links in end-to-end communication, the head node selects the best link with the lowest latency. Next, the head node sends forwarding policies to the nodes along the best link. The forwarding policy content is propagated downwards as packet content. After application parsing, each node independently generates routing table information and returns the results. The teardown process of link forwarding behavior is similar to the generation process, initiated by the head node and responded to by intermediate nodes along the path. Simultaneously, a heartbeat keep-alive mechanism is used to prevent problems such as abnormal power-off of the head node.

[0078] The process of creating, maintaining, and deleting links in an IPv6 network is similar to that of IPv6. The core difference is that IPv6 has built-in network path programming capabilities, which allow you to specify links using IPv6's source routing capabilities and complete the operation at the corresponding node.

[0079] In an SDN network, the controller is responsible for generating forwarding policies for all network nodes. The head node also receives the packet forwarding paths generated and distributed by the controller. The head node does not need to maintain the forwarding links. Each network node reports probe data to the controller. The controller integrates data such as network bandwidth, latency, and jitter of each node to perform calculations and distributes forwarding policies to each network node. Subsequent operations such as deleting forwarding policies are also handled by the controller.

[0080] To reduce the length of the path field in data packets, data is transmitted to the edge nodes of the network domain, and the calculation results of the edge nodes are then used to traverse the network domains.

[0081] Example 2:

[0082] The present invention also provides a hop-by-hop network detection and data diversion system. The hop-by-hop network detection and data diversion system can be implemented by executing the process steps of the hop-by-hop network detection and data diversion method. That is, those skilled in the art can understand the hop-by-hop network detection and data diversion method as a preferred embodiment of the hop-by-hop network detection and data diversion system.

[0083] Specifically, the hop-by-hop network detection and data diversion system includes:

[0084] Module M1 divides the network into multiple network domains, including IPv4 networks, IPv6 networks, and SDN networks. Network nodes for IPv4 and IPv6 networks are deployed separately, with applications deployed on the network nodes. The SDN network is centrally managed by the controller.

[0085] Module M2, if the network is an IPv4 network or an IPv6 network, initiates path probing at the edge nodes of each network domain. The probing content includes latency, bandwidth and jitter. The path probing is executed by the application, the probing results are obtained, and the probing results are returned to the edge node that initiated the probing.

[0086] If the network is an SDN network, then the network nodes in each network domain will report their own node link information to the controller;

[0087] Module M3, if the network is an IPv4 network or an IPv6 network, the edge node summarizes the detection results, selects the optimal path, converts the optimal path into the corresponding forwarding policy, and sends the forwarding policy to the downstream nodes on the optimal path.

[0088] If the network is an SDN network, the controller generates a forwarding policy based on the node link information and distributes the forwarding policy to each relevant network node.

[0089] Module M4: If the network is an IPv4 network, the edge nodes will convert the forwarding policy into an application layer forwarding table, and the network nodes will perform data forwarding according to the application layer forwarding table.

[0090] If the network is an IPv6 network, the edge nodes perform data forwarding through the source routing extension header;

[0091] If the network is an SDN network, each network node executes the forwarding policy issued by the controller;

[0092] Module M5 detects link status through a heartbeat keep-alive mechanism. If it detects that the service has ended or the link is abnormal, it triggers an adjustment mechanism and reclaims resources.

[0093] The key points of this invention are explained below:

[0094] 1. Link-based stateful inspection: After specifying a path between network nodes, inspection packets are sent along the specified path. The inspection result is returned to the head node at each hop, and the link inspection results are also summarized at the head node. The head node then generates the optimal path for transmission between nodes. In IPv6 networks, path probe packets are generated based on IPv6 source routing functionality. After hop-by-hop inspection, the head node summarizes the results and generates the optimal path.

[0095] 2. In IPv4 networks, source routing is generally not supported. Path establishment is initiated at the application layer by the head node, generating independent forwarding policies for each service application at downstream nodes. Downstream forwarding policies use path interfaces or loose specification to indirectly complete the source routing function of IPv4. The working status of the forwarding table is maintained through heartbeats between downstream nodes and the head node. Each downstream node checks its own network status and reports the results to the head node, which then aggregates the data to generate source routing link information.

[0096] 3. In an SDN network, each node actively reports its network status and neighbor link status values ​​to the controller. The controller generates forwarding policies through intelligent algorithms and generates source routing links for service header nodes based on the controller's existing functions. Network devices with source routing capabilities encapsulate packets using the source routes issued by the controller, while network nodes that do not support source routing forward data according to the single-hop ordinary forwarding table generated by the controller.

[0097] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0098] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A hop-by-hop network detection and data diversion method, characterized in that, Includes the following steps: Step S1: Divide the network into multiple network domains, including an IPv4 network, an IPv6 network, and an SDN network. The network nodes of the IPv4 network and the IPv6 network are deployed separately, and applications are deployed on the network nodes. The SDN network is centrally managed by the controller. Step S2: If the network is an IPv4 network or an IPv6 network, then path probing is initiated at the edge nodes of each network domain. The path probing is executed by the application to obtain the probing results and return the probing results to the edge node that initiated the probing. If the network is an SDN network, then the network nodes in each network domain will report their own node link information to the controller; Step S3: If the network is an IPv4 network or an IPv6 network, the edge node summarizes the detection results, selects the optimal path, converts the optimal path into a corresponding forwarding policy, and sends the forwarding policy to the downstream nodes on the optimal path. If the network is an SDN network, the controller generates a forwarding policy based on the node link information and sends the forwarding policy to each relevant network node. Step S4: If the network is an IPv4 network, the edge node converts the forwarding policy into an application layer forwarding table, and the network node performs data forwarding according to the application layer forwarding table. If the network is an IPv6 network, the edge node performs data forwarding through the source routing extension header; If the network is an SDN network, each network node executes the forwarding policy issued by the controller; Step S5: The link status is detected through the heartbeat keep-alive mechanism. If the service ends or the link is abnormal, the adjustment mechanism and resource reclamation are triggered.

2. The hop-by-hop network detection and data diversion method according to claim 1, characterized in that, In step S1, the network is divided into network domains based on network type, physical distance, logical function, security requirements, or management boundaries.

3. The hop-by-hop network detection and data diversion method according to claim 1, characterized in that, In step S1, edge nodes with dual-stack capabilities are deployed at the edge of the IPv4 network or the IPv6 network.

4. The hop-by-hop network detection and data diversion method according to claim 1, characterized in that, In step S2, the detection content includes latency, bandwidth, and jitter.

5. The hop-by-hop network detection and data diversion method according to claim 4, characterized in that, In step S2, in the IPv4 network, the edge node sends a path probe control message, the payload of which contains a list of probe nodes; the application initiates a probe through the path probe control message, each intermediate node performs hop-by-hop detection based on the next-hop information in the list of probe nodes, measures latency, bandwidth and jitter parameters, and returns the detection results to the edge node; In the IPv6 network, probe packets are generated using the built-in source routing function and sent to the target node through a specified transmission link. During the transmission of the probe packets, each intermediate node performs hop-by-hop detection based on the source routing information and records network parameters, and finally returns the probe results to the edge node. In the SDN network, the node link information includes bandwidth usage.

6. The hop-by-hop network detection and data diversion method according to claim 5, characterized in that, In step S3, in the IPv4 network or the IPv6 network, the edge node aggregates the detection results from each detection node and selects the transmission path with the lowest latency as the optimal path based on the latency data in the detection results.

7. The hop-by-hop network detection and data diversion method according to claim 6, characterized in that, In step S4, in the IPv4 network, user data packets are forwarded at the edge nodes of the network domain according to the forwarding policy. In the IPv6 network, when a user data packet already includes a source routing extension header, it is forwarded according to the routing policy in the source routing extension header; When a user data packet does not include a source routing extension header, the edge node encapsulates the forwarding policy into a newly generated source routing extension header, and downstream nodes perform data forwarding according to the source routing extension header of the user data packet.

8. The hop-by-hop network detection and data diversion method according to claim 7, characterized in that, In step S5, when the edge node detects the end of a service or a link abnormality in the IPv4 network or the IPv6 network, it triggers BFD detection or dynamic routing protocol to adjust the path and reclaim failed link resources. In the SDN network, if the controller detects that a service has ended or a link is abnormal, it updates the forwarding policy and reclaims the failed link resources.

9. The hop-by-hop network detection and data diversion method according to claim 8, characterized in that, The step S5 of recovering failed link resources includes: In the IPv4 network or the IPv6 network, the edge node sends a route cancellation instruction to the nodes along the optimal path. The route cancellation instruction contains routing table entry information to be deleted. After receiving and parsing the route cancellation instruction, the nodes along the path remove the specified routing table entry and return an operation confirmation to the edge node.

10. A hop-by-hop network detection and data diversion system, characterized in that, include: Module M1 divides the network into multiple network domains, including an IPv4 network, an IPv6 network, and an SDN network. The network nodes of the IPv4 network and the IPv6 network are deployed separately, and applications are deployed on the network nodes. The SDN network is centrally managed by the controller. Module M2, if the network is an IPv4 network or an IPv6 network, initiates path probing at the edge nodes of each network domain. The probing content includes latency, bandwidth and jitter. The path probing is executed by the application, the probing results are obtained, and the probing results are returned to the edge node that initiated the probing. If the network is an SDN network, then the network nodes in each network domain will report their own node link information to the controller; Module M3: If the network is an IPv4 network or an IPv6 network, the edge node summarizes the detection results, selects the optimal path, converts the optimal path into a corresponding forwarding policy, and sends the forwarding policy to the downstream nodes on the optimal path. If the network is an SDN network, the controller generates a forwarding policy based on the node link information and sends the forwarding policy to each relevant network node. Module M4: If the network is an IPv4 network, the edge node converts the forwarding policy into an application layer forwarding table, and the network node performs data forwarding according to the application layer forwarding table; If the network is an IPv6 network, the edge node performs data forwarding through the source routing extension header; If the network is an SDN network, each network node executes the forwarding policy issued by the controller; Module M5 detects link status through a heartbeat keep-alive mechanism. If it detects that the service has ended or the link is abnormal, it triggers an adjustment mechanism and reclaims resources.

Images