Local area network system and backbone network system

Abstract

The invention provides a local area network system and a backbone network system, belongs to the technical field of computers, and particularly relates to the technical field of communication, computer networks, routing and fault detection. According to the specific implementation scheme, the regional network system comprises a plurality of network nodes, each network node comprises a plurality of detectors and a plurality of forwarding devices, the detectors are in communication connection with the forwarding devices respectively, and the forwarding devices are in communication connection with the forwarding devices included in the other network nodes respectively; wherein the forwarding equipment is used for receiving a data packet and forwarding the data packet to the detector determined based on a destination address of the data packet under the condition that the destination address of the data packet belongs to a first routing network segment of the forwarding equipment; and the detector obtains fault information based on the received data packet.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to the fields of communication, computer networks, routing, and fault detection technology. More specifically, it relates to a regional network system and a backbone network system. Background Technology

[0002] With the rapid development of Internet technology, Internet cloud service providers and telecommunications operators are increasingly reliant on networks. The reliability, stability, and service quality of the network are directly related to the normal operation of business and user experience. Monitoring the operating status of the network system and handling faults are key links to ensure the reliability and stability of the network. Summary of the Invention

[0003] This disclosure provides a regional network system and a backbone network system.

[0004] According to one aspect of this disclosure, a regional network system is provided, comprising: multiple network nodes, each network node including multiple probes and multiple forwarding devices, wherein the multiple probes are communicatively connected to each of the multiple forwarding devices, and the forwarding devices are communicatively connected to multiple forwarding devices included in each of the multiple other network nodes; wherein the forwarding devices are used to receive data packets, and if the destination address of the data packet belongs to a first routing network segment of the forwarding device, forward the data packet to a probe determined based on the destination address of the data packet; the probe obtains fault information based on the received data packet.

[0005] According to another aspect of this disclosure, a backbone network system is provided, comprising a plurality of regional network systems as described above.

[0006] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0007] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0008] Figure 1 A schematic diagram of a local area network system according to an embodiment of the present disclosure is shown.

[0009] Figure 2 A schematic diagram of a local area network system according to another embodiment of the present disclosure is shown.

[0010] Figure 3 A schematic flowchart of fault detection according to an embodiment of the present disclosure is shown.

[0011] Figure 4 A schematic diagram of a backbone network system according to an embodiment of the present disclosure is shown.

[0012] Figure 5 A schematic diagram of a local area network system according to another embodiment of the present disclosure is shown.

[0013] Figure 6 A schematic diagram of fault detection according to another embodiment of the present disclosure is shown. Detailed Implementation

[0014] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0015] For internet cloud service providers, ensuring high availability of cloud services requires real-time and accurate monitoring of network quality. Data interaction between multiple regional data centers relies on a stable interconnection network; any failure could lead to data transmission interruptions or delays, impacting business continuity. Simultaneously, cross-regional disaster recovery also places extremely high demands on rapid response and handling of network faults. Only by promptly locating and repairing faults can uninterrupted service be guaranteed in emergencies. For telecommunications operators, the health of the backbone network directly determines the overall quality of communication services. Effective monitoring methods are needed to ensure network quality and provide early warnings of faults, preventing widespread communication disruptions due to failures. Furthermore, cross-regional network operation and maintenance management involves a wide range of areas and numerous nodes, demanding even more stringent requirements for the accuracy and efficiency of fault location. Efficient fault handling can reduce user complaints and improve service reputation.

[0016] However, current backbone network monitoring and fault location technologies still have significant limitations and cannot meet the actual needs of the aforementioned application areas. Among related technologies, backbone network fault analysis mainly employs two methods: The first is single-node detection, which relies on ADC (Advanced Datacenter Core) nodes independently sending probe information. This method can only determine whether the ADC node itself or the SC (Super Core) node has a fault. It cannot effectively identify specific fault points in the network link, such as link interruptions or link congestion, resulting in an extremely broad fault location range. The second method is second-level monitoring. Although this technology can quickly detect changes in network status and provide macro-level feedback on network operation, such as achieving second-level data acquisition and status monitoring through a Sonar system, it also cannot accurately locate the specific fault location. It can only indicate that an anomaly has occurred in the network, but cannot determine which link node or transmission path the anomaly occurred on.

[0017] The aforementioned limitations of these technologies have led to a series of serious problems. First, the limitations in fault location mean that when network problems occur, maintenance personnel cannot accurately pinpoint the source of the fault, and can only conduct extensive troubleshooting, which greatly affects the efficiency of fault diagnosis and repair. When a network failure occurs, maintenance personnel need to check each ADC node, SC node, and a large number of network links one by one, which not only consumes a lot of time, but may also further expand the scope of the fault due to the excessively long troubleshooting process. Second, the low efficiency of fault diagnosis and repair directly affects network stability and service quality. Prolonged failures can cause data transmission delays, packet loss, and even service interruptions. For Internet cloud service providers, this may lead to user data loss, business suspension, and thus user churn; for telecommunications operators, it may cause call interruptions, Internet lag, and other problems, seriously degrading the user experience. Finally, low efficiency in fault location and repair also significantly increases maintenance costs. The fault diagnosis process requires more maintenance personnel for manual troubleshooting, and the extended fault handling time will also increase additional labor costs and resource consumption. At the same time, the business losses caused by the failure also indirectly increase the overall operating costs.

[0018] In summary, the current backbone network monitoring and fault location technologies suffer from deficiencies in fault location accuracy and troubleshooting efficiency, severely impacting network reliability, stability, and service quality. They also increase operational costs and fail to meet the needs of internet cloud service providers and telecom operators in network quality assurance, fault handling, and operational management. Therefore, developing a technical solution capable of efficient and accurate fault location and network monitoring to improve network reliability and maintenance efficiency while reducing operating costs has become a pressing technical challenge in this field.

[0019] In view of this, embodiments of this disclosure provide an optimized detection scheme that combines multi-plane detection and a fully interconnected topology to modify the routing of ADC nodes and / or SC nodes in the network system to achieve accurate fault location. Specifically, embodiments of this disclosure provide a regional network system, including: multiple network nodes, each network node including multiple detectors and multiple forwarding devices, the multiple detectors communicating with each of the multiple forwarding devices, and the forwarding devices communicating with multiple forwarding devices included in each of the other network nodes; wherein, the forwarding devices are used to receive data packets, and if the destination address of the data packet belongs to a first routing network segment of the forwarding device, forward the data packet to a detector determined based on the destination address of the data packet; the detector obtains fault information based on the received data packet.

[0020] Figure 1 A schematic diagram of a local area network system according to an embodiment of the present disclosure is shown.

[0021] like Figure 1 As shown, a regional network system may include multiple network nodes, such as network node n1 and network node n2.

[0022] A regional network system can be defined as a network system composed of network nodes of a specific part of a backbone network system; that is, a backbone network system can be composed of multiple regional network systems.

[0023] A network node can be represented as a data center-level unit within the backbone network coverage area that has independent detection or forwarding capabilities, and each network node can correspond to a physical data center.

[0024] A network node can include multiple probes and multiple forwarding devices. Multiple probes can each communicate with multiple forwarding devices. Specifically, each probe can communicate with one forwarding device, and each forwarding device can communicate with one probe. The forwarding devices communicate with multiple forwarding devices included in each of the other network nodes, thus allowing the various network nodes to interconnect through the forwarding devices, forming a fully interconnected network architecture.

[0025] For example, network node n1 may include probe s1, probe s2, forwarding device d1, and forwarding device d2. Probe s1 can communicate with forwarding device d1, and probe s2 can communicate with forwarding device d2. Network node n2 may include probe s3, probe s4, forwarding device d3, and forwarding device d4. Probe s3 can communicate with forwarding device d3, and probe s4 can communicate with forwarding device d4. Forwarding device d1 can communicate with both forwarding device d3 and forwarding device d4, and forwarding device d2 can communicate with both forwarding device d3 and forwarding device d4.

[0026] Direct interconnection of multiple forwarding devices within each network node significantly increases network architecture complexity. For example, in a regional network system with 100 forwarding devices, direct interconnection results in 4950 communication links between these devices. Therefore, in some embodiments, the regional network system may also include a route reflector, which can communicate with each of the multiple forwarding devices within each network node. Using the route reflector, forwarding devices can learn the addresses of forwarding devices in other network nodes via the BGP (Border Gateway Protocol), enabling flow forwarding between these devices and reducing topology complexity. The method by which forwarding devices learn the network addresses of other forwarding devices using the route reflector is not limited here.

[0027] The probe can receive data packets that match its own network address and are forwarded by the forwarding device, but will not receive irrelevant traffic from other network segments or other nodes.

[0028] Forwarding devices can be used to forward traffic in a network system. Each forwarding device can be bound to a probe plane and will only process probe traffic for that corresponding probe plane, without forwarding traffic from other planes or other network segments.

[0029] For any given network node, "other network nodes" can refer to the equipment room units within the same regional network system that also contain probes and forwarding equipment. For example, a regional network system may include network nodes A, B, and C. Then, for network node A, multiple "other network nodes" can include network nodes B and C; similarly, for network node B, multiple "other network nodes" can include network nodes A and C. As another example, a regional network system may include multiple network nodes n1 and n2. Then, for network node n1, "other network nodes" can refer to network node n2; similarly, for network node n2, "other network nodes" can refer to network node n1.

[0030] In a regional network system, each forwarding device needs to forward data packets according to a configured routing policy. This routing policy can be configured so that the forwarding device processes data packets destined for specific addresses while filtering out data packets destined for other routing segments. Optionally, the forwarding settings can be configured with a first routing segment, where the specific address can be represented by the network addresses included in the first routing segment. That is, for any forwarding device, it can receive data packets and, if the destination address of the data packet belongs to the forwarding device's first routing segment, forward the data packet to a probe machine with which it has established a communication link. Conversely, if the destination address of the data packet does not belong to the forwarding device's first routing segment, the forwarding device can discard the data packet without processing it.

[0031] For example, if forwarding device d3 receives data packet Data1 and the destination address of data packet Data1 belongs to the first routing network segment of forwarding device d3, then forwarding device d3 can forward data packet Data1 to probe device s3; however, if the destination address of data packet Data1 does not belong to the first routing network segment of forwarding device d3, then forwarding device d3 can filter data packet Data1.

[0032] If forwarding device D4 receives data packet Data2, and the destination address of data packet Data2 belongs to the first routing network segment of forwarding device D4, then forwarding device D4 can forward data packet Data2 to probe device s4; however, if the destination address of data packet Data2 does not belong to the first routing network segment of forwarding device D4, then forwarding device D4 can filter data packet Data2.

[0033] Fault information can be used to indicate a communication link failure associated with any forwarding device of the current network node, or it can be used to indicate a failure of any forwarding device of the current network node itself; there is no limitation on this.

[0034] According to embodiments of this disclosure, by configuring the forwarding devices of each network node for routing, each forwarding device is used only to forward data packets of its local area network probe plane during fault detection. The probe can determine network faults based on the missing data packets received, thereby effectively improving the efficiency of fault location and reducing maintenance costs.

[0035] The local area network system provided in this disclosure will be further described below with reference to the accompanying drawings and specific embodiments. In the embodiments of this disclosure, the network address of the forwarding device or the destination address of the data packet can be an IPv4 (Internet Protocol version 4) address, that is, the network address of the forwarding device or the destination address of the data packet can be represented by a 32-bit binary number; in some embodiments, the source address of the forwarding device or the destination address of the data packet can be an IPv6 (Internet Protocol version 6) address, which is not limited here.

[0036] In a regional network system, the source address of a forwarding device or the destination address of a data packet can be divided into multiple feature fields, each of which can be a continuous range of binary bits. For example, the destination address of a data packet might be 192.108.52.1, which can be divided into four feature fields. The first feature field can include bits 1 to 8 of the address, represented as 192 in decimal; the second feature field can include bits 9 to 16 of the address, represented as 108 in decimal; the third feature field can include bits 17 to 24 of the address, represented as 52 in decimal; and the fourth feature field can include bits 25 to 32 of the address, represented as 1 in decimal.

[0037] In embodiments of this disclosure, the value of the second feature field of the address can be used to represent different local area network probe planes, and the value of the first feature field of the address can be used to represent different network nodes. The second feature field is a continuous range of binary bits, and the first feature field is a continuous range of binary bits.

[0038] According to embodiments of this disclosure, the destination address of the data packet belongs to the first routing network segment of the forwarding device, and the destination address of the data packet is consistent with the network address included in the first routing network segment of the forwarding device.

[0039] According to embodiments of this disclosure, the value of the first characteristic field of the destination address of a data packet can be used to indicate the network node to which the forwarding device belongs.

[0040] In one example, the destination address of the data packet and the first routing network segment of the forwarding device are IPv4 addresses. The second feature field can be represented as bits 17 to 24 of the address, the first feature field can be represented as bits 25 to 32 of the address, and bits 1 to 16 of the address can be fixed values. For example, if the destination address of the data packet is 100.0.1.1, based on bits 25 to 32 of the destination address, the node identifier of the network node receiving the data can be determined to be 1; based on bits 17 to 24 of the destination address, the network segment identifier of the local area network probe plane to which the destination address of the data packet belongs can be determined to be 1. Among the network nodes with node identifier 1, the first routing network segment of the forwarding device corresponding to the local area network probe plane with network segment identifier 1 can be represented as 100.0.1.1 / 32.

[0041] In the embodiments of this disclosure, under fault detection conditions, a network node can send data packets to multiple other network nodes in a full-coverage manner, so that each network node can determine the fault of the network node based on the data packets received by its detector and the data packets lost.

[0042] Figure 2 A schematic diagram of a local area network system according to another embodiment of the present disclosure is shown.

[0043] like Figure 2 As shown, the probe can also send data packets to other probes in other network nodes based on direct links between them. Specifically, the probe can send data packets to forwarding devices that it is communicatively connected to. These forwarding devices can send the data packets to multiple other forwarding devices through communication links between themselves and other network nodes. After receiving the data packets, the other forwarding devices can use their configured routing policies to determine whether to forward the data packets to the corresponding probe.

[0044] For example, a regional network system includes 5 network nodes. Each network node can be configured with 4 forwarding devices and 4 probes. The 4 forwarding devices can each be configured with an independent probe plane. For any given network node, each probe of that node can send data packets to the 4 probes of the other 4 network nodes. For the receiving probe, it can receive data packets forwarded by the other 4 forwarding devices located on the same probe plane. If the sending probe only sends one data packet to the 4 probes of the other 4 network nodes, then the receiving probe can receive 4 data packets located on the same probe plane.

[0045] For example, probe s1 can generate a data packet and send it to forwarding devices d3 and d4 via forwarding device d1; probe s2 can generate a data packet and send it to forwarding devices d3 and d4 via forwarding device d2.

[0046] Figure 3 A schematic flowchart of fault detection according to an embodiment of the present disclosure is shown.

[0047] like Figure 3 As shown, for network node n1, probe s1 can generate probe data packet 301, and the destination address of probe data packet 301 is X1.X2.0.n2.

[0048] Network node n1 may include three forwarding devices: forwarding device n1-1, forwarding device n1-2, and forwarding device n1-3. The first routing network segment of forwarding device n1-1 can be X1.X2.0.n1 / 32, the first routing network segment of forwarding device n1-2 can be X1.X2.1.n1 / 32, and the first routing network segment of forwarding device n1-3 can be X1.X2.2.n1 / 32.

[0049] Forwarding device n1-1 can communicate with probe s1. Therefore, for probe data packet 301 generated by probe s1, forwarding device n1-1 can forward probe data packet 301 to another forwarding device in the network node. For example, based on the values ​​of bits 25 to 32 of the destination address of probe data packet 301, the network node receiving probe data packet 301 can be determined to be network node n2. Network node n2 includes forwarding devices n2-1, n2-2, and n2-3. Then, forwarding device n1-1 can provide probe data packet 301 to forwarding devices n2-1, n2-2, and n2-3 respectively.

[0050] Optionally, under certain fault conditions, forwarding device n1-2 or forwarding device n1-3 may also receive the probe data packet 301. In this case, forwarding device n1-2 or forwarding device n1-3 can also forward the probe data packet 301. Taking forwarding device n1-2 as an example, after forwarding device n1-2 receives the probe data packet 301, it can also provide the probe data packet 301 to forwarding devices n2-1, n2-2, and n2-3 respectively.

[0051] In network node n2, the first routing network segment of forwarding device n2-1 can be X1.X2.0.n2 / 32, the first routing network segment of forwarding device n2-2 can be X1.X2.1.n2 / 32, and the first routing network segment of forwarding device n2-3 can be X1.X2.2.n2 / 32.

[0052] The three forwarding devices of network node n2 can perform route detection on the received probe data packets.

[0053] Specifically, forwarding device n2-1 can determine whether the destination address of probe data packet 301 belongs to the first routing network segment of forwarding device n2-1. The destination address of probe data packet 301 is X1.X2.0.n2, and the first routing network segment of forwarding device n2-1 is X1.X2.0.n2 / 32. Therefore, the destination address of probe data packet 301 belongs to the first routing network segment of forwarding device n2-1. Thus, forwarding device n2-1 can forward probe data packet 301 to a probe machine connected to it, such as probe machine s3.

[0054] Similarly, forwarding devices n2-2 and n2-3 also need to determine whether the destination address of probe data packet 301 belongs to the first routing network segment of forwarding device n2-2 or forwarding device n2-3. If the determination result indicates that the destination address of probe data packet 301 does not belong to the first routing network segment of forwarding device n2-2 or forwarding device n2-3, forwarding device n2-2 or forwarding device n2-3 can filter probe data packet 301 and not perform any processing on probe data packet 301.

[0055] If the forwarding device n2-1 determines that the destination address of the probe data packet 301 belongs to its first routing network segment, the forwarding device n2-1 can provide the probe data packet 301 to the probe machine s3.

[0056] In one example, the regional network system may include network node n3 in addition to network node n1 and network node n2. Network node n3 also includes three forwarding devices: forwarding device n3-1, forwarding device n3-2, and forwarding device n3-3. The first routing network segment of forwarding device n3-1 can be X1.X2.0.n3 / 32, the first routing network segment of forwarding device n3-2 can be X1.X2.1.n3 / 32, and the first routing network segment of forwarding device n3-3 can be X1.X2.2.n3 / 32. Multiple probes on network node n1 can generate probe data packets with destination addresses X1.X2.0.n2, X1.X2.1.n2, X1.X2.2.n2, X1.X2.0.n3, X1.X2.1.n3, and X1.X2.2.n3, respectively. These probe data packets with destination addresses X1.X2.0.n2, X1.X2.1.n2, X1.X2.2.n2, X1.X2.0.n3, X1.X2.1.n3, and X1.X2.2.n3 are then forwarded by forwarding devices n1-1, n1-2, and n1-3 to forwarding devices n2-1, n2-2, n2-3, n3-1, n3-2, and n3-3 of network node n2, respectively.

[0057] Accordingly, multiple probes of any network node can receive probe data packets from multiple forwarding devices included in each of the other network nodes. For example, assuming no failures in any of the devices and communication links, the probe data packets that the three probes of network node n2 can receive include: a probe data packet with a destination address of X1.X2.0.n2 received via forwarding device n2-1, a probe data packet with a destination address of X1.X2.1.n2 received via forwarding device n2-2, and a probe data packet with a destination address of X1.X2.2.n2 received via forwarding device n2-3.

[0058] If a fault occurs in the device or communication link, the probe may not be able to receive all the probe data packets, i.e., packet loss occurs. In this case, the probe can be used to count the packet loss, identify the unreceived probe data packets, obtain the first packet loss information, and obtain the fault information based on the first packet loss information.

[0059] According to an embodiment of this disclosure, the first packet loss information includes packet loss in the communication behavior between the first probe and the first probe, which is located in a first network node; wherein, based on the network address of the first probe, the probe determines a first forwarding device from a plurality of forwarding devices, and determines a second forwarding device from a plurality of forwarding devices included in the first network node, and determines that the fault information represents a communication link failure between the first forwarding device and the second forwarding device.

[0060] Specifically, the probe can receive multiple data packets from other network nodes located on the same local area network (LAN) probe plane. The probe can classify the received data packets according to their destination addresses and calculate the packet loss rate for each category. Each category can correspond to another network node. The first packet loss information can be expressed as follows: if the packet loss rate is high in a category, the network node corresponding to that category can be identified, i.e., the first network node. Based on the LAN probe plane to which the probe belongs, the forwarding device related to the LAN probe plane can be determined from among the multiple forwarding devices included in the first network node, thus identifying the second forwarding device. Similarly, based on the LAN probe plane to which the probe belongs, the forwarding device related to the LAN probe plane can be determined from among the multiple forwarding devices included in the current network node, thus identifying the first forwarding device; or, the forwarding device communicating with the probe can be directly identified to obtain the first forwarding device. Fault information determined based on this packet loss information can include communication link failures between the first and second forwarding devices.

[0061] For example, the probe data packets that prober s3 can receive may include multiple probe data packets with a destination address of X1.X2.0.n2 transmitted via forwarding device n2-1. These multiple probe data packets may include probe data packets from network node n1 and probe data packets from network node n3. If prober s3 determines statistically that a large number of probe data packets from network node n1 are lost, while probe data packets from network node n3 are not lost, then it can be determined that the communication between forwarding device n2-1 and network node n1 is faulty. This means that at least one of the following communication links is faulty: the communication link between forwarding device n2-1 and forwarding device n1-1, the communication link between forwarding device n2-1 and forwarding device n1-2, and the communication link between forwarding device n2-1 and forwarding device n1-3.

[0062] According to embodiments of this disclosure, the first packet loss information may also include packet loss in communication with multiple second probes, the multiple second probes being located at multiple second network nodes, and the values ​​of the second feature bits of the network addresses of the multiple second probes being consistent; wherein, the probes further determine a third forwarding device from multiple forwarding devices based on the second feature bits of the network addresses of the second probes, and determine that the fault information is represented as a fault of the third forwarding device.

[0063] Specifically, the probe can receive multiple data packets from other network nodes located on the same local area network probe plane. The probe can classify the received data packets according to their destination addresses and calculate the packet loss rate for each category. Each category can correspond to one other network node. The first packet loss information can be expressed as follows: if the packet loss rate of all categories of data packets is high, it can be determined that the probe is experiencing communication problems with all other network nodes. This indicates that the forwarding device forwarding the probe's data packets is malfunctioning, or that all downlink ports of the forwarding device are malfunctioning; in this case, the fault information can be expressed as a fault in the forwarding device.

[0064] For example, the probe data packets that probe machine s3 can receive may include multiple probe data packets with a destination address of X1.X2.0.n2 transmitted via forwarding device n2-1. These multiple probe data packets may include probe data packets from network node n1 and probe data packets from network node n3. If probe machine s3 determines statistically that both probe data packets from network node n1 and probe data packets from network node n3 experience a large number of packet losses, then it can be determined that the forwarding device n2-1, which is connected to probe machine s3, is faulty.

[0065] According to embodiments of this disclosure, by configuring the forwarding devices of each network node for routing, each forwarding device is used only to forward data packets of its local area network probe plane during fault detection. The probe can determine network faults based on the missing data packets received, thereby effectively improving the efficiency of fault location and reducing maintenance costs.

[0066] Figure 4 A schematic diagram of a backbone network system according to an embodiment of the present disclosure is shown.

[0067] like Figure 4 As shown, the backbone network system may include multiple regional network systems as described above, and these regional network systems communicate and connect with each other to form a decentralized, fully connected network topology.

[0068] Figure 5 A schematic diagram of a local area network system according to another embodiment of the present disclosure is shown.

[0069] like Figure 5As shown, each regional network system can have a super core data center, which may include multiple scheduling devices, such as scheduling device k1 and scheduling device k2. Within the regional network system, the scheduling devices can communicate with multiple forwarding devices included in each of the network nodes of the regional network system. Between regional network systems, the scheduling devices can communicate with scheduling devices in other regional network systems to achieve communication connections between multiple regional network systems.

[0070] For example, the regional network system 500 may include network node n1, network node n2, scheduling device k1, and scheduling device k2. Network node n1 may include probe s1, probe s2, forwarding device d1, and forwarding device d2. Probe s1 can communicate with forwarding device d1, and probe s2 can communicate with forwarding device d2. Network node n2 may include probe s3, probe s4, forwarding device d3, and forwarding device d4. Probe s3 can communicate with forwarding device d3, and probe s4 can communicate with forwarding device d4. Forwarding device d1 can communicate with forwarding devices d3 and d4 respectively, and forwarding device d2 can communicate with forwarding devices d3 and d4 respectively. Scheduling device k1 can communicate with forwarding devices d1, d2, d3, and d4 respectively, and scheduling device k2 can communicate with forwarding devices d1, d2, d3, and d4 respectively.

[0071] Similar to forwarding devices, scheduling devices can also be configured with routing policies. These policies allow traffic from specific network segments to pass through while filtering traffic from other network segments. This specific network segment can be a segment dedicated to communication between regional network systems, such as a second routing segment. In other words, the scheduling device's routing policy can be configured to allow traffic belonging to the corresponding second routing segment to pass through while filtering traffic from other network segments.

[0072] According to embodiments of this disclosure, the scheduling device is used to receive data packets and, if the destination address of the data packet belongs to the second routing network segment of the scheduling device, forward the data packet based on the destination address of the data packet.

[0073] For example, the second routing network segment of scheduling device k1 is 100.0.0.128 / 25. The data packets it can receive may include three data packets from the forwarding devices of the current area network system, with destination addresses of 100.0.0.129, 100.0.0.132, and 100.0.1.129, respectively. Since destination addresses 100.0.0.129 and 100.0.0.132 belong to the second routing network segment 100.0.0.128 / 25, while destination address 100.0.1.129 does not belong to the second routing network segment 100.0.0.128 / 25, scheduling device k1 can forward only the data packets with destination addresses of 100.0.0.129 and 100.0.0.132 to the scheduling devices of other area network systems. The data packets that the scheduling device can receive may also include two data packets from scheduling devices in other regional network systems, with destination addresses of 100.0.0.144 and 100.0.2.130, respectively. Since destination address 100.0.0.144 belongs to the second routing network segment 100.0.0.128 / 25, and destination address 100.0.2.130 does not belong to the second routing network segment 100.0.0.128 / 25, the scheduling device may forward only the data packet with destination address 100.0.0.144 to multiple forwarding devices in the current regional network system.

[0074] Forwarding devices can determine whether to forward a data packet to another forwarding device within the same regional network system, or to a scheduling device within that regional network system, based on the packet's destination address. Probes can also generate data packets for communication within the regional network system, as well as data packets for communication between regional network systems. In the backbone network system, each network node in each regional network system can be configured with a node identifier. The destination address of data packets for communication within the regional network system can be associated with the node identifier of each network node, while the destination address of data packets for communication between regional network systems can be associated with the node identifier of each network node and a fixed value.

[0075] For example, the destination address of data packets generated by the probe for transmission within the local area network system can be 100.0.plane identifier.node identifier, while the destination address of data packets generated for transmission between different local area network systems can be 100.0.plane identifier.node identifier + 128. If a forwarding device receives a data packet with a destination address of 100.0.plane identifier.node identifier, it can send the data packet to other forwarding devices in the current local area network system; if it receives a data packet with a destination address of 100.0.plane identifier.node identifier + 128, it can send the data packet to the scheduling device in the current local area network system so that the scheduling device can provide the data packet to other local area network systems.

[0076] In the backbone network, in addition to locating faults within the regional network system using the methods described above, the detector can also locate faults between regional network systems.

[0077] For example, the detector is used to collect data on packet loss, obtain second packet loss information, and obtain fault information based on the second packet loss information.

[0078] Figure 6 A schematic diagram of fault detection according to another embodiment of the present disclosure is shown.

[0079] like Figure 6 As shown, the backbone network system may include regional network system m1, regional network system m2, and regional network system m3. Regional network system m1 includes network nodes n1 and n2, regional network system m2 includes network nodes n3, n4, and n5, and regional network system m3 includes network node n6. Each of network nodes n1, n2, n3, n4, n5, and n6 may be configured with two probes and two forwarding devices.

[0080] In the regional network system m1, the first forwarding device m1n1-1 of network node n1 has a first routing network segment of X1.X2.0.n1 / 32, which is communicatively connected to probe s1; the first forwarding device m1n1-2 of network node n1 has a first routing network segment of X1.X2.1.n1 / 32, which is communicatively connected to probe s2. The first forwarding device m1n2-1 of network node n2 has a first routing network segment of X1.X2.0.n2 / 32, which is communicatively connected to probe s3; the first forwarding device m1n2-2 of network node n2 has a first routing network segment of X1.X2.1.n2 / 32, which is communicatively connected to probe s4.

[0081] In the regional network system m2, the first forwarding device m2n3-1 of network node n3 has a first routing network segment of X1.X2.0.n3 / 32, which is communicatively connected to probe s5; the first routing network segment of the second forwarding device m2n3-2 of network node n3 is X1.X2.1.n3 / 32, which is communicatively connected to probe s6. Similarly, the first forwarding device m2n4-1 of network node n4 has a first routing network segment of X1.X2.0.n4 / 32, which is communicatively connected to probe s7; the first routing network segment of the second forwarding device m2n4-2 of network node n4 is X1.X2.1.n4 / 32, which is communicatively connected to probe s8. The first forwarding device m2n5-1 of network node n5 has a first routing network segment of X1.X2.0.n5 / 32, which is connected to probe s9; the second forwarding device m2n5-2 of network node n5 has a first routing network segment of X1.X2.1.n5 / 32, which is connected to probe s10.

[0082] In the regional network system m3, the first routing network segment of the first forwarding device m3n6-1 of network node n6 is X1.X2.0.n6 / 32, which is connected to the probe s11; the first routing network segment of the second forwarding device m3n6-2 of network node n6 is X1.X2.1.n6 / 32, which is connected to the probe s12.

[0083] Regional network system m1 may further include scheduling devices k1 and k2; regional network system m2 may further include scheduling devices k3 and k4; and regional network system m3 may further include scheduling devices k5 and k6. The second routing network segment for scheduling devices k1, k3, and k5 can all be X1.X2.0.128 / 25; the second routing network segment for scheduling devices k2, k4, and k6 can all be X1.X2.1.128 / 25.

[0084] The following section will first use the example of network node n1 of regional network system m1 sending probe data packets to network node n3 of regional network system m2 to illustrate the sending and receiving process of probe data packets in the backbone network system.

[0085] The first probe s1 of network node n1 in the regional network system m1 can generate probe data packets with a destination address of X1.X2.0.(n3+128), belonging to the second routing network segment X1.X2.0.128 / 25. Probe s1 can provide this probe data packet to the forwarding device m1n1-1. During long-distance probing, the forwarding device m1n1-1 can learn the second routing network segments of various scheduling devices, as well as the first routing network segments of other forwarding devices and the network addresses of other probes. Since the first routing network segment or the network addresses of other probes are more granular than the second routing network segment, when sending data packets, the forwarding device will default to routing through the first routing network segment or the network addresses of other probes. Therefore, in the embodiments of this disclosure, when routing forwarding is required, the forwarding device m1n1-1 can mask the first routing network segments of other forwarding devices or the network addresses of other probes to facilitate routing the probe data packet through the second routing network segment. That is, the forwarding device m1n1-1 can forward the probe data packet to the scheduling device k1 and the scheduling device k2 respectively.

[0086] After receiving the probe data packet, scheduling device k1 or scheduling device k2 can determine whether to forward it. Specifically, scheduling device k1 or scheduling device k2 can determine whether the destination address of the probe data packet belongs to the second routing network segment of the scheduling device. The destination address of the probe data packet is X1.X2.0.(n3+128), and the second routing network segment it belongs to is X1.X2.0.128 / 25. This second routing network segment belongs to the second routing network segment of scheduling device k1. Therefore, scheduling device k1 can forward the probe data packet, and scheduling device k2 can filter the probe data packet.

[0087] The scheduling device k1 can also be used to determine the scheduling target based on the first characteristic bit field of the destination address of the probe data packet, namely the values ​​of bits 25 to 32. Based on the value of the first characteristic bit field (n3+128) of the destination address of the probe data packet, the scheduling target can be determined to be network node n3. Since network node n3 belongs to the regional network system m2, the scheduling device k1 can forward the probe data packet to the scheduling device related to the second routing network segment in the regional network system m2, namely the scheduling device k3.

[0088] The scheduling device k3 can provide the probe data packet to multiple forwarding devices included in the regional network system m2, so that the probe data packet can be forwarded to the network node n3. Specifically, the scheduling device k3 can send the probe data packet to forwarding devices m2n3-1, m2n3-2, m2n4-1, m2n4-2, m2n5-1, and m2n5-2 respectively. After receiving the probe data packet, each forwarding device can directly forward the probe data packet to the probe machine it is communicating with. That is, both forwarding devices m2n3-1 and m2n3-2 included in network node n3 can receive the probe data packet and send it to the corresponding probe machine. For example, forwarding device m2n3-1 sends the probe data packet to probe machine s5, and forwarding device m2n3-2 sends the probe data packet to probe machine s6, thus completing the transmission of the probe data packet to network node n3.

[0089] Under fault-free conditions, each probe can receive probe data packets forwarded by the forwarding devices in the network nodes of each regional network system and the scheduling devices in each regional network system. Therefore, backbone network fault detection can be performed based on unreceived probe data packets.

[0090] Taking network node n1 as an example, if we only consider fault detection between regional network systems, that is, if all forwarding devices of network node n1 can work normally, then the first probe s1 of network node n1 can receive probe data packets from scheduling device k3 and from scheduling device k4. Furthermore, the probe s1 can perform fault detection based on the packet loss of the received probe data packets.

[0091] According to an embodiment of this disclosure, the second packet loss information includes that the destination addresses of the received data packets do not belong to the second routing network segment of the first scheduling device, and the second packet loss information of the third detector indicates no packet loss. The third detector is located in the regional network system. The detector is used to determine the fifth forwarding device that is connected to the detector from multiple forwarding devices, and to determine that the fault information indicates a communication link failure between the first scheduling device and the fifth forwarding device.

[0092] For example, for probe s1, which communicates with forwarding device m1n1-1, scheduling devices that directly communicate with probe s1 may include scheduling devices k1 and k2. A third probe may also be located in the regional network system m1, such as probe s2 (communicating with forwarding device m1n1-2), probe s3 (communicating with forwarding device m1n2-1), and probe s4 (communicating with forwarding device m1n2-2). The second packet loss information for probe s1 can be expressed as follows: no packet loss occurred in the data packets transmitted within the regional network system that should have been received, no packet loss occurred in the data packets forwarded by scheduling device k2, but no data packets forwarded by scheduling device k1 were received. In this case, it can be assumed that forwarding device m1n1-1 itself is fault-free, scheduling device k2 itself is fault-free, and the communication link between forwarding device m1n1-1 and scheduling device k2 is also fault-free. In this case, the fault can be located at a fault in scheduling device k1 itself, or a fault in the communication link between scheduling device k1 and forwarding device m1n1-1. Subsequently, the detector s1 can also obtain the second packet loss information from the third detector. If the second packet loss information from the third detector indicates that there is no packet loss in the data packets transmitted within the regional network system that should be received, and there is no packet loss in the data packets forwarded by the scheduling devices k1 and k2, then it can be assumed that the scheduling device k1 itself is also fault-free. Therefore, the fault location can be determined to be a communication link failure between the forwarding device m1n1-1 and the scheduling device k1.

[0093] According to an embodiment of this disclosure, the second packet loss information includes destination addresses of received data packets that do not belong to the second routing network segment of the second scheduling device, and the second packet loss information of multiple fourth detectors also includes destination addresses of received data packets that do not belong to the second routing network segment of the second scheduling device, wherein the multiple fourth detectors are located in the regional network system; wherein, the detectors are further used to determine that the fault information indicates a fault of the second scheduling device.

[0094] For example, for probe s1, which communicates with forwarding device m1n1-1, scheduling devices that directly communicate with probe s1 may include scheduling devices k1 and k2. Multiple fourth probes may include all other probes also located in the regional network system m1, such as probe s2, which communicates with forwarding device m1n1-2; probe s3, which communicates with forwarding device m1n2-1; and probe s4, which communicates with forwarding device m1n2-2. The second packet loss information for probe s1 can be expressed as follows: no packet loss occurred in the data packets transmitted within the regional network system that should have been received; no packet loss occurred in the data packets forwarded by scheduling device k2; but no data packets were received forwarded by scheduling device k1. In this case, it can be assumed that forwarding device m1n1-1 itself is fault-free, scheduling device k2 itself is fault-free, and the communication link between forwarding device m1n1-1 and scheduling device k2 is also fault-free. In this case, the fault can be located at a fault in scheduling device k1 itself, or a fault in the communication link between scheduling device k1 and forwarding device m1n1-1. Subsequently, the probe can also acquire the second packet loss information of multiple fourth probes. If the second packet loss information of multiple fourth probes indicates that there is no packet loss in the data packets transmitted within the regional network system that should be received, no packet loss in the data packets forwarded by scheduling device k2, and no packet loss in the data packets forwarded by scheduling device k1, that is, the second packet loss information of the fourth probe is consistent with that of probe s1, then it can be considered that the fault in the backbone network system is not limited to the communication link fault between a single forwarding device and the scheduling device. Therefore, the fault can be located as a fault of scheduling device k1.

[0095] According to an embodiment of this disclosure, the second packet loss information includes packet loss in communication behavior between multiple sixth probes, all of which are located in a third regional network system; wherein, the probes further determine a third scheduling device from multiple scheduling devices based on the network address of the sixth probes, and determine a fourth scheduling device from multiple scheduling devices included in the third regional network system, and determine the fault information as a communication link failure between the third scheduling device and the fourth scheduling device.

[0096] For example, for probe s1, which communicates with forwarding device m1n1-1, multiple sixth probes may include probes located in regional network system m2, such as probe s5 (communicating with forwarding device m2n3-1), probe s6 (communicating with forwarding device m2n3-2), probe s7 (communicating with forwarding device m2n4-1), probe s8 (communicating with forwarding device m2n4-2), probe s9 (communicating with forwarding device m2n5-1), and probe s10 (communicating with forwarding device m2n5-2). Therefore, scheduling device k1 can forward probe data packets normally, but scheduling device k1 cannot receive probe data packets from regional network system m2. Furthermore, the probes can normally receive probe data packets from regional network system m3. In this case, the fault can be located at scheduling device k3 in regional network system m2, or at the communication link between scheduling device k3 and scheduling device k1. If the regional network system m2 reports that the scheduling device k3 is fault-free, then the fault information can be determined to be a communication link failure between scheduling device k3 and scheduling device k1.

[0097] According to embodiments of this disclosure, by configuring routes for each forwarding device and each scheduling device, each forwarding device or scheduling device is only used to forward data packets of its respective regional network probe plane during fault detection. The probe can determine network faults based on the missing data packets received, thereby effectively improving the efficiency of fault location and reducing operation and maintenance costs.

[0098] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0099] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A regional network system, comprising: a plurality of network nodes, the network nodes comprising a plurality of probe machines and a plurality of forwarding devices, the plurality of probe machines being communicatively connected with the plurality of forwarding devices respectively, the forwarding devices being communicatively connected with a plurality of forwarding devices comprised by a plurality of other network nodes respectively; wherein the forwarding devices are configured to receive a data packet and forward the data packet to a probe machine determined based on a destination address of the data packet, in a case that the destination address of the data packet belongs to a first routing network segment of the forwarding device; the probe machines are configured to obtain fault information based on the received data packet. 2.The regional network system of claim 1, wherein the probe machines are configured to obtain first packet loss information by counting packet loss conditions of the data packet, and obtain the fault information based on the first packet loss information.

3. The area network system according to claim 2, wherein the first packet loss information comprises packet loss existing in communication behaviors between a first probe machine and a first network node. wherein the probe machines are configured to determine a first forwarding device from the plurality of forwarding devices based on a network address of the first probe machine, and determine a second forwarding device from a plurality of forwarding devices comprised by the first network node, and determine the fault information to represent a communication link fault between the first forwarding device and the second forwarding device.

4. The area network system of claim 2, wherein, the first packet loss information comprises packet loss existing in communication behaviors between a plurality of second probe machines, the plurality of second probe machines being located in a plurality of second network nodes respectively, and a second characteristic bit segment of a network address of each of the plurality of second probe machines having a same value, the second characteristic bit segment being a continuous binary bit range. wherein the probe machines are further configured to determine a third forwarding device from the plurality of forwarding devices based on the second characteristic bit segment of the network address of the second probe machine, and determine the fault information to represent a fault of the third forwarding device.

5. The area network system of claim 1, wherein, the destination address of the data packet belonging to the first routing network segment of the forwarding device comprises the destination address of the data packet being identical to a network address comprised by the first routing network segment of the forwarding device.

6. The area network system of claim 1, wherein, a first characteristic bit segment of the destination address of the data packet is used to indicate a network node to which the forwarding device belongs, the first characteristic bit segment being a continuous binary bit range.

7. The area network system of claim 1, wherein, the probe machines are further configured to generate a data packet and send the data packet to the forwarding device; the forwarding devices are configured to send the data packet to a plurality of forwarding devices comprised by the other network nodes respectively. 8.A backbone network system, comprising a plurality of regional network systems of any one of claims 1-7.

9. The backbone system of claim 8, wherein, the regional network system further comprises a plurality of scheduling devices, the scheduling devices being communicatively connected with a plurality of forwarding devices comprised by a plurality of network nodes of the regional network system respectively; wherein the scheduling devices are configured to receive a data packet and perform forwarding processing on the data packet based on a destination address of the data packet, in a case that the destination address of the data packet belongs to a second routing network segment of the scheduling device.

10. The backbone system of claim 9, wherein, the probe machines are configured to obtain second packet loss information by counting packet loss conditions of the data packet, and obtain fault information based on the second packet loss information.

11. The backbone system of claim 10, wherein, The second packet loss information includes that the destination addresses of the received data packets do not belong to a second routing network segment of a second scheduling device, and second packet loss information of a third probe machine indicates no packet loss, and the third probe machine is located in the regional network system; The probe machine is configured to determine a fifth forwarding device in communication connection with the probe machine from a plurality of forwarding devices, and determine that the fault information indicates a communication link fault between the first scheduling device and the fifth forwarding device.

12. The backbone system of claim 10, wherein, The second packet loss information includes that the destination addresses of the received data packets do not belong to a second routing network segment of a second scheduling device, and second packet loss information of a plurality of fourth probe machines includes that the destination addresses of the received data packets do not belong to the second routing network segment of the second scheduling device, and the plurality of fourth probe machines are located in the regional network system. The probe machine is further configured to determine that the fault information indicates a fault of the second scheduling device.

13. The backbone system of claim 10, wherein, The second packet loss information includes packet loss in communication behaviors between a plurality of sixth probe machines, and the plurality of sixth probe machines are located in a third regional network system. The probe machine is further configured to determine a third scheduling device from the plurality of scheduling devices based on network addresses of the sixth probe machines, and determine a fourth scheduling device from a plurality of scheduling devices included in the third regional network system, and determine that the fault information indicates a communication link fault between the third scheduling device and the fourth scheduling device.

Images