SD-WAN whole network bearer evaluation driven routing and elastic linkage scheduling method, system and device, and medium

Abstract

The invention discloses a routing and elastic linkage scheduling method, system and device driven by SD-WAN whole network bearing evaluation and a medium, and relates to the field of software defined wide area networks, and the method comprises the steps: obtaining the topological information and state data of a backbone network; calculating the load saturation of the whole network based on the topological information and the state data, and marking overload nodes; determining a linkage scheduling scene according to the whole network bearing saturation and the overload node in combination with a pre-judgment result of a routing effect, and generating an execution instruction; wherein the execution instruction comprises a routing instruction, an elastic instruction and a capacity reduction instruction; and executing traffic grooming according to the routing instruction, synchronously creating a new elastic node and an associated elastic link according to the elastic instruction, and executing a capacity reduction operation according to the capacity reduction instruction. According to the invention, the cooperative scheduling of routing and elasticity can be realized, and the adaptation capability of the SD-WAN backbone network to the tidal fluctuation flow is improved.

Description

Technical Field

[0001] This application relates to the field of software-defined wide area networks (SD-WAN), and in particular to a method, system, device, and medium for routing and flexible linkage scheduling driven by SD-WAN whole-network bearer assessment. Background Technology

[0002] As enterprises continue to expand their cross-regional businesses, SD-WAN (Software-Defined Wide Area Network) backbones need to carry diverse traffic, including branch interconnections and cloud resource access. The network traffic they carry exhibits tidal fluctuations—that is, traffic changes periodically over time: peak hours are from 9:00 AM to 6:00 PM on weekdays, with traffic reaching its highest point (several times the daily average); while nighttime and weekends are off-peak periods. Against this backdrop, the stable operation of the SD-WAN backbone relies on two core functions: first, routing, which optimizes and manages existing traffic; and second, elastic scaling, which dynamically adds or removes nodes or link resources to exceed network capacity limits and adapt to periodic traffic fluctuations. However, in existing SD-WAN technologies, routing and elastic scaling functions are mostly independently designed modules, lacking an effective coordination mechanism. This leads to numerous defects in network scheduling, specifically in the following aspects: (1) Existing elastic scheduling mechanisms generally rely on the local utilization indicators of a single node or a single link to trigger expansion (such as when the node CPU utilization is ≥80% or the link utilization is ≥90%). They have not established a comprehensive evaluation system for the overall network carrying status, which can easily lead to misjudgments such as "local overload but the whole network still has redundant resources" or "the whole network is saturated but local indicators have not reached the trigger threshold", resulting in a lack of scientificity and rationality in elastic scheduling.

[0003] (2) The elastic operation of nodes and links is disconnected and lacks coordination. In the node expansion scenario, the bandwidth of the link associated with the new node is not adjusted synchronously, which makes the new node unable to fully play its carrying role and the link becomes a new performance bottleneck; while in the node recycling scenario, the resources of the associated elastic link are not released synchronously, resulting in idle waste of network resources and ultimately leading to a mismatch between the carrying capacity of nodes and links.

[0004] (3) There is no collaborative decision-making mechanism between routing and elastic scaling, and the two operate independently. When routing optimization cannot meet the traffic diversion requirements, the elastic scaling function is not triggered in time to expand resources, which may cause service interruption. After the elastic scaling operation is completed, the routing strategy is not dynamically adjusted according to the newly added or reclaimed resources, resulting in low utilization of newly added resources (usually less than 30%), or congestion of the remaining links after resource reclamation due to excessive traffic concentration, resulting in poor overall scheduling efficiency. Specifically, the existing technologies have clear shortcomings: In one type of existing technology, the elastic scaling function is triggered only based on the CPU utilization or bandwidth utilization of a single node. The link bandwidth is mostly statically configured and requires manual intervention for adjustment. The routing function only refers to the link quality (such as latency and packet loss rate) to select the path and has no data interaction with the elastic scaling module. Its core problem is the lack of network-wide load capacity assessment capability. The nodes and links are disconnected from elasticity, and the independent operation of routing and elasticity leads to low resource utilization efficiency. In another type of existing technology, although it attempts to achieve link elasticity, it still uses the utilization of a single link as the sole trigger condition and is not related to the elasticity of nodes. Moreover, the routing strategy is fixed (such as the default shortest path) and does not dynamically adjust with changes in elastic resources, which easily leads to problems such as over-expansion, low utilization of new resources, and link congestion after scaling down. In summary, existing SD-WAN technologies have significant shortcomings in areas such as overall network load status assessment, node-link elastic coordination, routing, and elastic linkage scheduling, making it difficult to adapt to the diverse traffic load requirements under tidal fluctuation characteristics. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, device, and medium for SD-WAN whole-network bearer assessment-driven routing and elastic linkage scheduling, which can realize the coordinated scheduling of routing and elasticity, and improve the adaptability of SD-WAN backbone network to tidal fluctuation traffic.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a routing and elastic linkage scheduling method driven by SD-WAN whole-network bearer evaluation, including: Obtain the topology information and status data of the backbone network; Based on the topology information and the status data, calculate the overall network load saturation and mark overloaded nodes; Based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, a coordinated scheduling scenario is determined, and an execution instruction is generated; wherein, the execution instruction includes a routing instruction, an elastic instruction, and a scaling-down instruction; Traffic diversion is performed according to the routing instructions, new elastic nodes and associated elastic links are created synchronously according to the elastic instructions, and a scaling-down operation is performed according to the scaling-down instructions.

[0007] Secondly, this application provides an SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling system, including: The data acquisition module is used to acquire the topology information and status data of the backbone network; The network load saturation calculation module allows users to calculate the network load saturation based on the topology information and the status data, and to mark overloaded nodes. The execution instruction generation module is used to determine the coordinated scheduling scenario based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, and generate execution instructions; wherein, the execution instructions include routing instructions, elastic instructions and scaling down instructions; The instruction execution module is used to perform traffic diversion according to the routing instruction, synchronously create new elastic nodes and associated elastic links according to the elastic instruction, and perform scaling down operation according to the scaling down instruction.

[0008] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described SD-WAN whole-network bearer assessment-driven routing and flexible linkage scheduling method.

[0009] Fourthly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described SD-WAN whole-network bearer evaluation-driven routing and flexible linkage scheduling method.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: (1) Through the core process of “acquiring topology and status data → calculating the network load saturation → marking overloaded nodes”, a network-wide load assessment system was established. It is no longer limited to local indicators, but rather quantifies the matching relationship between the actual load of the entire network and the load limit from a global perspective; accurately identifies the real load pressure of the entire network and reduces the business risks caused by hidden overload.

[0011] (2) Cooperative elastic scheduling of elastic nodes and associated elastic links binds node elasticity and associated elastic link elasticity into a unified operation, avoiding the load imbalance caused by the independent execution of the two; (3) By making a joint decision based on “the overall network load saturation + overload nodes + prediction of routing effect”, the blind expansion and ineffective routing are effectively reduced, and the accuracy of expansion decision is greatly improved. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating the steps of a routing and flexible linkage scheduling method driven by SD-WAN whole-network bearer evaluation, provided in an embodiment of this application. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0016] In one exemplary embodiment, such as Figure 1 As shown, a routing and elastic linkage scheduling method driven by SD-WAN whole-network bearer evaluation is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is described using a server as an example, including the following steps S1 to S5. Wherein: S1: Obtain the topology information and status data of the backbone network.

[0017] S2: Based on the topology information and the status data, calculate the overall network load saturation and mark overloaded nodes.

[0018] S3: Based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, determine the linkage scheduling scenario and generate execution instructions; wherein, the execution instructions include routing instructions, elastic instructions and scaling down instructions.

[0019] S4: Perform traffic diversion according to the routing instruction, synchronously create new elastic nodes and associated elastic links according to the elastic instruction, and perform a scaling-down operation according to the scaling-down instruction.

[0020] This application achieves coordinated scheduling of routing and flexibility by implementing a closed-loop process of the above steps: "data collection - network-wide evaluation - coordinated decision-making - flexible execution - route optimization".

[0021] In one specific embodiment, step S1 includes: Topology information and status data are obtained through the controller's northbound interface (such as a standardized API). The SD-WAN controller maintains topology information (PoP node list: node ID, node configured bandwidth; link list: link ID, end nodes, link bandwidth threshold) and status data (actual bandwidth used by nodes, actual bandwidth used by links).

[0022] The acquired data is formatted (e.g., converted into structured tables) and filtered for validity (removing offline node / link data).

[0023] In one specific embodiment, step S2 includes: S21: Calculate basic indicators based on the topology information and the status data; the basic indicators include the remaining capacity of nodes, the remaining capacity of links, and the actual total traffic of the entire network.

[0024] Node remaining capacity = Node configured bandwidth - Node actual bandwidth usage; Remaining link capacity = Link bandwidth threshold - Actual link bandwidth used; The total actual network traffic = the sum of the actual bandwidth used by all nodes.

[0025] S22: Based on the remaining capacity of the nodes and the remaining capacity of the links, the maximum flow algorithm is used to calculate the maximum capacity of the entire network.

[0026] Construct a weighted graph using topology information (node ​​weight = node remaining capacity, edge weight = link remaining capacity) and solve for the maximum capacity of the entire network.

[0027] S23: Calculate the network capacity saturation based on the actual total network traffic and the maximum network capacity.

[0028] Network saturation = (Actual total network traffic / Maximum network capacity) × 100%; Nodes (or links) with utilization ≥70% are marked as overloaded nodes (or overloaded links).

[0029] In a specific embodiment, the prediction of routing effectiveness is as follows: the excess traffic of the overloaded node is migrated according to the proportion of the remaining carrying capacity of each alternative path; the utilization rate decrease of the overloaded node and the utilization rate increase of the alternative path are calculated after migration; if the utilization rate decrease drops below a first local threshold and the utilization rate increase does not exceed a second local threshold, the routing is determined to be effective; otherwise, the routing is determined to be invalid; wherein, the first local threshold < the second local threshold.

[0030] Specifically, simulate traffic migration: traverse alternative paths (avoiding overloaded links), allocate traffic according to the remaining capacity of the path, and calculate the "decrease in utilization of overloaded nodes (or overloaded links)" and the "increase in utilization of alternative paths" after migration; determine the routing effect: if the utilization of overloaded nodes (or overloaded links) drops to <70%, and the increase in utilization of alternative paths is ≤75%, then the routing is "effective"; otherwise, the routing is "ineffective".

[0031] In one specific embodiment, the coordinated scheduling scenarios include route selection optimization only, route selection and elastic expansion coordination, direct elastic expansion, and reduction and route selection coordination.

[0032] The determination of the linkage scheduling scenario (based on three weighted thresholds) is as follows: (1) When the overall network load saturation is lower than the first global threshold, and there are overloaded nodes with local utilization exceeding the first local threshold, and routing is effective, the linkage scheduling scenario is determined to be routing optimization only; (2) When the overall network load saturation is between the first global threshold and the second global threshold, and there are overloaded nodes with local utilization exceeding the second local threshold, and routing is ineffective, the linkage scheduling scenario is determined to be routing and elastic expansion linkage; (3) When the overall network load saturation exceeds the second global threshold, and there are overloaded nodes with local utilization exceeding the third local threshold, and routing is ineffective, the linkage scheduling scenario is determined to be direct elastic expansion; (4) When the overall network load saturation is lower than the third global threshold, and there are elastic resources with continuous low load, and routing is effective, the linkage scheduling scenario is determined to be shrinkage and routing linkage scenario; wherein, the third global threshold < the first global threshold < the second global threshold, and the first local threshold < the second local threshold < the third local threshold.

[0033] The details are shown in Table 1: Table 1

[0034] Route selection instructions: including optimal path and traffic allocation ratio (for the "Route Selection Only" and "Route Selection + Elasticity (Route Selection Phase)" scenarios in Table 1); Elastic commands: include the number / location of elastic nodes and the associated elastic link specifications (for the "Route Selection + Elastic (Elastic Phase)" and "Direct Elastic" scenarios in Table 1); Shrink-down instructions: include nodes / links to be recycled and dual-dimensional verification standards (for the "shrink-down + routing" scenario in Table 1).

[0035] In one specific embodiment, step S5 specifically includes: (1) Expansion operation: 1) After receiving the elastic command, synchronously call the elastic node interface to create a new elastic node and call the link interface to open the associated elastic link (the node and link bandwidth specifications match, such as a node with 10G bandwidth corresponding to a link with 10G bandwidth). 2) Ensure that the time difference between node and link readiness is ≤10 seconds to avoid load disconnection.

[0036] (2) Volume reduction operation: 1) Upon receiving the scaling down instruction, first perform route selection and migration of the traffic of the node to be recycled (based on the route selection instruction). 2) Perform two-dimensional verification: Node verification: The total remaining capacity of the remaining nodes is greater than or equal to the traffic to be reclaimed, and the utilization rate of a single node is less than 70%; Link verification: The total remaining capacity of the remaining links is greater than or equal to the traffic to be migrated, and the utilization rate of a single link is less than 70%. 3) After the verification is passed, the elastic nodes and associated elastic links are synchronously recycled.

[0037] Output: Elastic execution report (success / failure indicator, resource status).

[0038] (3) Route selection and traffic optimization: 1) Route selection execution: Based on the priority of "low latency + high bandwidth redundancy" (latency weight 60%, remaining bandwidth weight 40%), the routing table is updated through the routing interface to divert overloaded traffic to alternative paths; 2) Post-elastic optimization: After elastic resources are launched, the routing weights are dynamically adjusted based on the proportion of the remaining capacity of the new elastic nodes and associated elastic links to the remaining capacity of the entire network (new node / link initial weight = its remaining capacity / the remaining capacity of the entire network) to balance the load of new and old resources. 3) Monitor traffic migration: Ensure that the interruption time of long connection migration is ≤5 seconds.

[0039] Output: Route selection execution report (migration completion rate, post-migration utilization rate).

[0040] In a specific embodiment, the method further includes: summarizing the elastic execution report and the routing execution report to support closed-loop decision-making (e.g., if the load is still overloaded after routing, trigger elasticity; if the load is uneven after elasticity, re-route); triggering alarms for abnormal situations (e.g., routing not taking effect, elasticity timeout), recording the operation trajectory (retained for ≥90 days), and then outputting a network status report (current load, resource status, and abnormality identifier). The network status report serves as the basis for the next round of data collection to achieve continuous optimization.

[0041] In a specific embodiment, for scenario 1: route selection optimization only (no flexible operation): Initial state: Overall network capacity saturation 65%, PoP1-PoP2 link utilization 76% (overloaded), alternative path PoP1-PoP3-PoP2 utilization 52%.

[0042] Execution steps: Data collected: The actual usage of the PoP1-PoP2 link is 7.6G (threshold 10G), and the remaining capacity of the alternative path is 4.8G.

[0043] Assessment: The overall network load saturation is 65%, and PoP1-PoP2 is marked as an overloaded link.

[0044] Decision: After migrating 2G traffic, the original link utilization rate is 56%, and the alternative path utilization rate is 68% (route selection is effective). Output the route selection command.

[0045] Execution: Migrated 2G traffic to the alternative path. The routing report showed "Migration complete, no overload". Monitoring: Status normal, proceed to the next cycle.

[0046] Result: No new elastic resources were added, and the overall network utilization rate increased by 10%.

[0047] In a specific embodiment, for scenario 2: routing + elastic scaling: Initial state: The overall network load saturation is 78%, the PoP4 utilization rate is 83% (overloaded), all associated links are fully loaded, and routing prediction is ineffective.

[0048] Execution steps: Decision: In the "routing + elasticity" scenario, first output the routing instruction (migrate some traffic, utilization drops to 77% but is still overloaded), then output the elasticity instruction (add PoP5 and associated links). Execution: Synchronously create PoP5 (10G bandwidth) and PoP5-PoP2 links (10G bandwidth), ready within 30 seconds; Optimization: Flexible routing was implemented, diverting 30% of the traffic from PoP4 to PoP5, reducing PoP4 utilization to 58%. Results: The overall network saturation was 62%, and the utilization rate of new resources was 65%.

[0049] This application has the following advantages: Improved network stability: By assessing the overall network load, hidden overloads are avoided, and the service interruption rate is reduced; Resource utilization optimization: Node-link collaborative elasticity + dynamic routing improve resource utilization; Enhanced scheduling accuracy: Three-dimensional threshold linkage reduces the rate of blind, elastic capacity expansion; Scaling down ensures security: Dual-dimensional verification prevents overload and reduces the risk of link congestion; Increased automation: closed-loop scheduling throughout the entire process reduces manual intervention and lowers operation and maintenance costs.

[0050] Based on the same inventive concept, this application also provides a routing and flexible linkage scheduling system for implementing the SD-WAN whole-network bearer evaluation-driven system described above. The solution provided by this system is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more embodiments of the SD-WAN whole-network bearer evaluation-driven routing and flexible linkage scheduling system provided below can be found in the limitations of the SD-WAN whole-network bearer evaluation-driven routing and flexible linkage scheduling method described above, and will not be repeated here.

[0051] In one exemplary embodiment, an SD-WAN whole-network bearer assessment-driven routing and elastic linkage scheduling system is provided, including the following modules.

[0052] The data acquisition module is used to acquire the topology information and status data of the backbone network.

[0053] The network load saturation calculation module allows users to calculate the network load saturation based on the topology information and the status data, and to mark overloaded nodes.

[0054] The execution instruction generation module is used to determine the coordinated scheduling scenario based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, and generate execution instructions; wherein, the execution instructions include routing instructions, elastic instructions, and scaling down instructions.

[0055] The instruction execution module is used to perform traffic diversion according to the routing instruction, synchronously create new elastic nodes and associated elastic links according to the elastic instruction, and perform scaling down operation according to the scaling down instruction.

[0056] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments. The computer device may be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, memory, and I / O are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device stores data to be processed. The I / O interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal via a network connection. When the computer program is executed by the processor, it implements the steps in the above-described method embodiments.

[0057] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0058] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0059] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0060] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0061] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A routing and elastic linkage scheduling method driven by SD-WAN whole-network bearer evaluation, characterized in that, include: Obtain the topology information and status data of the backbone network; Based on the topology information and the status data, calculate the overall network load saturation and mark overloaded nodes; Based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, a coordinated scheduling scenario is determined, and an execution instruction is generated; wherein, the execution instruction includes a routing instruction, an elastic instruction, and a scaling-down instruction; Traffic diversion is performed according to the routing instructions, new elastic nodes and associated elastic links are created synchronously according to the elastic instructions, and a scaling-down operation is performed according to the scaling-down instructions.

2. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 1, characterized in that, Based on the topology information and the status data, the overall network capacity saturation is calculated, specifically including: Based on the topology information and the status data, basic indicators are calculated; these basic indicators include node remaining capacity, link remaining capacity, and total actual network traffic. Based on the remaining capacity of the nodes and the remaining capacity of the links, the maximum flow algorithm is used to calculate the maximum capacity of the entire network. The network capacity saturation is calculated based on the actual total network traffic and the maximum network capacity.

3. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 1, characterized in that, The prediction results include valid route selection and invalid route selection; Specifically, the prediction of route selection effectiveness includes: The excess traffic of the overloaded node is migrated according to the proportion of the remaining carrying capacity of each alternative path; Calculate the decrease in utilization of the overloaded node after migration and the increase in utilization of the alternative path; If the utilization rate decreases to below the first local threshold and the utilization rate increase does not exceed the second local threshold, the routing is deemed valid; otherwise, the routing is deemed invalid. Wherein, the first local threshold is less than the second local threshold.

4. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 3, characterized in that, The coordinated scheduling scenarios include route selection optimization only, route selection and elastic expansion coordination, direct elastic expansion, and reduction and route selection coordination. Specifically, based on the overall network load saturation and the overloaded nodes, combined with the prediction results of routing effectiveness, the coordinated scheduling scenario is determined, including: When the overall network load saturation is lower than the first global threshold, and there are overloaded nodes with local utilization exceeding the first local threshold, and the routing is effective, the linkage scheduling scenario is determined to be the routing-only optimization. When the overall network load saturation is between the first global threshold and the second global threshold, and there are overloaded nodes with local utilization exceeding the second local threshold, and routing is invalid, the linkage scheduling scenario is determined to be a linkage between routing and elastic expansion. When the overall network saturation exceeds the second global threshold, and there are overloaded nodes with local utilization exceeding the third local threshold, and routing is invalid, the coordinated scheduling scenario is determined to be direct elastic expansion. When the overall network load saturation is lower than the third global threshold, and there are elastic resources with continuous low load, and the routing is effective, the linkage scheduling scenario is determined to be a scaling down and routing linkage scenario. Wherein, the third global threshold < the first global threshold < the second global threshold, and the first local threshold < the second local threshold < the third local threshold.

5. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 1, characterized in that, The scaling-down operation involves simultaneously reclaiming existing elastic nodes and associated elastic links.

6. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 1, characterized in that, Before performing the scaling-down operation, the process also includes: migrating traffic on the node to be recycled to other nodes based on routing instructions, and performing a two-dimensional bearer verification; the two-dimensional bearer verification includes: node bearer verification and link bearer verification.

7. The SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method according to claim 1, characterized in that, After performing elastic expansion, the following also applies: dynamically adjusting the routing weight based on the proportion of the remaining bearers of the new elastic nodes and associated elastic links to the total remaining bearers of the network.

8. A routing and flexible linkage scheduling system driven by SD-WAN whole-network bearer evaluation, characterized in that, include: The data acquisition module is used to acquire the topology information and status data of the backbone network; The network load saturation calculation module allows users to calculate the network load saturation based on the topology information and the status data, and to mark overloaded nodes. The execution instruction generation module is used to determine the coordinated scheduling scenario based on the overall network load saturation and the overloaded nodes, combined with the prediction results of the routing effect, and generate execution instructions; wherein, the execution instructions include routing instructions, elastic instructions and scaling down instructions; The instruction execution module is used to perform traffic diversion according to the routing instruction, synchronously create new elastic nodes and associated elastic links according to the elastic instruction, and perform scaling down operation according to the scaling down instruction.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the SD-WAN whole-network bearer assessment-driven routing and elastic linkage scheduling method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the SD-WAN whole-network bearer evaluation-driven routing and elastic linkage scheduling method as described in any one of claims 1-7.

Images