Network tunnel resource scheduling method and system, electronic equipment and storage medium

Abstract

The invention discloses a network tunnel resource scheduling method and system, and the method comprises the steps: obtaining tunnel data of a network tunnel, and the tunnel data comprises a service priority and a flow state; generating a control strategy according to the tunnel data; a target tunnel is forbidden according to the control strategy, and the target tunnel is a low-priority network tunnel determined based on the service priority or an idle network tunnel determined based on the traffic state; and after the high-priority service determined based on the service priority is ended, restarting the target tunnel according to the network state of the network tunnel. According to the method and the device, bandwidth resources are allocated according to needs, the transmission quality of high-priority services is guaranteed preferentially, meanwhile, tedious operation of manual intervention is avoided, and the operation and maintenance cost and potential risks of configuration errors are reduced.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of network, and particularly relates to a network tunnel resource scheduling method, a network tunnel resource scheduling system, an electronic device and a readable storage medium. BACKGROUND

[0002] In the field of high-definition video communication network, network tunnels are widely used in government, medical, education and other scenarios to support video conferencing, real-time monitoring and remote command and other businesses. With the increase in the number of terminals and the types of businesses, network bandwidth resources are increasingly strained.

[0003] The existing tunnel configuration method usually adopts fixed bandwidth allocation, which leads to the problem that high-priority businesses such as video conferencing or emergency command are easily affected by bandwidth occupation of low-priority businesses or unused tunnels under high load, resulting in transmission delay, lag or even interruption. In addition, manual adjustment of tunnel configuration is complex and time-consuming, which is difficult to meet the real-time business demand, increases the operation and maintenance cost and the potential risk of configuration errors. SUMMARY

[0004] The purpose of the embodiments of the application is to provide a network tunnel resource scheduling method, a network tunnel resource scheduling system, an electronic device and a readable storage medium, which can solve the problems that high-priority businesses are affected by bandwidth occupation of low-priority or unused tunnels and manual adjustment of configuration is complex and time-consuming in the prior art.

[0005] In order to solve the above technical problems, the application is implemented as follows: In a first aspect, the embodiments of the application provide a network tunnel resource scheduling method, which comprises: obtaining tunnel data of a network tunnel, wherein the tunnel data comprises business priority and traffic state; generating a control strategy according to the tunnel data; disabling a target tunnel according to the control strategy, wherein the target tunnel is a low-priority network tunnel determined based on the business priority, or an idle network tunnel determined based on the traffic state; re-enabling the target tunnel according to the network state of the network tunnel after a high-priority business determined based on the business priority ends.

[0006] Optionally, the generating a control strategy according to the tunnel data comprises: inputting the tunnel data into a time series prediction model and a strategy generation model deployed in an edge computing node or a central cloud; analyzing the tunnel data by the time series prediction model to obtain a bandwidth load prediction result of a future time period; The control strategy is generated by the strategy generation model according to the bandwidth load prediction result and the service priority.

[0007] Optionally, after the control strategy is generated according to the tunnel data and before the target tunnel is disabled according to the control strategy, the method further comprises: generating a disabling instruction and delivering the disabling instruction to a network device; suspending bandwidth occupation of the target tunnel by the network device according to the disabling instruction.

[0008] Optionally, the tunnel data of the network tunnel is obtained by: collecting a packet forwarding rate, a delay, a jitter, a bandwidth occupation and a data packet size distribution of the network tunnel as the traffic state; obtaining a service priority label associated with the network tunnel as the service priority, the service priority label being determined based on a service type, an initiating department or a time sensitivity in a tunnel configuration stage.

[0009] Optionally, after the high-priority service determined based on the service priority ends, the target tunnel is re-enabled according to a network state of the network tunnel, comprising: generating a recovery sequence of the target tunnel according to the network state of the network tunnel and a historical recovery mode; delivering an enabling instruction to the network device according to the recovery sequence, and recovering communication of the target tunnel by the network device.

[0010] Optionally, after the target tunnel is re-enabled according to the network state of the network tunnel after the high-priority service determined based on the service priority ends, the method further comprises: inputting an operation log into the time sequence prediction model and the strategy generation model; incrementally training the time sequence prediction model and the strategy generation model based on the operation log and real-time service performance data.

[0011] Optionally, before the tunnel data of the network tunnel is obtained, the method further comprises: configuring a static parameter for the network tunnel, the static parameter comprising a target address and a quality of service level; associating the static parameter and the service priority label with the network tunnel.

[0012] In a second aspect, an embodiment of the present application provides a network tunnel resource scheduling system, the system comprising: a tunnel data acquisition module configured to acquire tunnel data of a network tunnel, the tunnel data comprising a service priority and a traffic state. a control strategy generation module configured to generate a control strategy according to the tunnel data; a target tunnel disabling module configured to disable a target tunnel according to the control strategy, the target tunnel being a low-priority network tunnel determined based on the service priority or an idle network tunnel determined based on the traffic state; a target tunnel restarting module configured to re-enable the target tunnel according to a network state of the network tunnel after high-priority service determined based on the service priority ends.

[0013] Optionally, the control strategy generation module comprises: a tunnel data input module configured to input the tunnel data to a time series prediction model and a strategy generation model deployed at an edge computing node or a central cloud; a tunnel data analysis module configured to analyze the tunnel data by the time series prediction model to obtain a bandwidth load prediction result of a future time period; a control strategy determination module configured to generate the control strategy according to the bandwidth load prediction result and the service priority by the strategy generation model.

[0014] Optionally, the system further comprises: a disabling instruction issuing module configured to generate a disabling instruction and issue the disabling instruction to a network device after the control strategy generation module generates the control strategy according to the tunnel data and before the target tunnel disabling module disables the target tunnel according to the control strategy; a bandwidth occupation suspending module configured to suspend bandwidth occupation of the target tunnel by the network device according to the disabling instruction.

[0015] Optionally, the tunnel data acquisition module comprises: a traffic state acquisition module configured to acquire a packet forwarding rate, a delay, a jitter, a bandwidth occupation and a data packet size distribution of the network tunnel as the traffic state; a service priority acquisition module configured to acquire a service priority label associated with the network tunnel as the service priority, the service priority label being determined based on a service type, an initiating department or a time sensitivity in a tunnel configuration stage.

[0016] Optionally, the target tunnel restarting module comprises: a recovery sequence generation module configured to generate a recovery sequence of the target tunnel according to a network state and a historical recovery mode of the network tunnel; a starting instruction issuing module configured to issue an enabling instruction to the network device according to the recovery sequence step by step to recover communication of the target tunnel by the network device.

[0017] Optionally, the system further comprises: an operation log input module, configured to input operation logs to the time series prediction model and the policy generation model after the target tunnel restart module re-enables the target tunnel according to the network state of the network tunnel after the high-priority service determined based on the service priority ends; a model incremental training module, configured to perform incremental training on the time series prediction model and the policy generation model based on the operation logs and real-time service performance data.

[0018] Optionally, the system further comprises: a static parameter configuration module, configured to configure static parameters for the network tunnel before the tunnel data acquisition module acquires tunnel data of the network tunnel, the static parameters including a target address and a quality of service level; a network tunnel association module, configured to associate the static parameters and the service priority label with the network tunnel.

[0019] In a third aspect, an electronic device is provided, which includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, and the program or instructions, when executed by the processor, implement the steps of the method according to the first aspect.

[0020] In a fourth aspect, a readable storage medium is provided, which stores a program or instructions, and the program or instructions, when executed by a processor, implement the steps of the method according to the first aspect.

[0021] In the embodiments of the present application, by acquiring tunnel data containing service priority and traffic state, a control policy is generated to dynamically disable low-priority or idle network tunnels, and after the high-priority service ends, the target tunnel is re-enabled according to the network state of the whole network. Compared with the bandwidth resource shortage caused by static configuration in the background technology and the complexity of manual adjustment, the on-demand allocation of bandwidth resources is realized, the transmission quality of high-priority services is ensured to be prioritized, manual intervention is avoided, and the potential risk of configuration errors is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a step flowchart of a network tunnel resource scheduling method according to an embodiment of the present application; Figure 2 is a structural schematic diagram of a network tunnel resource scheduling system according to an embodiment of the present application; Figure 3 is a hardware structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0024] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the front and rear associated objects are in a "or" relationship.

[0025] The embodiments of the present application provide a network tunnel resource scheduling scheme. Tunnel data of a network tunnel is obtained, including service priority and traffic state, a control strategy is generated to dynamically disable low-priority or idle tunnels, thereby releasing bandwidth resources to guarantee the transmission quality of high-priority services. After the high-priority services end, the disabled tunnels are re-enabled based on the network state of the whole network, and subsequent optimization is supported by recording operation logs. Further, the tunnel data is analyzed by using a timing prediction model and a strategy generation model to generate bandwidth load prediction and control strategies, the disable and enable operations are executed by a router, and logs and real-time service performance data are fed back to the model for incremental training, thereby realizing intelligent allocation of resources and automatic management of the system.

[0026] The network tunnel resource scheduling scheme provided by the embodiments of the present application will be described in detail below with reference to the drawings, specific embodiments and application scenarios.

[0027] Referring to Figure 1 , a step flowchart of a network tunnel resource scheduling method according to an embodiment of the present application is shown.

[0028] In step 101, tunnel data of a network tunnel is obtained.

[0029] Tunnel data refers to the operational status and traffic characteristics information related to network tunnels, specifically including service priority and traffic state. Service priority reflects the importance and real-time requirements of the services carried by network tunnels, and is usually pre-configured based on service type, initiating department, or time sensitivity during tunnel establishment. For example, video conference services may be assigned a higher priority, while monitoring backhaul services may be assigned a lower priority. Service priority is associated with network tunnels in the form of labels and stored in routers or central policy libraries for use during scheduling. Traffic state covers real-time operational data of network tunnels, including packet forwarding rate, delay, jitter, peak and average bandwidth occupancy, traffic timing pattern, and packet size distribution. Real-time operational data is collected in real time by a perception module in the router, which is usually integrated into a gigabit encryption router and is responsible for monitoring the communication status of network tunnels. The collection process is carried out through periodic scanning or event triggering to ensure that real-time operational data can reflect the current operation of network tunnels in a timely manner. For example, packet forwarding rate reflects the activity level of data transmission, while delay and jitter indicate whether the transmission quality meets the service requirements. The collected tunnel data is pre-processed (such as data format standardization or denoising) in a lightweight manner and uploaded to the analysis platform in a telemetry manner to ensure data integrity and real-time performance.

[0030] Step 102, generating control policy according to tunnel data.

[0031] Service priority identifies the importance of the service, and traffic state reflects the real-time communication situation of the network tunnel. The process of generating control policy is a comprehensive analysis of tunnel data to determine which network tunnels need to adjust their operational status, thereby optimizing bandwidth resource allocation. The analysis process first evaluates the service priority to identify the difference between high-priority services (such as emergency command) and low-priority services (such as daily monitoring), ensuring that high-priority services are prioritized in resource competition. At the same time, analyze the traffic state, for example, by checking the packet forwarding rate and bandwidth occupancy, to determine whether the network tunnel is in an active or idle state. For example, a network tunnel may carry low-priority services but currently has no data transmission, indicating that its bandwidth occupancy can be temporarily released. The generation of control policy is based on these analysis results to develop specific scheduling instructions for each network tunnel, such as marking certain network tunnels as target tunnels that need to be disabled. In the process of generating the policy, the bandwidth utilization and service load distribution of the entire network are considered to ensure that the scheduling instructions can balance network resources and service demand. The control policy is output in the form of a data structure, which may include network tunnel identification, priority weight, and whether it needs to be disabled, etc. These information is stored in the form of instruction set for subsequent steps. The implementation of generating control policy relies on the computing power of the network management platform, which is usually completed on edge computing nodes or central cloud to ensure processing speed and accuracy.

[0032] Step 103, disable the target tunnel according to the control strategy.

[0033] The target tunnel refers to a low-priority network tunnel determined based on service priority, or an idle network tunnel determined based on traffic state. The low-priority network tunnel usually carries services with low real-time requirements, such as monitoring backhaul or daily data transmission, and the priority label thereof is explicitly marked in the tunnel configuration stage. The idle network tunnel refers to a tunnel with no data transmission at present, which is identified by traffic state data (such as a packet forwarding rate close to zero or zero bandwidth occupation). The execution of the disabling operation is completed by the router, which suspends the bandwidth occupation of the target tunnel after receiving the instruction in the control strategy, while retaining the configuration information thereof for subsequent recovery. The disabling process is usually completed in milliseconds, ensuring minimal impact on the overall operation of the network. For example, for a low-priority network tunnel, its data forwarding function can be suspended, and the bandwidth occupied thereby can be redistributed to high-priority services; for an idle network tunnel, the disabling operation directly stops its resource occupation, releasing the idle bandwidth. The selection of the target tunnel to be disabled is based on the explicit instructions in the control strategy, which has comprehensively considered service priority and traffic state to avoid mistakenly disabling critical service tunnels.

[0034] Step 104, after the high-priority service determined based on service priority is completed, the target tunnel is re-enabled according to the network state of the network tunnel.

[0035] The high-priority service usually refers to a service with high real-time requirements, such as a video conference or emergency command, and the priority label thereof is explicitly marked in the tunnel configuration stage, for example, by assigning the highest priority through service type or time sensitivity. By monitoring service priority data, it is determined whether the high-priority service is completed, for example, when the video conference data stream stops or the emergency command task is completed, triggering the re-enabling process. The network state of the entire network includes information such as bandwidth utilization and service load distribution, reflecting the resource margin and operating condition of the current network. These information are obtained through real-time monitoring, which may include whether the available bandwidth of the entire network is sufficient to support the recovery of the disabled tunnel, or whether the current service load allows the addition of data transmission. The re-enabling process is executed by the router, which restores the bandwidth allocation and data forwarding function of the target tunnel after receiving the enabling instruction. The target tunnel includes a network tunnel that was previously disabled due to low priority or idle state, and the configuration information thereof (such as destination address, quality of service level) has been retained during the disabling period, so that recovery does not require reconfiguration, reducing the complexity of the operation. When re-enabling, the target tunnels can be restored in batches according to the network state to avoid congestion caused by instantaneous bandwidth competition. For example, if the bandwidth margin of the entire network is sufficient, the tunnels carrying higher-priority services are preferentially restored.

[0036] In the embodiments of the present application, by obtaining tunnel data containing service priority and traffic state, a control strategy is generated to dynamically disable low-priority or idle network tunnels, and after high-priority service ends, the target tunnel is re-enabled according to the network state of the whole network. Compared with the bandwidth resource shortage caused by static configuration in the background technology and the complexity of manual adjustment, the on-demand allocation of bandwidth resources is realized, ensuring that the transmission quality of high-priority services is preferentially guaranteed, while avoiding the cumbersome operation of manual intervention, reducing the potential risk of operation and maintenance cost and configuration errors.

[0037] In an exemplary embodiment of the present application, one implementation of generating a control strategy according to tunnel data is as follows: input the tunnel data into a time series prediction model and a strategy generation model deployed in an edge computing node or a central cloud; analyze the tunnel data by the time series prediction model to obtain bandwidth load prediction results in the future time period; and generate a control strategy according to the bandwidth load prediction results and service priority by the strategy generation model.

[0038] The time series prediction model first processes the tunnel data to analyze its time series characteristics, such as trend analysis of the bandwidth occupancy peak and mean value in the traffic state, to predict the bandwidth load in the future time period (such as the next few minutes or hours). The prediction results are output in a quantitative form, which may be a bandwidth demand curve of each network tunnel or a load distribution diagram of the whole network. Then, the strategy generation model combines the prediction results and service priority to calculate the scheduling weight of each network tunnel. For example, the network tunnel carrying high-priority services is given a higher weight, while the low-priority or idle network tunnel has a lower weight. The strategy generation model generates a control strategy according to these weights, and the output content includes network tunnel identification, scheduling instructions (such as disabling or keeping active), and bandwidth allocation suggestions. The control strategy is stored in a structured data form, which may use JavaScript Object Notation (JSON) or Extensible Markup Language (XML) format. The edge computing node is suitable for handling real-time scenarios, while the central cloud is suitable for complex computing or multi-node coordination scenarios. The whole process runs automatically without human intervention, ensuring that the scheduling decision quickly responds to network changes and is suitable for complex environments with concurrent services in high-definition video communication networks.

[0039] The embodiments of the present application input the tunnel data into the time series prediction model and the strategy generation model, generate a control strategy by combining bandwidth load prediction and service priority, realize accurate analysis of the running state of the network tunnel and intelligent generation of scheduling decisions, and can dynamically develop resource allocation schemes according to service demand and network conditions.

[0040] In an example embodiment of the present application, after the control policy is generated according to the tunnel data, and before the target tunnel is disabled according to the control policy, one implementation is to generate a disabling instruction and send it to the network device; and the network device suspends the bandwidth occupation of the target tunnel according to the disabling instruction.

[0041] The control policy is generated by the analysis platform, and contains scheduling instructions for each network tunnel, such as marking some network tunnels as target tunnels that need to be disabled. The target tunnel is usually a low-priority network tunnel (such as a tunnel carrying monitoring backhaul traffic) determined based on traffic priority or an idle network tunnel (such as a tunnel with a packet forwarding rate of zero) determined based on traffic state. The process of generating the disabling instruction is completed by the scheduling module, which parses the control policy into a specific instruction set, each instruction containing network tunnel identification, disabling action, and timestamp information. The instruction set is sent to the network device, usually a gigabit encrypted router, through a secure communication protocol. After receiving the instruction, the network device performs the disabling operation and suspends the bandwidth occupation of the target tunnel. Specifically, the network device suspends the packet forwarding function of the target tunnel by modifying the forwarding rules or closing the related ports, while retaining its configuration information (such as destination address, quality of service level) for subsequent recovery. The disabling operation is completed in milliseconds, ensuring minimal impact on the overall operation of the network. To avoid misoperation, the network device verifies the legality of the instruction before execution, such as checking whether the network tunnel identification matches. After disabling, the status of the target tunnel is updated to suspended and recorded in the local log of the network device or the central policy library. This process is fully automated, avoiding the complexity of traditional manual configuration, and is suitable for the need of quickly adjusting bandwidth allocation in high-load scenarios.

[0042] The embodiments of the present application suspend the bandwidth occupation of the target tunnel by generating a disabling instruction and sending it to the network device, achieving fast execution of the control policy, and can timely release the bandwidth resources of low-priority or idle network tunnels to provide sufficient network support for high-priority traffic, while maintaining the automation and efficiency of the operation.

[0043] In an example embodiment of the present application, one implementation of obtaining tunnel data of the network tunnel is to collect the packet forwarding rate, delay, jitter, bandwidth occupation, and packet size distribution of the network tunnel as the traffic state; and obtain the traffic priority label associated with the network tunnel as the traffic priority, which is determined based on the traffic type, the initiating department, or the time sensitivity in the tunnel configuration stage.

[0044] The perception module is a built-in functional unit of the router, responsible for real-time monitoring of the running status of the network tunnel. The packet forwarding rate reflects the data transmission frequency of the network tunnel, such as the number of packets forwarded per second; the delay and jitter indicate the time stability and quality of data transmission; the bandwidth occupation includes the peak and average, reflecting the resource usage of the network tunnel; the packet size distribution provides detailed information of the traffic pattern, such as whether there is a burst of large packet transmission. These traffic status data are collected through periodic scanning or event triggering, such as once per second or triggered when traffic is abnormal. The service priority label is determined during the tunnel configuration stage, based on service type (such as video conference, monitoring backhaul), initiating department (such as emergency command department) or time sensitivity (such as real-time requirement). For example, video conference service may be marked as "highest priority", while monitoring backhaul service is marked as "low priority". These labels are stored in the router or central policy library, associated with the network tunnel through a unique identifier. The perception module formats the collected traffic status data and obtained service priority labels, such as converting to a unified data structure (such as JSON), and then uploads them to the analysis platform through the telemetry protocol. The uploading process ensures the real-time and integrity of the data, and uses encrypted channels to protect data security. The entire acquisition process does not require human intervention, relying on the automation function of the router to complete, suitable for multi-terminal, high-concurrency high-definition video communication network, ensuring that the tunnel data can accurately reflect the network status and service demand, providing reliable support for subsequent scheduling decisions.

[0045] The embodiments of the present application realize comprehensive monitoring of the running status of the network tunnel and accurate identification of the importance of the service by collecting detailed traffic status data through the perception module and obtaining pre-configured service priority labels, providing high-quality input data for the generation of control policies, ensuring the accuracy and applicability of scheduling decisions.

[0046] In an exemplary embodiment of the present application, after the high-priority service determined based on the service priority is ended, the re-activation of the target tunnel according to the network status of the network tunnel is implemented in one embodiment as follows: generating a recovery sequence of the target tunnel according to the network status of the network tunnel and the historical recovery mode; gradually issuing activation instructions to network devices according to the recovery sequence, and recovering the communication of the target tunnel through the network devices.

[0047] The network state of the whole network includes bandwidth utilization, service load distribution and other information, which is obtained by real-time monitoring, for example, by analyzing the platform to aggregate all network tunnel traffic data, calculating the current available bandwidth and load. The historical recovery mode records the sequence and effect of previous recovery operations, for example, preferentially recovering network tunnels carrying high-priority services, or recovering in batches according to bandwidth margins. The process of generating the recovery sequence is completed in the analysis platform, which integrates the network state of the whole network and the historical recovery mode to determine the activation sequence of the target tunnel. For example, if the bandwidth of the whole network is sufficient, the network tunnel with low priority but long-term disabled may be preferentially recovered; if the bandwidth is limited, the network tunnel with greater impact on services is preferentially recovered. The recovery sequence is output in the form of an instruction set, including network tunnel identification, activation time, priority and other information. The activation instruction is issued to the network device (usually a gigabit encrypted router) through a secure communication protocol, and the network device recovers the bandwidth allocation and data forwarding function of the target tunnel according to the instruction. The recovery process uses the configuration information (such as destination address, quality of service level) retained during the disabled state, without the need for reconfiguration. The network device verifies the reachability and transmission quality of the target tunnel after recovery, for example, checks whether the data packet is normally forwarded and whether the delay meets the requirements. The recovery operation is performed in batches to avoid congestion caused by instantaneous bandwidth competition.

[0048] The embodiments of the present application generate a recovery sequence according to the network state of the whole network and the historical recovery mode, and gradually activate the target tunnel, which realizes the ordered recovery of the disabled network tunnel, dynamically balances the network resources after the high-priority service ends, avoids congestion during the recovery process, and ensures the overall availability and service continuity of the network.

[0049] In an exemplary embodiment of the present application, after the high-priority service determined based on the service priority ends, and the target tunnel is reactivated according to the network state of the network tunnel, one implementation is: inputting the operation log into the time sequence prediction model and the strategy generation model; based on the operation log and real-time service performance data, incrementally training the time sequence prediction model and the strategy generation model.

[0050] The operation log records key information during the scheduling process, including the generation record of the control strategy, the disabling and enabling time of the target tunnel, the execution status of the network device, etc. These logs are stored in a structured format (such as JSON or database table) in the central policy library or analysis platform. The input process is completed through internal data interfaces, such as transmitting the operation log to the model training platform deployed on the edge computing node or the central cloud through the message queue. Real-time service performance data includes service recovery delay (such as the first data transmission time after the target tunnel is recovered) or user satisfaction data (such as the smoothness feedback of a video conference), which is collected in real time through network devices or user terminals. These data reflect the actual effect of the scheduling operation, such as whether disabling and recovery cause service interruption or delay. The incremental training process updates the model parameters using the operation log and real-time service performance data, such as adjusting the weight of the timing prediction model to improve the bandwidth prediction accuracy, or optimizing the decision rules of the strategy generation model to reduce the recovery delay. The training process uses online learning, triggered periodically or when the data accumulates to a certain amount, avoiding interrupting normal scheduling tasks. The whole process runs automatically on the edge or cloud, without human intervention, ensuring that the model can adapt to changes in the network environment, such as fluctuations in business load or the addition of new terminals.

[0051] The embodiments of the present application realize continuous optimization of the scheduling mechanism by inputting the operation log and real-time service performance data into the model and performing incremental training, which can dynamically adapt to changes in the network environment and business demand, and improve the accuracy of the control strategy and the stability of long-term operation.

[0052] In an exemplary embodiment of the present application, before acquiring the tunnel data of the network tunnel, one implementation is to configure static parameters for the network tunnel, including target address and quality of service level; and associate the static parameters and the business priority label with the network tunnel.

[0053] The destination address refers to the communication target of the network tunnel, such as the IP address or domain name of the target server, used to determine the end point of data transmission; the quality of service level defines the transmission priority and bandwidth guarantee of the network tunnel, such as the high bandwidth and low delay level suitable for video conference service. The service priority label is determined based on the service type (such as video conference, monitoring backhaul), the initiating department (such as emergency command department), or the time sensitivity (such as real-time requirement). For example, the emergency command service can be assigned a "highest priority" label, and the monitoring backhaul service is assigned a "low priority" label. The configuration process is completed in the network tunnel establishment stage, usually performed by the network administrator through the management platform or the configuration interface of the router. The static parameters and the service priority label are associated through the unique identifier of the network tunnel to form a metadata record. For example, the record of a certain network tunnel can include its destination address as "192.168.1.1", the quality of service level as "high priority", and the service priority label as "video conference - highest priority". These metadata are stored in the local database of the router or the central policy library, synchronized to the analysis platform through a secure communication protocol, and ensured to be directly obtained in the subsequent scheduling process. After the configuration is completed, the network tunnel can enter the running state, and the traffic state data can be collected by the perception module.

[0054] The embodiments of the present application configure the destination address and the quality of service level for the network tunnel and associate the service priority label, establish the standardized initial configuration of the network tunnel, provide accurate service identification and communication parameters for subsequent data acquisition and scheduling, and ensure the pertinence of resource allocation and the standardization of network management.

[0055] Based on the above related description of the network tunnel resource scheduling method embodiment, a tunnel resource scheduling method based on dynamic disabling is introduced. The method automatically collects tunnel data of the network tunnel, generates a control strategy combining the service priority and the traffic state, dynamically disables the low-priority or idle network tunnel, and restores the tunnel communication according to the network state after the high-priority service ends, while optimizing the scheduling model through log feedback, which is suitable for complex network environment with concurrent services.

[0056] First, in the network tunnel establishment stage, static parameters and service priority labels are configured for each network tunnel. The static parameters include the destination address (such as the IP address of the target server, for example "192.168.1.100") and the quality of service level. The service priority label is determined based on the service type, the initiating department, or the time sensitivity, for example, the emergency command service is marked as "highest priority", the video conference service is "medium priority", and the monitoring backhaul service is "low priority". These static parameters and service priority labels are associated through the unique identifier of the network tunnel, and synchronized to the central policy library through a secure communication protocol, stored as JSON format metadata records for subsequent scheduling.

[0057] Next, the tunnel data of the network tunnel is collected by the perception module in the router. The perception module monitors the running state of the network tunnel in real time, collects the packet forwarding rate (such as 1000 packets per second), the delay (such as 10 milliseconds), the jitter (such as 2 milliseconds), the bandwidth occupation (such as currently occupying 50 Mbps, and the peak value is 100 Mbps), and the packet size distribution (such as 80% for small packets) as the traffic state. At the same time, the service priority label associated with the network tunnel is obtained from the central policy library or the local router as the service priority. The collected tunnel data is formatted and converted into a unified data structure (such as JSON) and uploaded to the analysis platform deployed in the central cloud through the telemetry protocol. The entire collection process is executed every second to ensure real-time data and protect data security through an encrypted channel.

[0058] Subsequently, the analysis platform inputs the tunnel data into the time series prediction model and the policy generation model to generate the control policy. The time series prediction model analyzes the traffic state in the tunnel data, combines the historical traffic pattern (such as the bandwidth occupation trend in the past hour), predicts the bandwidth load in the future time period (such as the next 10 minutes), and outputs the bandwidth demand estimation of each network tunnel, for example, "tunnel S001 is expected to occupy 80 Mbps, and tunnel S002 is expected to occupy 20 Mbps". The policy generation model then integrates the bandwidth load prediction result and the service priority to calculate the scheduling weight of each network tunnel. For example, the network tunnel S001 carrying "the highest priority" service is assigned a high weight, while the network tunnel S002 carrying "low priority" service or the idle network tunnel S003 (packet forwarding rate close to zero) is assigned a low weight. Based on the weight, the policy generation model generates the control policy, outputs an instruction set containing the network tunnel identifier and the scheduling instruction (such as "disable S002, S003"), and stores it in XML format for the scheduling module to call.

[0059] According to the control policy, the scheduling module determines the target tunnel, i.e. the low priority network tunnel (such as S002) determined based on the service priority or the idle network tunnel (such as S003) determined based on the traffic state. The scheduling module generates a disable instruction containing the network tunnel identifier and the disable action, and sends it to the gigabit encryption router. After receiving the instruction, the router verifies the legality of the instruction (such as checking the matching of the network tunnel identifier), and then suspends the bandwidth occupation of the target tunnel. The specific operation includes modifying the forwarding rule or closing the related port, stopping the data forwarding function of the target tunnel, while retaining its static parameters and priority label. The disable operation is completed in milliseconds to ensure minimal impact on the overall operation of the network. After disabling, the state of the target tunnel is updated to "suspended" and recorded in the local log of the router.

[0060] After the high-priority service (such as emergency command) based on service priority determination ends, the target tunnel is re-enabled according to the network state of the whole network. The analysis platform monitors the bandwidth utilization (such as the current available bandwidth 200 Mbps) and service load distribution (such as the proportion of high-priority service decreasing to 10%) of the whole network in real time, and generates a recovery sequence of the target tunnel in combination with the historical recovery mode (such as preferentially recovering the tunnel of the priority service in the bearer). For example, the recovery sequence may specify to recover S002 (medium priority) first and then recover S003 (low priority). The scheduling module gradually issues the enable instruction to the router according to the recovery sequence, and the router recovers the bandwidth allocation and data forwarding function of the target tunnel by using the reserved static parameters. After recovery, the router verifies the reachability and transmission quality (such as delay less than 15 milliseconds) of the target tunnel to ensure normal operation of the service. The recovery operation is carried out in batches to avoid congestion caused by instantaneous bandwidth competition.

[0061] Finally, the whole-process operation log is recorded, including control policy generation record, disable / enable instruction, router execution state, etc., and stored in the central policy library. The operation log and real-time service performance data (such as service recovery delay 10 milliseconds, user satisfaction score 4.8) are input to the time series prediction model and the policy generation model through the message queue. The model is incrementally trained based on these data to adjust the prediction weights and decision rules, such as optimizing the accuracy of bandwidth prediction or reducing the recovery delay. The training process is triggered periodically (such as once a day) to ensure that the model adapts to changes in the network environment.

[0062] The embodiments of the present application realize the on-demand allocation of bandwidth resources and the continuous optimization of the scheduling mechanism by automatically collecting tunnel data, generating control strategies, dynamically disabling low-priority or idle network tunnels, and recovering tunnel communication after high-priority services end, combined with log feedback and model training, effectively guaranteeing the transmission quality of high-priority services, reducing the complexity of manual configuration and operation and maintenance costs, and being suitable for complex scenarios of high-definition video communication networks.

[0063] It should be noted that the network tunnel resource scheduling method provided by the embodiments of the present application can be executed by a network tunnel resource scheduling system, or a control module in the network tunnel resource scheduling system for executing the network tunnel resource scheduling method. In the embodiments of the present application, the network tunnel resource scheduling system is taken as an example to execute the network tunnel resource scheduling method, and the network tunnel resource scheduling method provided by the embodiments of the present application is described.

[0064] Referring to Figure 2 , a structural schematic diagram of a network tunnel resource scheduling system according to an embodiment of the present application is shown. The system can specifically include the following modules: The tunnel data acquisition module 21 is configured to acquire tunnel data of a network tunnel, wherein the tunnel data includes service priority and traffic state. a control strategy generation module 22 configured to generate a control strategy according to the tunnel data; a target tunnel disabling module 23 configured to disable a target tunnel according to the control strategy, the target tunnel being a low-priority network tunnel determined based on the service priority or an idle network tunnel determined based on the traffic state; a target tunnel restarting module 24 configured to restart the target tunnel according to a network state of the network tunnel after high-priority service determined based on the service priority ends.

[0065] In an example embodiment of the present application, the control strategy generation module 22 comprises: a tunnel data input module configured to input the tunnel data to a time series prediction model and a strategy generation model deployed at an edge computing node or a central cloud; a tunnel data analysis module configured to analyze the tunnel data by the time series prediction model to obtain a bandwidth load prediction result of a future time period; a control strategy determination module configured to generate the control strategy according to the bandwidth load prediction result and the service priority by the strategy generation model.

[0066] In an example embodiment of the present application, the system further comprises: a disabling instruction issuing module configured to generate a disabling instruction and issue the disabling instruction to a network device after the control strategy generation module 22 generates the control strategy according to the tunnel data and before the target tunnel disabling module 23 disables the target tunnel according to the control strategy; a bandwidth occupation suspending module configured to suspend bandwidth occupation of the target tunnel by the network device according to the disabling instruction.

[0067] In an example embodiment of the present application, the tunnel data acquisition module 21 comprises: a traffic state collection module configured to collect a packet forwarding rate, a delay, a jitter, a bandwidth occupation and a data packet size distribution of the network tunnel as the traffic state; a service priority acquisition module configured to acquire a service priority label associated with the network tunnel as the service priority, the service priority label being determined based on a service type, an initiating department or a time sensitivity in a tunnel configuration stage.

[0068] In an example embodiment of the present application, the target tunnel restarting module 24 comprises: a recovery sequence generation module configured to generate a recovery sequence of the target tunnel according to a network state and a historical recovery mode of the network tunnel; The starting instruction issuing module is configured to issue the starting instruction to the network device step by step according to the recovery sequence, and recover the communication of the target tunnel through the network device.

[0069] In an example embodiment of the present application, the system further comprises: The operation log input module is configured to input operation logs to the time series prediction model and the policy generation model after the target tunnel restart module 24 restarts the target tunnel according to the network state of the network tunnel after the high-priority service determined based on the service priority ends. The model incremental training module is configured to perform incremental training on the time series prediction model and the policy generation model based on the operation logs and real-time service performance data.

[0070] In an example embodiment of the present application, the system further comprises: The static parameter configuration module is configured to configure static parameters for the network tunnel before the tunnel data acquisition module 21 acquires the tunnel data of the network tunnel, wherein the static parameters include a target address and a quality of service level. The network tunnel association module is configured to associate the static parameters and the service priority label with the network tunnel.

[0071] The network tunnel resource scheduling system in the embodiments of the present application can be a device, a component, an integrated circuit, or a chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Examples of the mobile electronic device include a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an Ultra-Mobile Personal Computer (UMPC), a netbook, or a Personal Digital Assistant (PDA), and examples of the non-mobile electronic device include a server, a Network Attached Storage (NAS), a Personal Computer (PC), a television (TV), a teller machine, or a self-service machine, without limitation.

[0072] The network tunnel resource scheduling system in the embodiments of the present application can be a device with an operating system. The operating system can be an Android operating system, an IOS operating system, or other possible operating systems, without limitation.

[0073] The network tunnel resource scheduling system provided in the embodiments of the present application canFigure 1 The processes in the method embodiments of the network tunnel resource scheduling system are not repeated here to avoid repetition.

[0074] Optionally, the embodiments of the present application also provide an electronic device, which includes a processor, a memory, a program or instructions stored in the memory and executable on the processor, the program or instructions being executed by the processor to implement the processes of the network tunnel resource scheduling method embodiments and achieve the same technical effects. To avoid repetition, the processes are not repeated here.

[0075] It should be noted that the electronic device in the embodiments of the present application includes the mobile electronic device and the non-mobile electronic device described above.

[0076] Figure 3 A hardware structure diagram of an electronic device for implementing the embodiments of the present application.

[0077] The electronic device 1000 includes, but is not limited to, a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and a processor 1010, etc. The input unit 1004 can include a graphics processor 10041 and a microphone 10042. The display unit 1006 can include a display panel 10061. The user input unit 1007 can include a touch panel 10071 and other input devices 10072. The memory 1009 can include an application program and an operating system.

[0078] Those skilled in the art can understand that the electronic device 1000 can also include a power supply (such as a battery) for supplying power to each component, and the power supply can be logically connected to the processor 1010 through a power management system, so as to realize the functions of power management, such as charging, discharging, and power consumption management, through the power management system. Figure 3 The electronic device structure shown in the above embodiment does not constitute a limitation on the electronic device, and the electronic device can include more or fewer components than those shown, or combine certain components, or have a different arrangement of components, which is not repeated here.

[0079] The embodiments of the present application also provide a readable storage medium, which stores a program or instructions, the program or instructions being executed by a processor to implement the processes of the network tunnel resource scheduling method embodiments and achieve the same technical effects. To avoid repetition, the processes are not repeated here.

[0080] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0081] It should be noted that the terms "comprising", "including", or any other variant thereof are intended to cover a non-exclusive inclusion, such that processes, methods, articles, or apparatuses that comprise a list of elements not only include those elements, but also include other elements not expressly listed or inherent to such processes, methods, articles, or apparatuses. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element. In addition, it should be pointed out that the scope of the methods and systems in the embodiments of the present application is not limited to performing functions in the order shown or discussed, but can also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method can be performed in an order different from that described, and various steps can also be added, omitted, or combined. In addition, the features described with reference to certain examples can be combined in other examples.

[0082] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a plurality of instructions for making a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the methods described in various embodiments of the present application.

[0083] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above specific embodiments, the above specific embodiments are only illustrative, not restrictive, and those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims.

Claims

1. A network tunnel resource scheduling method, characterized in that, The method comprises: acquiring tunnel data of a network tunnel, the tunnel data comprising a service priority and a traffic state; generating a control strategy according to the tunnel data; disabling a target tunnel according to the control strategy, the target tunnel being a low-priority network tunnel determined based on the service priority or an idle network tunnel determined based on the traffic state; after high-priority service determined based on the service priority ends, re-enabling the target tunnel according to a network state of the network tunnel.

2. The method of claim 1, wherein, The generating of the control strategy according to the tunnel data comprises: inputting the tunnel data into a time series prediction model and a strategy generation model deployed on an edge computing node or a central cloud; analyzing the tunnel data by the time series prediction model to obtain a bandwidth load prediction result for a future time period; generating the control strategy according to the bandwidth load prediction result and the service priority by the strategy generation model.

3. The method of claim 2, wherein, After the generating of the control strategy according to the tunnel data and before the disabling of the target tunnel according to the control strategy, the method further comprises: generating a disabling instruction and delivering it to a network device; suspending, by the network device, bandwidth occupation of the target tunnel according to the disabling instruction.

4. The method of claim 1, wherein, The acquiring of the tunnel data of the network tunnel comprises: collecting a packet forwarding rate, a delay, a jitter, a bandwidth occupation and a data packet size distribution of the network tunnel as the traffic state; acquiring a service priority label associated with the network tunnel as the service priority, the service priority label being determined based on a service type, an initiating department or a time sensitivity in a tunnel configuration stage.

5. The method of claim 3, wherein, The re-enabling of the target tunnel according to the network state of the network tunnel after the high-priority service determined based on the service priority ends comprises: generating a recovery sequence of the target tunnel according to the network state of the network tunnel and a historical recovery mode; delivering an enabling instruction to the network device according to the recovery sequence to recover communication of the target tunnel by the network device.

6. The method of claim 2, wherein, After the re-enabling of the target tunnel according to the network state of the network tunnel after the high-priority service determined based on the service priority ends, the method further comprises: inputting an operation log into the time series prediction model and the strategy generation model; based on the operation log and real-time service performance data, incrementally training the time series prediction model and the strategy generation model.

7. The method of claim 4, wherein, Before the acquiring of the tunnel data of the network tunnel, the method further comprises: configuring a static parameter for the network tunnel, the static parameter comprising a target address and a quality of service level; associating the static parameter and the service priority label with the network tunnel.

8. A network tunnel resource scheduling system, characterized by, The system comprises: a tunnel data acquisition module configured to acquire tunnel data of a network tunnel, the tunnel data comprising a service priority and a traffic state; a control strategy generation module configured to generate a control strategy according to the tunnel data; a target tunnel disabling module configured to disable a target tunnel according to the control policy, the target tunnel being a low-priority network tunnel determined based on the service priority, or an idle network tunnel determined based on the traffic state; a target tunnel restarting module configured to restart the target tunnel according to a network state of the network tunnel after high-priority service determined based on the service priority ends.

9. An electronic device, comprising: A computer readable storage medium having stored thereon program code or instructions executable by a processor to perform the steps of the network tunnel resource scheduling method according to any one of claims 1-7.

10. A readable storage medium, characterized by, A computer readable storage medium having stored thereon program code or instructions executable by a processor to perform the steps of the network tunnel resource scheduling method according to any one of claims 1-7.

Images