Bandwidth allocation method and device, electronic equipment and nonvolatile storage medium

Abstract

The invention discloses a bandwidth allocation method and device, electronic equipment and a nonvolatile storage medium. The method comprises the steps that multi-dimensional feature data of a plurality of virtual private network subnets are acquired, and the multi-dimensional feature data at least comprise bandwidth use conditions, subnet scales of the plurality of virtual private network subnets and service types corresponding to the virtual private network subnets; a bandwidth allocation prediction model is adopted to analyze the multi-dimensional feature data to obtain a prediction result, and the prediction result is used for predicting the bandwidth allocated to each virtual private network sub-network; and dynamically allocating bandwidths to the plurality of virtual private network subnets based on the prediction result and a real-time monitoring result, the real-time monitoring result being used for adjusting the weights of the plurality of virtual private network subnets. According to the method and the device, the technical problems that only fixed bandwidths can be allocated when the bandwidths are allocated to a plurality of VPN subnets and dynamic adjustment of the VPN subnet bandwidths cannot be realized in related technologies are solved.

Description

Technical Field

[0001] The present application relates to the field of network technology, and in particular to a bandwidth allocation method, device, electronic device and non-volatile storage medium. Background Art

[0002] With the rapid development of cloud-network integration services, efficient and automated network operation and management has become a focus of attention. For example, in order to meet business needs, specific bandwidth is allocated to the enterprise network to ensure that key businesses can obtain sufficient network resources and that ordinary office businesses can also reasonably allocate bandwidth. Efficient and intelligent management of private networks has become a necessary measure for enterprises and organizations.

[0003] In the private networks of various industries, the access layer dedicated lines are often divided into multiple virtual private networks (VPN) subnets. For example, the education private network is divided into: global Internet access, financial management, electronic inspection, video monitoring, physical and chemical examinations, etc. These VPN subnets share a physical dedicated line from the school to the nearby telecommunications bureau. The overall bandwidth of this dedicated line is fixed. In the related technology, when bandwidth is allocated to these subnets, a specified bandwidth is allocated to different VPNs, and it cannot be adjusted after allocation. However, in the actual business environment, the bandwidth required by these VPN subnets changes dynamically in different usage time periods and different situations. Therefore, in the related technology, when bandwidth is allocated to multiple virtual private networks (VPN) subnets, only a fixed bandwidth can be allocated, and dynamic adjustment of VPN subnet bandwidth cannot be achieved.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a bandwidth allocation method, device, electronic device and non-volatile storage medium to at least solve the technical problem in the related art that when allocating bandwidth to multiple VPN subnets, only fixed bandwidth can be allocated and dynamic adjustment of VPN subnet bandwidth cannot be achieved.

[0006] According to one aspect of an embodiment of the present application, a bandwidth allocation method is provided, comprising: obtaining multi-dimensional feature data of multiple virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of multiple virtual private network subnets, and a service type corresponding to each virtual private network subnet; using a bandwidth allocation prediction model to analyze the multi-dimensional feature data to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; and dynamically allocating bandwidth to the multiple virtual private network subnets based on the prediction result and real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of the multiple virtual private network subnets.

[0007] According to some embodiments of the present application, the subnet scale of multiple virtual private network subnets is obtained in the following manner: obtaining the subnet scale of multiple virtual private network subnets from a database of a virtual private network subnet control center, wherein the subnet scale includes at least the address space size of each virtual private network subnet and the number of devices included.

[0008] According to some embodiments of the present application, bandwidth usage is obtained in the following manner: obtaining network traffic data of multiple virtual private network subnets; performing data cleaning on the network traffic data; extracting characteristic data that affects bandwidth allocation from the network traffic data after data cleaning, and using the characteristic data as bandwidth usage, wherein the characteristic data includes at least a peak value of network traffic and an average value of network traffic for each virtual private network subnet within a preset time period.

[0009] According to some embodiments of the present application, each virtual private network subnet corresponds to a service type, and a first weight is assigned to each virtual private network subnet based on the service type, wherein the first weight is determined based on a bandwidth allocation prediction model.

[0010] According to some embodiments of the present application, the bandwidth allocation prediction model is trained in the following manner: Step 1: Obtain historical multi-dimensional feature data, wherein the historical multi-dimensional feature data at least includes historical bandwidth usage, subnet sizes of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet; Step 2: Use the historical multi-dimensional feature data to train the initial bandwidth allocation prediction model, and adjust the weight of each virtual private network subnet based on the weight of the previous iteration, a preset learning rate, and a preset loss function, wherein the weight of the previous iteration in the first iteration is the preset weight; iterate step 2 until the iteration number is reached and the iteration is stopped to obtain the bandwidth allocation prediction model, and the weight of each virtual private network subnet in the last iteration is used as the first weight of each virtual private network subnet.

[0011] According to some embodiments of the present application, bandwidth is dynamically allocated to multiple virtual private network subnets based on prediction results and real-time monitoring results, including: determining the bandwidth occupancy rate of each virtual private network subnet at intervals of a preset time period as a real-time monitoring result, wherein the bandwidth occupancy rate is determined by the ratio of the real-time bandwidth of each virtual private network subnet to the sum of the bandwidths of multiple virtual private network subnets, and the real-time bandwidth is used to indicate the bandwidth corresponding to the timestamp indicated by the current time period; adjusting the first weight of each virtual private network subnet based on the bandwidth occupancy rate; sorting the multiple virtual private network subnets in descending order according to the adjusted first weights to obtain a priority sequence; adjusting the bandwidth allocated to each virtual private network subnet indicated in the prediction result based on the priority sequence to obtain a bandwidth allocation result; and allocating bandwidth to the multiple virtual private network subnets based on the bandwidth allocation result.

[0012] According to some embodiments of the present application, the first weight of each virtual private network subnet is adjusted based on the bandwidth occupancy rate, including: increasing the first weight of the first virtual private network subnet by a preset value, and reducing the first weight of the second virtual private network subnet by a preset value, wherein the first virtual private network subnet is a virtual private network subnet whose bandwidth occupancy rate is greater than or equal to a first preset threshold value, and the second virtual private network subnet is a virtual private network subnet whose bandwidth occupancy rate is less than the first preset threshold value.

[0013] According to another aspect of an embodiment of the present application, a bandwidth allocation device is also provided, including: an acquisition module, used to acquire multi-dimensional feature data of multiple virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet; an analysis module, used to analyze the multi-dimensional feature data using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; an allocation module, used to dynamically allocate bandwidth to multiple virtual private network subnets based on the prediction result and real-time monitoring result, wherein the real-time monitoring result is used to adjust the weights of multiple virtual private network subnets.

[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is provided, in which a program is stored, wherein when the program is running, a device where the non-volatile storage medium is located is controlled to execute any one of the above bandwidth allocation methods.

[0015] According to another aspect of an embodiment of the present application, an electronic device is provided, including: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes any one of the above-mentioned bandwidth allocation methods when running.

[0016] In an embodiment of the present application, a method is adopted to obtain multi-dimensional feature data of multiple virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet scales of multiple virtual private network subnets, and service types corresponding to each virtual private network subnet; a bandwidth allocation prediction model is used to analyze the multi-dimensional feature data to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; bandwidth is dynamically allocated to multiple virtual private network subnets based on the prediction result and the real-time monitoring result, wherein the real-time monitoring result is used to adjust the weights of multiple virtual private network subnets, by obtaining multi-dimensional feature data of multiple virtual private network subnets, using the bandwidth allocation prediction model to analyze the multi-dimensional feature data to obtain the prediction result, and dynamically allocating bandwidth to multiple virtual private network subnets based on the prediction result and the real-time monitoring result, thereby achieving the purpose of dynamically allocating bandwidth to multiple virtual private network subnets, thereby solving the technical problem in the related art that only fixed bandwidth can be allocated when allocating bandwidth to multiple VPN subnets, and dynamic adjustment of VPN subnet bandwidth cannot be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal for implementing a bandwidth allocation method provided in an embodiment of the present application;

[0019] Figure 2 is a flow chart of a bandwidth allocation method provided according to an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of a VPN subnet adaptive configuration adjustment device method provided according to an embodiment of the present application;

[0021] Figure 4 It is an example diagram of a VPN subnet adaptive configuration adjustment device method provided according to an embodiment of the present application;

[0022] Figure 5 It is a structural schematic diagram of a bandwidth allocation device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] The information collected in the embodiments of the present application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or reject automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] In the related art, when bandwidth is allocated to these subnets, a specified bandwidth is allocated to different VPNs, and it cannot be adjusted after allocation. However, in an actual business environment, the bandwidth required by these VPN subnets changes dynamically in different usage time periods and different situations. Therefore, there is a problem in the related art that when bandwidth is allocated to multiple virtual private network (VPN) subnets, only a fixed bandwidth can be allocated, and the dynamic adjustment of the VPN subnet bandwidth cannot be achieved. In order to solve this problem, a relevant solution is provided in the embodiment of the present application, which is described in detail below.

[0027] According to an embodiment of the present application, an embodiment of a bandwidth allocation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a bandwidth allocation method. Figure 1 As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission module 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0029] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0030] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the bandwidth allocation method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned bandwidth allocation method is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0033] In the above operating environment, an embodiment of the present application provides an embodiment of a bandwidth allocation method. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0034] like Figure 2 FIG. 1 is a flow chart of a bandwidth allocation method provided according to an embodiment of the present application, including:

[0035] Step S202, obtaining multi-dimensional feature data of multiple virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of multiple virtual private network subnets, and a service type corresponding to each virtual private network subnet.

[0036] In the technical solution provided in step S202, the subnet scales of multiple virtual private network subnets can be obtained in the following manner: the subnet scales of multiple virtual private network subnets are obtained from the database of the virtual private network subnet control center, wherein the subnet scales at least include the address space size of each virtual private network subnet and the number of devices included. By accurately grasping the subnet scales, bandwidth requirements can be more accurately predicted, ensuring that bandwidth allocation in a large-scale network environment is more reasonable, and improving network operation efficiency, which is suitable for scenarios that require refined management, such as corporate internal networks and educational institution networks.

[0037] Bandwidth usage can be obtained in the following ways: obtaining network traffic data of multiple virtual private network subnets; performing data cleaning on the network traffic data; extracting characteristic data that affects bandwidth allocation from the cleaned network traffic data, and using the characteristic data as bandwidth usage, wherein the characteristic data at least includes the network traffic peak value and the network traffic mean value of each virtual private network subnet within a preset time period. This method improves the accuracy of bandwidth usage analysis through data cleaning and feature extraction, helps to make a quick response in a dynamically changing network environment, and is suitable for network applications that require real-time bandwidth adjustment.

[0038] It should be noted that each virtual private network subnet corresponds to a service type, and a first weight is assigned to each virtual private network subnet based on the service type, wherein the first weight is determined based on a bandwidth allocation prediction model. By assigning different weights to different service types, the bandwidth requirements of key services can be prioritized, thereby improving the stability and reliability of network services.

[0039] The following are specific embodiments:

[0040] Before obtaining the multi-dimensional feature data of multiple virtual private network subnets, initialization is required. First, the virtual private network (VPN) subnet control center device is initialized. The VPN subnet control center is a communication bridge that allocates bandwidth to each VPN subnet. After the VPN subnet control center device is initialized, it can receive registration applications from each VPN subnet and other operations. Each VPN subnet initiates a registration application to the control center by connecting to the network dedicated line reserved by the VPN subnet control center. First, the identity authentication of each VPN subnet is performed. After the authentication is passed, the VPN subnet control center stores the information of each VPN subnet (for example, the subnet scale of multiple virtual private network subnets is stored in the subnet database, and the network scale at least includes the address space size of each virtual private network subnet and the number of devices included). In order to ensure the comprehensiveness and accuracy of bandwidth allocation, it is necessary to obtain multi-dimensional feature data from multiple aspects that affect bandwidth allocation, such as bandwidth usage, subnet scale of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet. Each virtual private network subnet corresponds to a service type. Different service types have different bandwidth requirements, such as video conferencing, data transmission type, etc. Therefore, it is necessary to consider the impact of service type on bandwidth allocation.

[0041] When acquiring multi-dimensional feature data, a data receiving and processing device can be used to acquire the subnet scales of multiple virtual private network subnets from the database of the virtual private network subnet control center, and a network traffic monitoring tool can be used to acquire real-time network traffic data of multiple virtual private network subnets (i.e., the network traffic data of the above-mentioned multiple virtual private network subnets). This step can be performed by a timer, for example, acquiring data every five minutes to ensure the real-time nature of the data. Data cleaning is performed on the acquired real-time network traffic data. For example, data cleaning can be performed in the following ways: by calculating the mean and standard deviation of the data, data points that exceed a certain range (such as mean ± 3 times standard deviation) are identified as outliers and deleted or replaced; data points that are obviously inconsistent with the time series trend are identified and deleted. For example, if the network traffic is stable for most of the time, but an extremely high peak suddenly appears, the data of this point is deleted, and data consistency checks are performed: ensure that the timestamps of all data are continuous. If data is missing, interpolation or filling with the nearest normal value can be performed. Convert all traffic data to the same unit, such as from kbps to Mbps, to avoid analysis errors caused by inconsistent units. Finally, the traffic data is normalized or standardized so that data of different sizes can be compared on the same scale. Feature data that affects bandwidth allocation is extracted from the network traffic data after data cleaning. Feature data is data associated with and affecting bandwidth allocation. For example, features such as traffic peaks (P1, P2, P3...Pk) and traffic averages (A1, A2, A3...Ak) in a preset time period (1-k minutes) are extracted from the network traffic data, and the feature data is used as bandwidth usage.

[0042] Step S204, using a bandwidth allocation prediction model to analyze the multi-dimensional feature data to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets.

[0043] In the technical solution provided in step S204, the bandwidth allocation prediction model is trained in the following manner: Step 1: Obtain historical multi-dimensional feature data, wherein the historical multi-dimensional feature data at least includes historical bandwidth usage, subnet sizes of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet; Step 2: Use the historical multi-dimensional feature data to train the initial bandwidth allocation prediction model, and adjust the weight of each virtual private network subnet based on the weight of the previous iteration, a preset learning rate, and a preset loss function, wherein the weight of the previous iteration in the first iteration is the preset weight; iterate step 2 until the iteration number is reached and the iteration is stopped to obtain the bandwidth allocation prediction model, and the weight of each virtual private network subnet in the last iteration is used as the first weight of each virtual private network subnet.

[0044] The following are specific embodiments:

[0045] Step 1: Obtain historical multi-dimensional feature data (e.g., hourly network traffic data in the past year), Step 2: Use the historical multi-dimensional feature data to train the initial bandwidth allocation prediction model (e.g., an artificial intelligence model of a multi-layer perceptron (MPL)), adjust the weight of each virtual private network subnet based on the weight of the previous iteration, the preset learning rate, and the preset loss function, iterate step 2 until the iteration number is reached, and stop the iteration to obtain the bandwidth allocation prediction model, and use the weight of each virtual private network subnet at the last iteration as the first weight of each virtual private network subnet. For example, the weight of each virtual private network subnet can be updated by the following formula:

[0046]

[0047] Among them, W i represents the weight determined in the last iteration (the preset weight in the first iteration), W i ′ represents the updated weight of the weight determined in the previous iteration (the first weight in the last iteration), α represents the learning rate, and E represents the preset loss function. Represents the preset loss function E with respect to weight W i The partial derivative of .

[0048] Step S206: dynamically allocating bandwidth to the multiple virtual private network subnets based on the prediction results and the real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of the multiple virtual private network subnets.

[0049] In the technical solution provided in step S206, there are multiple ways to dynamically allocate bandwidth for multiple virtual private network subnets based on the prediction results and the real-time monitoring results. For example, the following methods can be used to dynamically allocate bandwidth for multiple virtual private network subnets based on the prediction results and the real-time monitoring results: determine the bandwidth occupancy rate of each virtual private network subnet as the real-time monitoring result at intervals of a preset time period, wherein the bandwidth occupancy rate is determined by the ratio of the real-time bandwidth of each virtual private network subnet to the total bandwidth of multiple virtual private network subnets, and the real-time bandwidth is used to indicate the bandwidth corresponding to the timestamp indicated by the current time period; adjust the first weight of each virtual private network subnet based on the bandwidth occupancy rate; sort the multiple virtual private network subnets in order from large to small according to the adjusted first weight to obtain a priority sequence; adjust the bandwidth allocated to each virtual private network subnet indicated in the prediction result based on the priority sequence to obtain a bandwidth allocation result; and allocate bandwidth to multiple virtual private network subnets based on the bandwidth allocation result. This method can quickly respond to network changes through real-time monitoring and dynamic adjustment, effectively improve the flexibility and efficiency of bandwidth allocation, and is suitable for scenarios with large fluctuations in network traffic, such as network management during holidays and large-scale events.

[0050] There are multiple ways to adjust the first weight of each virtual private network subnet based on the bandwidth occupancy rate in the above steps, for example: increase the first weight of the first virtual private network subnet by a preset value, and reduce the first weight of the second virtual private network subnet by a preset value, wherein the first virtual private network subnet is a virtual private network subnet whose bandwidth occupancy rate is greater than or equal to the first preset threshold, and the second virtual private network subnet is a virtual private network subnet whose bandwidth occupancy rate is less than the first preset threshold. This weight adjustment mechanism can ensure that high-load subnets obtain more bandwidth resources, while low-load subnets release resources, thereby realizing reasonable redistribution of resources, and is particularly suitable for scenarios of resource optimization and cost control, such as network resource management in data centers.

[0051] The following are specific embodiments:

[0052] The bandwidth allocation prediction model includes an input layer, a subnet feature fusion layer, a hidden layer, and an output layer, wherein the input layer is used to input multi-dimensional feature data. The subnet feature fusion layer is used to learn the weights between different features and fuse them into a more representative feature vector (i.e., determine the first weight of each VPN subnet). The hidden layer is the core part of the bandwidth allocation prediction model, responsible for extracting nonlinear features from the input data. The number of hidden layers and the number of neurons in each layer can be adjusted according to the complexity of the problem. The more layers and neurons there are, the more overfitting problems need to be considered. The output layer is used to output the prediction results, which indicate the bandwidth expected to be allocated to each virtual private network subnet. For example, the hidden layer and the output layer process data through the following formula to obtain the prediction results:

[0053]

[0054] Where y represents the prediction result (i.e., the bandwidth size expected to be allocated to each VPN subnet, which meets the bandwidth requirements of the service corresponding to the service type), and X i represents the input value (multi-dimensional feature data of the i-th VPN subnet), W i represents the first weight corresponding to the i-th VPN subnet, b represents the bias term, which is a preset constant, and f represents the preset activation function.

[0055] In order to realize dynamic allocation of bandwidth, the network status is continuously monitored in real time through the real-time monitoring device to analyze the current bandwidth occupancy, user behavior, etc. The bandwidth occupancy rate of each virtual private network subnet is determined at intervals of a preset time period (for example, 1 minute) as a real-time monitoring result. The bandwidth occupancy rate of each VPN subnet connection (U1, U2, U3...Uk) is realized in real time. For example, the bandwidth occupancy rate can be obtained and updated in real time by the following formula:

[0056]

[0057] Among them, U i Indicates the bandwidth utilization of the ith subnet, BW i represents the real-time bandwidth of the i-th VPN subnet, It represents the total bandwidth of all VPN subnets (the total bandwidth of the above multiple virtual private network subnets), that is, the bandwidth of the entire dedicated line, which is a fixed value.

[0058] The first weight of each virtual private network subnet is adjusted based on the bandwidth occupancy rate by the decision-making device and the dynamic adjustment device; the plurality of virtual private network subnets are sorted in descending order according to the adjusted first weights to obtain a priority sequence: a first preset threshold (for example, 90%) is set, and when the bandwidth occupancy rate of a VPN subnet is greater than the first preset threshold, it indicates that the bandwidth resources of the VPN subnet are tight and it may be in the service execution stage, and more bandwidth needs to be allocated preferentially, and the first weight is increased by a preset value (for example, 0.1), and the weight of the VPN subnet whose bandwidth occupancy rate is less than the first preset threshold is reduced. For example, if the bandwidth occupancy rate of a VPN subnet is 10 %, at this time, the VPN subnet is not in the peak business period, the first weight is reduced by a preset value (for example, 0.1), and multiple virtual private network subnets are sorted in descending order according to the adjusted first weight to obtain a priority sequence, which can ensure that the priority sequence is sorted according to the degree of tension of VPN subnet bandwidth resources. The more VPN subnets in the front need to allocate more bandwidth. The bandwidth allocated to each virtual private network subnet indicated in the prediction result is adjusted based on the priority sequence to obtain a bandwidth allocation result; bandwidth is allocated to multiple virtual private network subnets based on the bandwidth allocation result, and the adjusted bandwidth allocation result can meet the real-time bandwidth requirements of each VPN subnet. Because a timer is used, that is, the bandwidth occupancy rate of each virtual private network subnet is determined at intervals of a preset time period as a real-time monitoring result, it is ensured that the bandwidth allocated to different VPN subnets is dynamically adjusted according to the real-time network status, so that the utilization rate of each VPN subnet reaches the optimal level. For example, when the network is congested at a certain node in the first cycle (the corresponding VPN subnet bandwidth occupancy rate is as high as 90%), the bandwidth allocated to the VPN subnet at the end of the priority sequence will be reduced based on the adjusted bandwidth allocation result (after the reduction, the normal connectivity of the VPN subnet can still be guaranteed), and the reduced bandwidth will be allocated to the VPN subnet first. In the second time period, it is detected that the bandwidth occupancy rate of the VPN subnet has become 20%, indicating that the VPN subnet service has ended. Based on the priority sequence adjusted in this cycle, the bandwidth previously allocated to the VPN subnet will be allocated to the VPN subnets at the top of other priority sequences.

[0059] Through the above steps, it is possible to intelligently predict and allocate bandwidth based on the actual needs and business types of the virtual private network subnet, effectively avoiding resource waste and network congestion. It is particularly suitable for bandwidth management in large-scale network environments such as cloud computing, data centers, and the Internet of Things.

[0060] The embodiment of the present application also provides a VPN subnet adaptive configuration adjustment device method schematic diagram, such as Figure 3As shown, a schematic diagram of a VPN subnet adaptive configuration adjustment device method applicable to the implementation of the present application is shown. The adaptive configuration adjustment device and method module is responsible for realizing the core function, dynamic bandwidth allocation, and dynamic bandwidth allocation for multiple VPN subnets 1-VPN subnet k. The VPN subnet control center is a communication bridge that allocates bandwidth to each VPN subnet. Each VPN subnet initiates a registration application to the control center by connecting to the network dedicated line reserved by the VPN subnet control center. First, identity authentication of each VPN subnet is performed. After the authentication is passed, the VPN subnet control center stores the information of each VPN subnet.

[0061] The present application also provides a VPN subnet adaptive configuration adjustment device method example diagram, such as Figure 4 As shown, the implementation process of the method in the embodiment of the present application is presented. First, VPN subnet registration is performed: each VPN subnet initiates a registration application to the control center by connecting to the network dedicated line reserved by the VPN subnet control center; each VPN subnet information is obtained from the VPN subnet control center and data acquisition and data cleaning are performed through a data collection and preprocessing device and a feature extraction device to obtain multi-dimensional feature data and historical multi-dimensional feature data, and an initial bandwidth allocation prediction model is trained through a data training device to obtain a bandwidth allocation prediction model, and real-time monitoring results are obtained through a real-time monitoring device. Then, bandwidth is dynamically allocated to each VPN subnet through a decision-making device and a dynamic adjustment device, that is, bandwidth is dynamically allocated to multiple virtual private network subnets based on the prediction results and real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of multiple virtual private network subnets.

[0062] The present application also provides a schematic diagram of the structure of a bandwidth allocation device, such as Figure 5 As shown, including:

[0063] The acquisition module 502 is used to acquire multi-dimensional feature data of multiple virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet.

[0064] The acquisition module 502 is further configured to acquire the subnet sizes of multiple virtual private network subnets from a database of the virtual private network subnet control center, wherein the subnet size at least includes the address space size of each virtual private network subnet and the number of devices included.

[0065] The acquisition module 502 is also used to obtain network traffic data of multiple virtual private network subnets; perform data cleaning on the network traffic data; and extract characteristic data that affects bandwidth allocation from the network traffic data after data cleaning, wherein the characteristic data at least includes the network traffic peak value and the network traffic mean value of each virtual private network subnet within a preset time period.

[0066] The analysis module 504 is used to analyze the multi-dimensional feature data using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model through historical multi-dimensional feature data of multiple virtual private network subnets.

[0067] The allocation module 506 is used to dynamically allocate bandwidth to the multiple virtual private network subnets based on the prediction results and the real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of the multiple virtual private network subnets.

[0068] The allocation module 506 is also used to determine the bandwidth occupancy rate of each virtual private network subnet at intervals of a preset time period, wherein the bandwidth occupancy rate is determined by the ratio of the real-time bandwidth of each virtual private network subnet to the sum of the bandwidths of multiple virtual private network subnets, and the real-time bandwidth is used to indicate the bandwidth corresponding to the timestamp indicated by the current time period; adjust the first weight of each virtual private network subnet based on the bandwidth occupancy rate; sort the multiple virtual private network subnets in descending order according to the adjusted first weights to obtain a priority sequence; adjust the bandwidth allocated to each virtual private network subnet indicated in the prediction result based on the priority sequence to obtain a bandwidth allocation result; and allocate bandwidth to the multiple virtual private network subnets based on the bandwidth allocation result.

[0069] The allocation module 506 is also used to increase the first weight of the first virtual private network subnet by a preset value, and reduce the first weight of the second virtual private network subnet by a preset value, wherein the first virtual private network subnet is a virtual private network subnet whose bandwidth occupancy is greater than or equal to a first preset threshold, and the second virtual private network subnet is a virtual private network subnet whose bandwidth occupancy is less than the first preset threshold.

[0070] It should be noted that Figure 5 The bandwidth allocation device shown is used to perform Figure 2 The bandwidth allocation method shown is therefore Figure 2 The relevant explanations in the bandwidth allocation method in are also applicable to the bandwidth allocation device and will not be repeated here.

[0071] It should be noted that each module in the above-mentioned bandwidth allocation device can be a program module (for example, a set of program instructions that implement a certain specific function) or a hardware module. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0072] The embodiment of the present application also provides a non-volatile storage medium, the non-volatile storage medium includes a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the above bandwidth allocation method. For example, multi-dimensional feature data of multiple virtual private network subnets are obtained, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet size of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet; the multi-dimensional feature data is analyzed using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training the initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; based on the prediction result and the real-time monitoring result, bandwidth is dynamically allocated to multiple virtual private network subnets, wherein the real-time monitoring result is used to adjust the weights of multiple virtual private network subnets.

[0073] The embodiment of the present application also provides an electronic device, the electronic device includes a processor, the processor is used to run a program, wherein the above bandwidth allocation method is executed when the program is running. For example, multi-dimensional feature data of multiple virtual private network subnets are obtained, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet size of multiple virtual private network subnets, and the service type corresponding to each virtual private network subnet; the multi-dimensional feature data is analyzed using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; bandwidth is dynamically allocated to multiple virtual private network subnets based on the prediction results and real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of multiple virtual private network subnets.

[0074] According to another aspect of the embodiment of the present application, a computer program product is also provided, including a computer program, which implements the above bandwidth allocation method when executed by a processor. For example, multi-dimensional feature data of multiple virtual private network subnets are obtained, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet size of multiple virtual private network subnets, and service type corresponding to each virtual private network subnet; the multi-dimensional feature data is analyzed using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each virtual private network subnet, and the bandwidth allocation prediction model is obtained by training the initial bandwidth allocation prediction model with historical multi-dimensional feature data of multiple virtual private network subnets; bandwidth is dynamically allocated to multiple virtual private network subnets based on the prediction result and real-time monitoring result, wherein the real-time monitoring result is used to adjust the weights of multiple virtual private network subnets.

[0075] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0076] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0077] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0078] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0079] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program code.

[0080] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A bandwidth allocation method, characterized in that: include: Acquire multi-dimensional feature data of a plurality of virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of the plurality of virtual private network subnets, and a service type corresponding to each virtual private network subnet; The multi-dimensional feature data is analyzed using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each of the virtual private network subnets, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model using the historical multi-dimensional feature data of the multiple virtual private network subnets; Dynamically allocating bandwidth to the multiple virtual private network subnets based on the prediction results and the real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of the multiple virtual private network subnets.

2. The method according to claim 1, characterized in that: The subnet sizes of the multiple virtual private network subnets are obtained by: The subnet scales of the multiple virtual private network subnets are obtained from a database of a virtual private network subnet control center, wherein the subnet scales at least include the address space size of each virtual private network subnet and the number of devices included.

3. The method according to claim 1, characterized in that The bandwidth usage is obtained in the following way: Obtaining network traffic data of the plurality of virtual private network subnets; Performing data cleaning on the network traffic data; Extract characteristic data that affects bandwidth allocation from the network traffic data after data cleaning, and use the characteristic data as the bandwidth usage, wherein the characteristic data at least includes a network traffic peak value and a network traffic mean value of each virtual private network subnet within a preset time period.

4. The method according to claim 1, characterized in that: Each virtual private network subnet corresponds to a service type, and a first weight is assigned to each virtual private network subnet based on the service type, wherein the first weight is determined based on the bandwidth allocation prediction model.

5. The method according to claim 4, characterized in that The bandwidth allocation prediction model is trained in the following way: Step 1: Acquire the historical multi-dimensional feature data, wherein the historical multi-dimensional feature data at least includes historical bandwidth usage, subnet sizes of the multiple virtual private network subnets, and a service type corresponding to each of the virtual private network subnets; Step 2: Use the historical multi-dimensional feature data to train the initial bandwidth allocation prediction model, and adjust the weight of each virtual private network subnet based on the weight of the previous iteration, the preset learning rate, and the preset loss function, wherein the weight of the previous iteration in the first iteration is the preset weight; Iterate step 2 until the iteration number is reached and then stop iterating to obtain the bandwidth allocation prediction model, and use the weight of each virtual private network subnet in the last iteration as the first weight of each virtual private network subnet.

6. The method according to claim 4, characterized in that The dynamically allocating bandwidth to the multiple virtual private network subnets based on the prediction results and the real-time monitoring results includes: Determine the bandwidth occupancy rate of each virtual private network subnet at intervals of a preset time period as the real-time monitoring result, wherein the bandwidth occupancy rate is determined by the ratio of the real-time bandwidth of each virtual private network subnet to the sum of the bandwidths of the multiple virtual private network subnets, and the real-time bandwidth is used to indicate the bandwidth corresponding to the timestamp indicated by the current time period; Adjusting the first weight of each virtual private network subnet based on the bandwidth occupancy rate; Sorting the plurality of virtual private network subnets in descending order according to the adjusted first weights to obtain a priority sequence; Adjust the bandwidth allocated to each virtual private network subnet indicated in the prediction result based on the priority sequence to obtain a bandwidth allocation result; Bandwidth is allocated to the plurality of virtual private network subnets based on the bandwidth allocation result.

7. The method according to claim 6, characterized in that The adjusting the first weight of each virtual private network subnet based on the bandwidth occupancy rate includes: The first weight of the first virtual private network subnet is increased by a preset value, and the first weight of the second virtual private network subnet is reduced by the preset value, wherein the first virtual private network subnet is a virtual private network subnet whose bandwidth occupancy is greater than or equal to a first preset threshold, and the second virtual private network subnet is a virtual private network subnet whose bandwidth occupancy is less than the first preset threshold.

8. A bandwidth allocation device, characterized in that: include: An acquisition module, used to acquire multi-dimensional feature data of a plurality of virtual private network subnets, wherein the multi-dimensional feature data at least includes bandwidth usage, subnet sizes of the plurality of virtual private network subnets, and a service type corresponding to each virtual private network subnet; An analysis module, configured to analyze the multi-dimensional feature data using a bandwidth allocation prediction model to obtain a prediction result, wherein the prediction result is used to predict the bandwidth allocated to each of the virtual private network subnets, and the bandwidth allocation prediction model is obtained by training an initial bandwidth allocation prediction model using the historical multi-dimensional feature data of the multiple virtual private network subnets; An allocation module is used to dynamically allocate bandwidth to the multiple virtual private network subnets based on the prediction results and the real-time monitoring results, wherein the real-time monitoring results are used to adjust the weights of the multiple virtual private network subnets.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the bandwidth allocation method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the bandwidth allocation method according to any one of claims 1 to 7 is executed when the program is run.

Images