Network configuration methods, devices, electronic equipment and storage media

Abstract

This application discloses a network configuration method, apparatus, electronic device, and storage medium, relating to the field of Internet technology. The method includes: acquiring first state data of the network devices; inputting the first state data into a preset time series model to obtain a prediction result, wherein the time series model is a combination of an ARIMA model and an LSTM model; and dynamically adjusting the network configuration of the network devices based on the prediction result. This application can improve the adaptability and resource utilization efficiency of network configuration.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to networking configuration methods, devices, electronic devices and storage media. Background Technology

[0002] Network configuration refers to the process of planning, setting up, and adjusting various devices (such as routers, switches, servers, etc.) and their connections in a network according to specific network requirements, with the aim of building a network environment that can meet specific functional and performance requirements.

[0003] Traditional network configuration methods often struggle to effectively address complex network demands when faced with dynamic changes in device status and environment, easily leading to resource waste, network performance degradation, and device failures.

[0004] In summary, improving the adaptability and resource utilization efficiency of network configuration has become a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0005] The main objective of this application is to provide a network configuration method, apparatus, electronic device, and storage medium, which aims to improve the adaptability and resource utilization efficiency of network configuration.

[0006] To achieve the above objectives, this application proposes a network configuration method, which includes:

[0007] Obtain the initial status data of the network devices;

[0008] The first state data is input into a preset time series model to obtain the prediction result. The time series model is a combination of ARIMA (integrated autoregressive moving average model) and LSTM (Long Short-Term Memory) model.

[0009] The networking configuration of the networking devices is dynamically adjusted based on the prediction results.

[0010] In one embodiment, the step of acquiring the first status data of the networking device includes:

[0011] Obtain the initial first state data of the network devices;

[0012] The initial first state data is sequentially subjected to noise reduction, anomaly detection, and data format standardization to obtain the first state data of the networking device.

[0013] In one embodiment, before the step of inputting the first state data into a preset time series model to obtain the prediction result, the method further includes:

[0014] The ARIMA model and the LSTM model are trained and validated using historical state data, so that the ARIMA model learns the short-term change characteristics of the historical state data and the LSTM model learns the long-term change characteristics of the historical state data.

[0015] The trained ARIMA model and the trained LSTM model are combined to obtain the time series model. The combination method is either front-to-back combination or weighted combination.

[0016] In one embodiment, the step of dynamically adjusting the networking configuration of the networking devices based on the prediction result includes:

[0017] Based on the prediction results, determine whether the resource allocation of the networking equipment is abnormal;

[0018] When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm to obtain a configuration optimization strategy;

[0019] The network configuration of the networking devices is dynamically adjusted according to the configuration optimization strategy described above.

[0020] In one embodiment, the prediction results include CPU (Central Processing Unit) load, network traffic, and device load. The step of determining whether the resource allocation of the network devices is abnormal based on the prediction results includes:

[0021] When the CPU load exceeds a preset first load threshold, it is determined that there is an anomaly in the resource allocation of the networking device, and a first anomaly prompt is output.

[0022] When the network traffic exceeds a preset traffic threshold, it is determined that there is an anomaly in the resource allocation of the networking device, and a second anomaly prompt is output;

[0023] When the device load exceeds a preset second load threshold, it is determined that there is an abnormality in the resource allocation of the network device, and a third abnormality prompt is output.

[0024] In one embodiment, the step of using the prediction result as input data for a preset intelligent optimization algorithm to obtain a configuration optimization strategy when the resource allocation is abnormal includes:

[0025] When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, a configuration optimization strategy is determined through selection, crossover, and mutation operations.

[0026] or,

[0027] When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined by simulating the behavior of a particle swarm.

[0028] In one embodiment, the method further includes:

[0029] Obtain the second status data of the networking device, and calculate the error value between the second status data and the prediction result;

[0030] When the error value exceeds a preset error threshold, the time series model is corrected, and the networking configuration of the networking device is adjusted according to the second state data.

[0031] Furthermore, to achieve the above objectives, this application also proposes a network configuration device, which includes:

[0032] The data acquisition module is used to acquire the first status data of the network devices;

[0033] The prediction module is used to input the first state data into a preset time series model to obtain the prediction result, wherein the time series model is a combination of ARIMA model and LSTM model;

[0034] The configuration adjustment module is used to dynamically adjust the networking configuration of the networking devices based on the prediction results.

[0035] In addition, to achieve the above objectives, this application also proposes an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the networking configuration method described above.

[0036] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the network configuration method described above.

[0037] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the networking configuration method described above.

[0038] This application proposes a network configuration method. First-state data of the network devices is acquired and input into a preset time-series model. This time-series model is a combination of ARIMA and LSTM models. It leverages the advantages of ARIMA in analyzing linear time-series relationships and LSTM in handling complex nonlinear time-series data to more accurately predict the future state of the network devices. Then, based on the prediction results of the time-series model, the network configuration of the network devices is dynamically adjusted. Specifically, this adjustment can include modifying the network topology, optimizing resource allocation, and adjusting operating modes. This enhances the adaptability of the network configuration, enabling it to better adapt to dynamic changes in the network environment, optimize resource allocation, improve resource utilization efficiency, and ensure network stability and efficiency. Attached Figure Description

[0039] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating the network configuration method of this application in Embodiment 1.

[0042] Figure 2 A flowchart illustrating the network configuration method of this application in Embodiment 2;

[0043] Figure 3 This is another flowchart illustrating the network configuration method in Embodiment 2 of this application.

[0044] Figure 4 This is a schematic diagram of the module structure of the networking configuration device according to an embodiment of this application;

[0045] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the networking configuration method in this application embodiment.

[0046] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0047] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0048] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0049] This application provides a solution to obtain first state data of networking devices; input the first state data into a preset time series model to obtain prediction results, wherein the time series model is a combination of ARIMA model and LSTM model; and dynamically adjust the networking configuration of networking devices according to the prediction results.

[0050] Traditional network configuration methods often struggle to effectively address complex network demands when faced with dynamic changes in device status and environment, easily leading to resource waste, network performance degradation, and device failures.

[0051] Specifically, the Internet of Things (IoT) involves a large number of devices whose status and environmental parameters change constantly over time. Traditional static network configurations are mostly based on fixed configurations and cannot adapt to these changes in real time, thus affecting the stability and efficiency of the network. For example, if the CPU utilization of a device suddenly increases, and the network configuration cannot be adjusted in time, the device may become a network bottleneck, thereby affecting the performance of the entire network.

[0052] Furthermore, while existing dynamic networking can cope with changes in device status to some extent, it can usually only configure the network based on the current or recent device status, and cannot accurately predict future device status and network requirements. It lacks sufficient predictive capabilities and global optimization strategies, and cannot achieve optimal network configuration.

[0053] In summary, this embodiment of the application acquires the first state data of the networking devices and inputs it into a preset time series model. This time series model is a combination of the ARIMA model and the LSTM model. This time series model can leverage the advantages of the ARIMA model in analyzing linear relationships in time series and the advantages of the LSTM model in processing complex nonlinear time series data to more accurately predict the future state of the networking devices. Then, based on the prediction results of the time series model, the networking configuration of the networking devices is dynamically adjusted. Specifically, the adjustment can be to adjust the network topology of the networking devices, optimize resource allocation, and adjust the working mode, thereby improving the adaptability of the networking configuration. This allows it to better adapt to the dynamic changes in the network environment, optimize resource allocation, improve resource utilization efficiency, and ensure the stability and efficiency of the network.

[0054] It should be noted that the executing entity in this embodiment can be an electronic device with data processing, network communication, and program execution functions, such as a computer, server, or any other electronic device capable of performing the aforementioned functions. The following description uses a network configuration terminal as an example to illustrate this embodiment and the subsequent embodiments.

[0055] Based on this, the embodiments of this application provide a network configuration method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the networking configuration method of this application.

[0056] In this embodiment, the network configuration method includes steps S10 to S30:

[0057] Step S10: Obtain the first status data of the networking devices;

[0058] It should be noted that networking devices are used in various device networking scenarios, such as multi-device management in the Internet of Things, edge computing node collaboration, and even distributed cloud computing environments. Networking devices connect different devices and networks to achieve data transmission and resource sharing. Networking devices can specifically be routers, switches, super devices, gateways, etc. This embodiment does not specifically limit the type and number of networking devices.

[0059] The data acquisition module deployed on the network edge computing node acquires the current status data of the network devices, i.e., the first status data, through real-time monitoring or periodic collection. This first status data includes:

[0060] Hardware utilization: such as CPU utilization, memory utilization, storage utilization, etc.

[0061] Network status: such as network bandwidth utilization, packet loss rate, communication latency, etc.

[0062] Equipment power consumption: Monitors the power consumption and battery status of networked devices, especially suitable for mobile devices.

[0063] Environmental information: Collect parameters such as temperature, humidity, and light intensity of the environment in which the networked devices are located.

[0064] Geographic location information: The geographic location of the networked devices is obtained through GPS (Navigation Satellite Timing and Ranging / Global Positioning System) or other positioning technologies.

[0065] It should also be noted that, in this embodiment, the frequency of status data collection can be adjusted according to the type of networking device and the dynamics of the environment. For example, mobile networking devices in a highly dynamic environment require higher collection frequencies.

[0066] Step S20: Input the first state data into the preset time series model to obtain the prediction result, wherein the time series model is a combination of ARIMA model and LSTM model;

[0067] After obtaining the first state data of the network devices, the first state data is input into a pre-built time series model. This time series model is a combination of the ARIMA model and the LSTM model. The ARIMA model is good at capturing linear relationships and time trends in the data, while the LSTM model can handle more complex nonlinear relationships and long-term dependency problems. By combining the two models, the time series model can more accurately predict the future state of the network devices, so as to prevent potential problems such as possible failures and performance bottlenecks of the network devices in advance.

[0068] Step S30: Dynamically adjust the networking configuration of the networking devices based on the prediction results.

[0069] After the time series model outputs the prediction results, the network configuration of the networking devices is dynamically adjusted based on the prediction results. The adjustments can include adjusting the network topology of the networking devices, optimizing resource allocation, and working modes, in order to prevent network failures, improve network stability and performance, and maximize resource utilization.

[0070] In summary, this embodiment of the application acquires the first state data of the networking devices and inputs it into a preset time series model. This time series model is a combination of the ARIMA model and the LSTM model. This time series model can leverage the advantages of the ARIMA model in analyzing linear relationships in time series and the advantages of the LSTM model in processing complex nonlinear time series data to more accurately predict the future state of the networking devices. Then, based on the prediction results of the time series model, the networking configuration of the networking devices is dynamically adjusted. Specifically, the adjustment can be to adjust the network topology of the networking devices, optimize resource allocation, and adjust the working mode, thereby improving the adaptability of the networking configuration. This allows it to better adapt to the dynamic changes in the network environment, optimize resource allocation, improve resource utilization efficiency, and ensure the stability and efficiency of the network.

[0071] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, step S10 may include steps S101 to S102:

[0072] Step S101: Obtain the initial first state data of the networking devices;

[0073] Step S102: The initial first state data is sequentially processed by noise reduction, anomaly detection and data format standardization to obtain the first state data of the network device.

[0074] To ensure the quality and consistency of the acquired first-state data, the initial first-state data of the network devices is preprocessed. The preprocessing specifically includes:

[0075] Noise reduction: Kalman filtering, low-pass filtering and other algorithms are used to remove noise from the initial first-state data to ensure data smoothness.

[0076] Anomaly detection: By setting multiple thresholds and rules (such as the Z-score algorithm), possible abnormal data points are detected and marked, such as a sudden jump in CPU utilization or an abnormal increase in network latency.

[0077] Data format standardization: Standardize the format of initial first-state data from different sources, and unify the timestamps, units and structures of the data.

[0078] In one feasible embodiment, before step S20, such as Figure 2 As shown, it may also include steps A10 to A20:

[0079] Step A10: Train and validate the ARIMA model and LSTM model using historical state data, so that the ARIMA model learns the short-term change characteristics of the historical state data and the LSTM model learns the long-term change characteristics of the historical state data.

[0080] It should be noted that, since the status data of networked devices has obvious time dependence and trends, this embodiment selects two main time series models for combination. One is the ARIMA model, which is mainly used to process linear time series data and is suitable for analyzing and predicting the periodic or trend changes of devices. The other is the LSTM model, which is a neural network-based Long Short-Term Memory (LSTM) model. It is good at capturing nonlinear changes and long-term dependencies in device status and is particularly suitable for predicting complex dynamic behaviors.

[0081] Historical state data of networked devices are divided into training and validation sets to ensure that the time series model can learn the main patterns of historical state data without overfitting. Then, through cross-validation and grid search, the hyperparameters of the time series model, such as the order (p, d, q) of ARIMA and the number of layers and hidden units of LSTM, are adjusted to minimize prediction error. For the ARIMA model, historical state data is used to estimate the parameters of autoregression and moving average; for the LSTM model, the network parameters are trained through feedforward and backpropagation to enable it to accurately predict the future state of devices. Then, historical state data in the validation set is used to evaluate the prediction accuracy of the time series model. The performance of the time series model is judged by indicators such as MSE (mean squared error) and MAE (mean absolute error), and the model is further adjusted based on the validation results.

[0082] Step A20: Combine the trained ARIMA model and the trained LSTM model to obtain a time series model. The combination method is either front-to-back combination or weighted combination.

[0083] By combining the trained ARIMA model and the trained LSTM model, a time series model is obtained. In one feasible implementation, the ARIMA model is first used to perform short-term linear predictions to obtain preliminary prediction results. Then, these results are used as input to the LSTM model, which further predicts the nonlinear changes and long-term trends of the device based on the ARIMA prediction results.

[0084] In another feasible implementation, the prediction results of ARIMA and LSTM are combined in a weighted manner to obtain the final prediction result. The weights can be adjusted according to the verification accuracy of historical state data to ensure the prediction accuracy of the combined time series model.

[0085] It is worth mentioning that compared with using the ARIMA model or LSTM model alone, this embodiment shows better performance in terms of prediction accuracy, robustness, and adaptability. The combined time series model can not only capture the subtle fluctuations in the short-term status data of network devices, but also accurately predict long-term development trends, which is beneficial for status monitoring, fault early warning, and resource scheduling of network devices. Furthermore, this embodiment proposes two strategies: pre- and post-combination and weighted combination. Pre- and post-combination means first using the ARIMA model for short-term prediction, and then using the prediction result as input to the LSTM model for long-term trend correction and prediction. This combination method can fully utilize the predictive advantages of the two models to form a complementary effect. Weighted combination, on the other hand, dynamically adjusts the weights of the prediction results based on the predictive performance of the two models to optimize the prediction results.

[0086] In one feasible embodiment, such as Figure 3 As shown, step S30 may include steps S301 to S303:

[0087] Step S301: Determine whether the resource allocation of the networking devices is abnormal based on the prediction results;

[0088] Based on the prediction results output by the time series model, a comprehensive assessment of the resource allocation status of network devices is conducted. Specifically, by comparing the predicted data of network devices with the actual resource requirements, it is determined whether there are any abnormalities in the current resource allocation of network devices. Abnormalities refer to situations such as resource surplus, resource shortage, or uneven resource allocation.

[0089] Step S302: When there is an anomaly in resource allocation, the prediction result is used as the input data of the preset intelligent optimization algorithm to obtain the configuration optimization strategy;

[0090] When an anomaly is identified in resource allocation, the prediction results output by the time series model are used as input data and fed into a preset intelligent optimization algorithm. This intelligent optimization algorithm can be a genetic algorithm or a particle swarm optimization algorithm. The intelligent optimization algorithm can automatically search for and find the optimal or near-optimal configuration optimization strategy based on the input data. By executing the configuration optimization strategy, resource allocation can be optimized.

[0091] Step S303: Dynamically adjust the networking configuration of the networking devices according to the configuration optimization strategy.

[0092] The configuration optimization strategy dynamically adjusts the network devices. The adjustment actions may include adjusting the working mode of the network devices, reallocating resources, and optimizing the network topology, so as to adjust the network configuration of the network devices in real time according to changes in the network environment and adapt to the ever-changing network requirements.

[0093] In one feasible embodiment, the prediction results include CPU load, network traffic, and device load, and step S301 may include steps S3011 to S3013:

[0094] It should be noted that the prediction results based on the time series model include the CPU load, network traffic and device load of the networking devices. This can help identify potential resource bottlenecks or anomalies that the networking devices may face in advance, and make corresponding network configuration adjustments.

[0095] Step S3011: When the CPU load exceeds the preset first load threshold, determine that there is an abnormality in the resource allocation of the networking device and output the first abnormality prompt.

[0096] When the prediction results show that the CPU load exceeds the preset first load threshold, it is determined that the computing capacity of the network equipment is about to approach or reach saturation, and there is an abnormal risk in resource allocation. At this time, it is determined that there is an abnormality in resource allocation, and the first abnormality prompt is output to prompt the corresponding measures to be taken, such as increasing computing resources and optimizing task scheduling.

[0097] Step S3012: When the network traffic exceeds the preset traffic threshold, determine that there is an abnormality in the resource allocation of the networking device and output a second abnormality prompt.

[0098] If the prediction results show that the network traffic exceeds the preset traffic threshold, it is determined that the network communication capacity of the networking equipment cannot meet the current or future needs, and there is an abnormal risk in resource allocation. At this time, it is determined that there is an abnormality in resource allocation, and a second abnormality prompt is output to prompt the corresponding measures to be taken, such as increasing bandwidth and optimizing the network topology.

[0099] Step S3013: When the device load exceeds the preset second load threshold, it is determined that there is an abnormality in the resource allocation of the network device, and a third abnormality prompt is output.

[0100] When the prediction results show that the device load exceeds the preset second load threshold, it is determined that the overall performance of the network devices is close to the limit and there is an abnormal risk in resource allocation. At this time, it is determined that there is an abnormality in resource allocation and a third abnormality prompt is output to prompt corresponding measures, such as upgrading hardware devices or adjusting resource allocation strategies.

[0101] In one feasible embodiment, step S302 may include step S3021:

[0102] Step S3021: When there is an anomaly in resource allocation, the prediction result is used as the input data of the preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined through selection, crossover and mutation operations.

[0103] After determining that there is an anomaly in resource allocation, the predicted results of CPU load, network traffic, and device load obtained through time series modeling are used as input data for the intelligent optimization algorithm. In this embodiment, the intelligent optimization algorithm is preferably a genetic algorithm. The intelligent optimization algorithm uses the prediction results to perform a series of selection, crossover (or hybridization), and mutation operations. These operations simulate natural selection, the transmission of genetic information, and gene mutation in the process of biological evolution. The goal is to gradually optimize the configuration parameters of the network devices through iterative search to achieve the optimal state of resource allocation. Finally, the algorithm will output one or more configuration optimization strategies. These configuration optimization strategies can significantly reduce the probability of resource allocation anomalies and improve the overall performance of the network.

[0104] In another feasible embodiment, step S302 may include step S3022:

[0105] Step S3022: When there is an anomaly in resource allocation, the prediction result is used as the input data for the preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined by simulating the behavior of the particle swarm.

[0106] After determining that there is an anomaly in resource allocation, the predicted results of CPU load, network traffic, and device load obtained through time series modeling are used as input data for the intelligent optimization algorithm. In this embodiment, the intelligent optimization algorithm is preferably the particle swarm optimization algorithm. The intelligent optimization algorithm finds the optimal solution by simulating the flight behavior of a group of particles in the search space. Each particle represents a possible configuration optimization strategy, and the flight direction and speed of the particle are affected by its own historical best position and the historical best position of the entire group. The algorithm iterates continuously, updating the position and speed of the particles until it finds the optimal or near-optimal configuration optimization strategy. These configuration optimization strategies aim to optimize resource allocation and improve network performance and stability by adjusting the configuration parameters of the networking devices.

[0107] For example, in one feasible implementation, the specific measures for configuring the optimization strategy include:

[0108] Task migration: When it is predicted that the CPU load of a certain network device is about to reach a bottleneck, some computing tasks are automatically migrated to another network device or multiple other network devices with lower load to balance network resources.

[0109] Network bandwidth adjustment: Dynamically adjust the bandwidth allocation for data transmission based on network traffic forecasts, prioritizing the communication quality of critical tasks.

[0110] Backup device activation: When it is predicted that some network devices will be overloaded or fail, backup devices will be automatically activated to take over tasks or share the load.

[0111] Dynamic network topology adjustment: The network topology is dynamically adjusted based on prediction results to improve network robustness and efficiency. Specifically, this includes:

[0112] Communication path optimization: Adjust the data transmission path between network devices to reduce transmission delay and avoid network congestion.

[0113] Node role redefinition: Dynamically adjust the role of network devices (such as core nodes and edge nodes) based on their status and load, thereby optimizing the load distribution of the entire network.

[0114] Furthermore, in one feasible embodiment, the networking configuration strategy can be continuously optimized based on feedback data analysis and historical records from the networking devices. For example, the priority of tasks can be dynamically adjusted according to changes in the status of the networking devices to ensure that critical tasks are executed first, or the resource allocation strategy can be adjusted based on the real-time load of the networking devices to ensure optimal resource utilization of the devices in different states.

[0115] In one feasible embodiment, the network configuration method of this application may further include steps S40 to S50:

[0116] Step S40: Obtain the second state data of the networking device and calculate the error value between the second state data and the prediction result;

[0117] It should be noted that as time goes by and the status of network devices changes, the time series model needs to be updated regularly. The time series model should be retrained or fine-tuned based on the newly collected status data of the network devices in order to maintain its predictive ability.

[0118] The second state data of the network devices is obtained. This second state data is the state data of the network devices obtained after the first state data through real-time monitoring or periodic collection after the prediction result. The error value between the second state data and the prediction result is calculated. This error value reflects the degree of deviation between the time series model prediction and the actual state.

[0119] Step S50: When the error value exceeds the preset error threshold, the time series model is corrected, and the networking configuration of the networking devices is adjusted according to the second state data.

[0120] If the calculated error value exceeds the preset error threshold, it indicates that the prediction performance of the time series model needs to be corrected. The correction may include adjusting the model parameters, updating the training data, and introducing new feature variables to improve the prediction accuracy of the time series model. At the same time, the network configuration of the networking devices is fine-tuned based on the second state data to better adapt to the current network environment and needs, and further improve the stability and performance of the network.

[0121] For example, to help understand the implementation flow of the network configuration method obtained by combining this embodiment with the above embodiment one, an overall flow embodiment description is provided:

[0122] 1. Data acquisition and preprocessing.

[0123] 1.1 Input, i.e., obtaining the initial first state data of the network devices:

[0124] Device status data: X t= [CPU utilization t, memory utilization t, network latency t, device power consumption t, ambient temperature t, ...] (network device status data at multiple times);

[0125] Time series T = [t1, t2, ..., tn].

[0126] 1.2 Processing involves sequentially performing noise reduction, anomaly detection, and data format standardization on the initial first-state data to obtain the first-state data of the network devices:

[0127] Data standardization: Units of equipment status data are normalized using the mean-standard deviation formula: X tnorm =σX t -μ where μ is the mean and σ is the standard deviation.

[0128] Denoising: Denoising methods such as Kalman filtering are used to ensure data smoothness. The filtering model is as follows: x t =x t-1 +K t (z t -x t-1 Among them, K t It is the Kalman gain, z t These are noise measurement data.

[0129] 1.3 Output:

[0130] High-quality time series data after standardization, denoising, and anomaly processing: X clean =[X t1norm ,X t2norm ,…,X tnnorm ].

[0131] 2. Time series model construction and training, which involves training and validating the ARIMA and LSTM models using historical state data, so that the ARIMA model learns the short-term change characteristics of the historical state data and the LSTM model learns the long-term change characteristics of the historical state data.

[0132] 2.1 Input:

[0133] Preprocessed device status time series data: X clean .

[0134] Historical data: H = [H t-n H t-(n-1) ,…,H t (Historical device status data is used to train the model).

[0135] 2.2 Processing:

[0136] ARIMA model: used for linear trend forecasting, its mathematical model is: X t =c+φ1X t-1 +φ2X t-2 +…+∈ t Where φ1, φ2, ... are autoregressive coefficients, ∈ t This is the error term.

[0137] LSTM model: used to capture long-term dependencies in the states of network devices. The model structure is: h t =LSTM(X) t-n ,X t-n+1 ,…,X t ), where the computation of the LSTM unit includes a forget gate, an input gate, and an output gate, respectively: f t =σ(W f· [h t-1 ,X t ]+b f )i t =σ(W i· [h t-1 ,X t ]+b i )o t =σ(W o· [h t-1 ,X t ]+b o (where f) t i t ,o t W represents the activation values ​​of the forget gate, input gate, and output gate, respectively. f W i W o It is a weight matrix.

[0138] 2.3 Output:

[0139] The trained ARIMA and LSTM models are used for short-term linear prediction and long-term complex prediction, respectively.

[0140] Prediction Result: Prediction of Network Equipment Status X for Future Time Period t+1 ,X t+2 ,…

[0141] 3. Optimization of dynamic networking configuration.

[0142] 3.1 Input: The first-state data is input into a preset time series model to obtain the prediction results.

[0143] Time series prediction results: X t+1 ,X t+2 ,…

[0144] Current network device status: X t .

[0145] 3.2 Processing, i.e., dynamically adjusting the network configuration of network devices based on the prediction results:

[0146] Task migration algorithm: When the CPU or network load of network devices is predicted to reach a bottleneck, a load balancing algorithm is used to calculate the task allocation among network devices. The objective is to minimize the maximum load, as shown in the formula:

[0147]

[0148] Among them, L i For the load of network device i, C i The computing power of network device i.

[0149] Network bandwidth adjustment: Based on predicted traffic demand, adjust the bandwidth allocation among network devices. The goal is to maximize network utilization, and the optimization problems are as follows:

[0150]

[0151] Among them, B i It is the bandwidth allocated to network device i, U(B) i ) is the utility function for bandwidth allocation.

[0152] Backup device activation: When a network device is predicted to fail, the backup device is activated. The decision function is:

[0153]

[0154] Among them, P failure It is the predicted probability of network equipment failure.

[0155] 3.3 Output:

[0156] Adjusted network configuration: task allocation scheme, bandwidth allocation scheme, and list of enabled backup devices.

[0157] 4. Real-time feedback and adaptive optimization

[0158] 4.1 Input:

[0159] Real-time network device status: X treal ;

[0160] Model predicted state: X tpredict .

[0161] 4.2 Processing involves acquiring the second-state data of the network devices and calculating the error between the second-state data and the prediction results. When the error exceeds a preset error threshold, the time series model is corrected, and the network configuration of the network devices is adjusted based on the second-state data.

[0162] Feedback mechanism: Real-time monitoring of the difference between the current status of network devices and the predicted status, calculating the error: E t = |X treal -X tpredict When the error exceeds the threshold, the system retrains the model and updates the network configuration.

[0163] 4.3 Output:

[0164] Adjusted model parameters and optimized configurations ensure the continuous and efficient operation of the device network.

[0165] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the networking configuration method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0166] This application also provides a network configuration device; please refer to... Figure 4 The networking configuration device includes:

[0167] Data acquisition module 10 is used to acquire the first status data of the network devices;

[0168] Prediction module 20 is used to input the first state data into a preset time series model to obtain prediction results, wherein the time series model is a combination of ARIMA model and LSTM model;

[0169] The configuration adjustment module 30 is used to dynamically adjust the networking configuration of the networking devices based on the prediction results.

[0170] Optionally, the data acquisition module 10 is also used for:

[0171] Obtain the initial first state data of the network devices;

[0172] The initial first-state data is sequentially processed by denoising, anomaly detection, and data format standardization to obtain the first-state data of the network devices.

[0173] Optionally, the network configuration device also includes a model training module (not shown), which is used for:

[0174] The ARIMA and LSTM models are trained and validated using historical state data, so that the ARIMA model learns the short-term change characteristics of the historical state data and the LSTM model learns the long-term change characteristics of the historical state data.

[0175] The trained ARIMA model and the trained LSTM model are combined to obtain a time series model. The combination method is either front-to-back combination or weighted combination.

[0176] Optionally, the configuration adjustment module 30 is also used for:

[0177] Determine whether the resource allocation of the networking equipment is abnormal based on the prediction results;

[0178] When there are anomalies in resource allocation, the prediction results are used as input data for a preset intelligent optimization algorithm to obtain a configuration optimization strategy.

[0179] The network configuration of the networking devices is dynamically adjusted according to the configuration optimization strategy.

[0180] Optionally, the prediction results include CPU load, network traffic, and device load, and the configuration adjustment module 30 is also used for:

[0181] When the CPU load exceeds the preset first load threshold, it is determined that there is an abnormality in the resource allocation of the network devices, and the first abnormality prompt is output.

[0182] When network traffic exceeds a preset traffic threshold, it is determined that there is an anomaly in the resource allocation of the networking devices, and a second anomaly prompt is output.

[0183] When the device load exceeds the preset second load threshold, it is determined that there is an abnormality in the resource allocation of the network devices, and a third abnormality prompt is output.

[0184] Optionally, the configuration adjustment module 30 is also used for:

[0185] When there are anomalies in resource allocation, the prediction results are used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined through selection, crossover and mutation operations.

[0186] or,

[0187] When there are anomalies in resource allocation, the prediction results are used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined by simulating the behavior of the particle swarm.

[0188] Optionally, the network configuration device also includes a data feedback module (not shown), which is used for:

[0189] Acquire the second state data of the network devices and calculate the error value between the second state data and the prediction result;

[0190] When the error value exceeds the preset error threshold, the time series model is corrected, and the networking configuration of the networking devices is adjusted according to the second state data.

[0191] The network configuration device provided in this application, employing the network configuration method described in the above embodiments, can improve the adaptability and resource utilization efficiency of network configuration. Compared with the prior art, the beneficial effects of the network configuration device provided in this application are the same as those of the network configuration method provided in the above embodiments, and other technical features in the network configuration device are the same as those disclosed in the network configuration method of the above embodiments, and will not be repeated here.

[0192] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the networking configuration method in Embodiment 1 above.

[0193] The following is for reference. Figure 5 The diagram illustrates a structural schematic of an electronic device suitable for implementing embodiments of this application. The electronic devices in these embodiments may include, but are not limited to, mobile terminals such as laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), and PMPs (Portable Media Players), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0194] like Figure 5As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0195] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0196] The electronic device provided in this application, employing the networking configuration method described in the above embodiments, can improve the adaptability and resource utilization efficiency of the networking configuration. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the networking configuration method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the networking configuration method of the previous embodiment, and will not be repeated here.

[0197] It should be understood that the various parts disclosed in the embodiments of this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0198] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0199] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the networking configuration method in the above embodiments.

[0200] The computer-readable storage medium provided in this application embodiment may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0201] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0202] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by an electronic device, the electronic device causes the electronic device to: acquire first state data of the network devices; input the first state data into a preset time series model to obtain a prediction result, wherein the time series model is a combination of an ARIMA model and an LSTM model; and dynamically adjust the network configuration of the network devices according to the prediction result.

[0203] Computer program code for performing the operations of the embodiments of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0204] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0205] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0206] The readable storage medium provided in this application embodiment is a computer-readable storage medium. This computer-readable storage medium stores computer-readable program instructions (i.e., a computer program) for executing the above-described network configuration method, which can improve the adaptability and resource utilization efficiency of the network configuration. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application embodiment are the same as the beneficial effects of the network configuration method provided in the above embodiments, and will not be repeated here.

[0207] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the networking configuration method described above.

[0208] The computer program product provided in this application can solve the technical problem of network configuration. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the network configuration method provided in the above embodiments, and will not be repeated here.

[0209] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A network configuration method, characterized in that, The method includes: Obtain the initial status data of the network devices; The first state data is input into a preset time series model to obtain the prediction result, wherein the time series model is a combination of ARIMA model and LSTM model; The networking configuration of the networking devices is dynamically adjusted based on the prediction results. The implementation process of the network configuration method includes: (1) Data acquisition and preprocessing; (1.1) Input, i.e., obtaining the initial first state data of the network devices: Device status data: X t =[CPU utilization t, memory utilization t, network latency t, device power consumption t, ambient temperature t, ...]; Time series T = [t1, t2, ..., tn]; (1.2) Processing, namely, performing noise reduction, anomaly detection, and data format standardization on the initial first state data in sequence to obtain the first state data of the network device: Data standardization: Units of equipment status data are normalized using the mean-standard deviation formula: X tnorm =σX t -μ, where μ is the mean and σ is the standard deviation; Denoising: Kalman filtering is used for denoising. The filtering model is as follows: x t =x t-1 +K t (z t -x t-1 ), where K t It is the Kalman gain, z t These are noise measurement data; (1.3) Output: High-quality time series data after standardization, denoising, and anomaly processing: X clean =[X t1norm ,X t2norm ,…,X tnnorm ]; (2) Time series model construction and training, that is, training and validating the ARIMA model and LSTM model through historical state data, so that the ARIMA model learns the short-term change characteristics of historical state data and the LSTM model learns the long-term change characteristics of historical state data. (2.1) Input: Preprocessed device status time series data: X clean ; Historical data: H=[H t-n H t-(n-1) ,…,H t ]; (2.2) Processing: ARIMA model: used for linear trend prediction, the mathematical model is: X t =c+ X t-1 + X t-2 +…+ ,in, It is the autoregressive coefficient. It is an error term; LSTM model: used to capture long-term dependencies in the states of network devices. The model structure is: h t =LSTM(X t-n ,X t-n+1 ,…,X t ), where the computation of the LSTM unit includes a forget gate, an input gate, and an output gate, respectively: f t =σ(W f▪ [h t-1 ,X t ]+b f );i t =σ(W i▪ [h t-1 ,X t ]+b i );o t =σ(W o▪ [h t-1 ,X t ]+b o ); where f t 、i t o t W represents the activation values ​​of the forget gate, input gate, and output gate, respectively. f 、W i 、W o It is a weight matrix; (2.3) Output: The trained ARIMA and LSTM models are used for short-term linear prediction and long-term complex prediction, respectively. Prediction Result: Prediction of Network Equipment Status X for Future Time Period t+1 ,X t+2 ,…; (3) Optimization of dynamic networking configuration; (3.1) Input: The first state data is input into the preset time series model to obtain the prediction results. Time series prediction results: X t+1 ,X t+2 ,…; Current network device status: X t ; (3.2) Processing, i.e., dynamically adjusting the networking configuration of networking devices based on the prediction results: The step of dynamically adjusting the networking configuration of the networking devices based on the prediction results includes: Based on the prediction results, determine whether the resource allocation of the networking equipment is abnormal; When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm to obtain a configuration optimization strategy; The network configuration of the networking devices is dynamically adjusted according to the configuration optimization strategy described above. The prediction results include CPU load, network traffic, and device load. The step of determining whether the resource allocation of the network devices is abnormal based on the prediction results includes: When the CPU load exceeds a preset first load threshold, it is determined that there is an anomaly in the resource allocation of the networking device, and a first anomaly prompt is output. When the network traffic exceeds a preset traffic threshold, it is determined that there is an anomaly in the resource allocation of the networking device, and a second anomaly prompt is output; When the device load exceeds a preset second load threshold, it is determined that there is an anomaly in the resource allocation of the network device, and a third anomaly prompt is output; The step of using the prediction result as input data for a preset intelligent optimization algorithm to obtain a configuration optimization strategy when an anomaly occurs in resource allocation includes: When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, a configuration optimization strategy is determined through selection, crossover, and mutation operations. or, When there is an anomaly in the resource allocation, the prediction result is used as input data for a preset intelligent optimization algorithm. Based on the intelligent optimization algorithm, the configuration optimization strategy is determined by simulating the behavior of a particle swarm. Task migration algorithm: When the CPU or network load of network devices is predicted to reach a bottleneck, a load balancing algorithm is used to calculate the task allocation among network devices; the objective is to minimize the maximum load, as shown in the formula: ； Among them, L i For the load of network device i, C i The computing power of network device i; Network bandwidth adjustment: Based on predicted traffic demand, adjust the bandwidth allocation among network devices; the goal is to maximize network utilization, and the optimization issues are as follows: ； Among them, B i It is the bandwidth allocated to network device i, U(B) i ) is the utility function for bandwidth allocation; Backup device activation: When a network device is predicted to fail, the backup device is activated. The decision function is: ； Among them, P failure It is the predicted probability of network equipment failure; (3.3) Output: Adjusted network configuration: task allocation scheme, bandwidth allocation scheme, list of enabled backup devices; (4) Real-time feedback and adaptive optimization (4.1) Input: Real-time network device status: X treal ; Model predicted state: X tpredict ; (4.2) Processing, namely, acquiring the second state data of the network devices and calculating the error value between the second state data and the prediction result, wherein the second state data is the state data of the network devices obtained after the first state data through real-time monitoring or periodic collection after the prediction result, and the error value between the second state data and the prediction result is calculated. The error value reflects the degree of deviation between the time series model prediction and the actual state; when the error value exceeds the preset error threshold, the time series model is corrected, and the network configuration of the network devices is adjusted according to the second state data: Feedback mechanism: Real-time monitoring of the difference between the current status of network devices and the predicted status, calculating the error: E t =∣X treal -X tpredict | When the error exceeds the threshold, the system retrains the model and updates the network configuration; (4.3) Output: Adjusted model parameters and optimized configurations ensure the continuous and efficient operation of the device network.

2. A networking configuration device, characterized in that, The network configuration device employs the network configuration method as described in claim 1, and the network configuration device includes: The data acquisition module is used to acquire the first status data of the network devices; The prediction module is used to input the first state data into a preset time series model to obtain the prediction result, wherein the time series model is a combination of ARIMA model and LSTM model; The configuration adjustment module is used to dynamically adjust the networking configuration of the networking devices based on the prediction results.

3. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the networking configuration method as described in claim 1.

4. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the network configuration method as described in claim 1.

Images