Distributed network node management system based on Internet of Things

Abstract

The invention relates to the technical field of network node management, in particular to a distributed network node management system based on the Internet of Things, which comprises a network performance evaluation module, a node load optimization module, a node state monitoring module and a fault prevention operation module. According to the invention, by collecting the delay and data packet loss rate data of each network node in real time and evaluating the real-time performance of the network node, performance monitoring and stability evaluation are realized, the node processing capability is evaluated in combination with bandwidth information, and load distribution is optimized, so that network state changes can be responded more flexibly, resource utilization is optimized, and the network quality is improved. The deviation degree and the abnormal state of the node can be found in time through the operation key indexes of the Internet of Things technology monitoring equipment, the continuity and the stability of the network service are guaranteed, and the network service interruption is prevented by carrying out detailed health assessment on the network node and adjusting the network configuration parameters and activating the standby node according to the assessment result. And the reliability and efficiency of the network are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network node management, and particularly to a decentralized network node management system based on the Internet of Things. Background Art

[0002] The technical field of network node management involves systems for monitoring and controlling devices or nodes in a network. This field mainly focuses on the configuration, performance, fault diagnosis, and security of nodes. Network nodes can be routers, switches, servers, or any other network communication devices. Managing nodes typically requires centralized or decentralized systems to ensure network stability and efficiency. This technology includes not only the physical management of hardware but also strategies for remote management and maintenance through software. As the complexity of network architectures increases, effective node management becomes particularly crucial to ensure smooth information transmission and optimal allocation of network resources.

[0003] Among them, a decentralized network node management system is a system used to control and monitor multiple nodes distributed in a network. The main use of this topic is to optimize communication and data exchange between nodes through centralized or decentralized methods in a large-scale distributed environment. The system can automatically handle node failures, reconfigure the network to adapt to changing requirements, and maintain the stable operation of the network. Using a decentralized network node management system can significantly improve the reliability and efficiency of the network, especially in scenarios with a wide geographical distribution and a high number of devices.

[0004] Traditional management systems rely on centralized or partially decentralized systems for network node management and cannot handle rapidly changing network states and node failures in a large-scale distributed network in a timely and accurate manner. In particular, there is insufficient coordination between physical hardware management and software remote management strategies, resulting in uneven resource allocation and response delays. Traditional systems are not precise enough in monitoring network node performance and predicting faults and lack deep integration of real-time data analysis. As a result, when problems occur in network nodes, response measures may be delayed or inaccurate, affecting the overall operation efficiency and data transmission quality of the entire network. For example, when a node is overloaded and the load distribution is not adjusted in a timely manner, it may cause network congestion and data loss, thereby affecting the overall reliability of network services and user experience. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose a decentralized network node management system based on the Internet of Things.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: A decentralized network node management system based on the Internet of Things, the system includes:

[0007] The network performance evaluation module collects the latency and packet loss rate data of each network node based on decentralized network nodes, evaluates the real-time performance of the network nodes, analyzes the performance data of the network nodes over a period of time, evaluates the stability of the nodes according to the changes in the performance data, and generates network node performance data information;

[0008] The node load optimization module evaluates the processing capacity of each node based on the network node performance data information and combines the bandwidth of each network node, and redistributes the node load according to the current total network load and the load status of each node to obtain the node load optimization result;

[0009] The node status monitoring module, based on the node load optimization result, detects the CPU usage rate, memory usage rate, and device operating temperature of the network node device through the Internet of Things, compares the real-time operating data with the preset health standards, evaluates the deviation degree of the network node device operating data, and combines the time since the last maintenance of the node device to evaluate the abnormality degree of the node and generate a node abnormality evaluation result;

[0010] The fault prevention operation module evaluates the current limiting ratio of the network node based on the node abnormality evaluation result, adjusts the network configuration parameters, and activates the standby node to maintain the uninterrupted network service and obtains the network abnormality prevention operation result.

[0011] The improvement of the present invention is that the method for evaluating the real-time performance of the network nodes is as follows:

[0012] Based on decentralized network nodes, collect the latency and packet loss rate data of each network node to obtain the basic network status information;

[0013] Based on the basic network status information, through the formula:

[0014]

[0015] Calculate the performance score of the node, and evaluate the real-time performance of the network node according to the size of the performance score to obtain the network node performance evaluation result;

[0016] Among them, α S represents the weight of the latency time, β S represents the weight of the packet loss rate, d S is the latency time of the node, p S is the packet loss rate of the node, e is the base of the natural logarithm, and S is the performance score of the node.

[0017] The improvement of the present invention is that the steps for obtaining the network node performance data information are as follows:

[0018] Based on the network node performance evaluation results, collect the performance scores of network nodes over a period of time to obtain node stability correlation data;

[0019] Based on the node stability correlation data, through the formula:

[0020]

[0021] Calculate the stability score of the node;

[0022] where R is the stability score of the node, n R is the total number of data points, S i is the i-th performance score; is the average value of the performance scores, and σ max is the maximum standard deviation threshold;

[0023] Based on the stability score of the node, integrate the performance scores of network nodes to obtain network node performance data information.

[0024] The improvement of the present invention is that the method for evaluating the processing capacity of each node is:

[0025] Based on the network node performance data information, extract the performance score, stability score, and bandwidth of each node to obtain processing capacity correlation data;

[0026] Based on the processing capacity correlation data, through the formula:

[0027]

[0028] Calculate the processing capacity index of the node;

[0029] where S is the performance score of the node, R is the stability score of the node, B is the bandwidth of the node, and B max is the maximum value of the bandwidths of all nodes in the network, and P is the processing capacity index of the node;

[0030] Based on the processing capacity index of the node, evaluate the processing performance of the node according to the size of the processing capacity index to obtain the processing performance evaluation result.

[0031] The improvement of the present invention is that the steps for obtaining the node load optimization result are:

[0032] Based on the processing performance evaluation result, collect the processing capacity index of each node and the actual load of the current node, and calculate the total processing capacity of all nodes to obtain load distribution correlation data;

[0033] Based on the load distribution correlation data, through the formula:

[0034]

[0035] Calculate the new load distribution value;

[0036] Wherein, L new is the new load distribution value, P is the processing capacity index of the node, L total is the total load of the network, P total is the sum of the processing capacity indexes of all nodes;

[0037] Based on the new load distribution value, redistribute the node loads to obtain the optimized node load result.

[0038] The improvement of the present invention is that the method for evaluating the deviation degree of the operation data of the network node device is:

[0039] Based on the optimized node load result, detect the CPU usage rate, memory usage rate and device operation temperature of the network node device through the Internet of Things to obtain real-time operation data;

[0040] Based on the real-time operation data, compare it with the preset health standard through the formula:

[0041]

[0042] Calculate the operation parameter deviation index D to obtain the operation deviation evaluation result:

[0043] Wherein, C u is the current CPU usage rate, M u is the current memory usage rate, T u is the current device operation temperature, C s is the health standard value of the CPU usage rate, M s is the health standard value of the memory usage rate, T s is the health standard value of the device operation temperature, w C is the weight of the CPU usage rate, w M is the weight of the memory usage rate, w T is the weight of the temperature.

[0044] The improvement of the present invention is that the steps for obtaining the node abnormality evaluation result are

[0045] Based on the operation deviation evaluation result, extract the time since the last maintenance of the node device according to the maintenance record of the network node device to obtain the abnormal correlation data;

[0046] Based on the abnormal correlation data, through the formula:

[0047]

[0048] Calculate the abnormality index;

[0049] Among them, β E is the adjustment coefficient, θ E is the standard number of days threshold for the equipment maintenance cycle, T E is the time since the last maintenance, D is the operation parameter deviation index, e is the base of the natural logarithm, and E is the anomaly index;

[0050] Based on the anomaly index, compare it with a preset anomaly threshold to determine whether the network node device is abnormal, and obtain the node anomaly evaluation result.

[0051] The improvement of the present invention is that the obtaining step of the network anomaly prevention operation result is:

[0052] Based on the node anomaly evaluation result, through the formula:

[0053]

[0054] Calculate the bandwidth ratio after flow limiting;

[0055] Among them, E is the anomaly index, E max is the historical maximum anomaly index of similar devices, α Z is the adjustment coefficient, and Z is the bandwidth ratio after flow limiting;

[0056] Based on the bandwidth ratio after flow limiting, adjust the network configuration parameters and activate the standby node to maintain the uninterrupted network service, and obtain the network anomaly prevention operation result.

[0057] Compared with the prior art, the advantages and positive effects of the present invention are:

[0058] In the present invention, by collecting the delay and packet loss rate data of each network node in real time and evaluating its real-time performance, performance monitoring and stability evaluation are realized. Combining the bandwidth information to evaluate the node processing ability and optimizing the load distribution can respond more flexibly to network state changes, optimize resource utilization. By monitoring the key operation indicators of the device through the Internet of Things technology and comparing them with the preset health standards, the deviation degree and abnormal state of the node can be found in time, and possible failures can be prevented in advance, ensuring the continuity and stability of the network service. By conducting a detailed health assessment of the network nodes and adjusting the network configuration parameters according to the assessment results, activating the standby node to prevent network service interruption, the fault recovery ability is realized, and the reliability and efficiency of the network are improved. Description of the Drawings

[0059] Figure 1 is the system flow chart of the present invention;

[0060] Figure 2 is the flow chart for evaluating the real-time performance of the network nodes of the present invention;

[0061] Figure 3 Flow chart for obtaining network node performance data information in the present invention;

[0062] Figure 4 Flow chart for evaluating the processing capacity of each node in the present invention;

[0063] Figure 5 Flow chart for obtaining the node load optimization result in the present invention;

[0064] Figure 6 Flow chart for evaluating the deviation degree of the operation data of network node devices in the present invention;

[0065] Figure 7 Flow chart for obtaining the node anomaly evaluation result in the present invention;

[0066] Figure 8 Flow chart for obtaining the network anomaly prevention operation result in the present invention. Detailed implementation manners

[0067] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0068] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation of the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0069] Please refer to Figure 1 , the present invention provides a technical solution: a decentralized network node management system based on the Internet of Things, the system includes:

[0070] The network performance evaluation module is based on the decentralized network nodes, collects the delay and packet loss rate data of each network node, evaluates the real-time performance of the network nodes, analyzes the performance data of the network nodes over a period of time, evaluates the stability of the nodes according to the changes in the performance data, and generates network node performance data information;

[0071] The node load optimization module evaluates the processing capacity of each node based on the network node performance data information and combines the bandwidth of each network node, and redistributes the node load according to the current total network load and the load status of each node to obtain the node load optimization result;

[0072] The node status monitoring module, based on the node load optimization result, detects the CPU usage rate, memory usage rate and device operating temperature of the network node device through the Internet of Things, compares the real-time operation data with the preset health standard, evaluates the deviation degree of the network node device operation data, and combines the time since the last maintenance of the node device to evaluate the abnormality degree of the node and generates the node abnormality evaluation result;

[0073] The fault prevention operation module, based on the node abnormality evaluation result, evaluates the flow limiting ratio of the network node, adjusts the network configuration parameters, and activates the standby node to maintain the uninterrupted network service and obtains the network abnormality prevention operation result.

[0074] The network node performance data information includes node response speed, data loss frequency and bandwidth stability indicators. The node load optimization result includes the adjusted node traffic ratio, improved data processing time and optimized network response speed. The node abnormality evaluation result includes health index, parameter types exceeding the standard and operation and maintenance measures to be adjusted. The network abnormality prevention operation result includes the number of activated standby nodes, adjusted communication paths and optimized data transmission frequencies.

[0075] Please refer to Figure 2 , the method for evaluating the real-time performance of network nodes is:

[0076] Based on the decentralized network nodes, collect the delay and packet loss rate data of each network node to obtain the basic network status information;

[0077] Based on the basic network status information, through the formula:

[0078]

[0079] Calculate the performance score of the node, and evaluate the real-time performance of the network node according to the size of the performance score to obtain the network node performance evaluation result;

[0080] Among them, α S represents the weight of the delay time, β S represents the weight of the packet loss rate, d S is the delay time of the node, p S is the packet loss rate of the node, e is the base of the natural logarithm, and S is the performance score of the node.

[0081] Formula:

[0082]

[0083] Detailed Explanation and Acquisition Method of Parameters:

[0084] α S : The weight of the delay time, which is a weight parameter for the impact of the delay time on the performance score of network nodes. The way to obtain this parameter is usually through statistical analysis based on historical data, and the degree of its impact on performance is determined through regression analysis. In practical applications, this weight can be optimized according to the historical network performance dataset to maximize the prediction accuracy of the model for future data.

[0085] β S : The weight of the packet loss rate, which is a weight parameter for the impact of the packet loss rate on the performance score. It is obtained through historical data and statistical analysis, reflecting the relative importance of the packet loss rate in the overall network performance score.

[0086] d S : The delay time of the node, which is directly obtained from network monitoring tools. For example, using the Ping command or more advanced network monitoring systems to record the time from sending data to receiving a response.

[0087] p S : The packet loss rate of the node, which is also calculated through network monitoring tools. Usually, after sending a certain number of packets, the ratio of packets not successfully received is calculated.

[0088] Calculation Example:

[0089] Set the parameters as follows: α S = 0.007, β S = 3, d S = 100ms (delay time), p S = 0.1 (packet loss rate).

[0090] Calculate S:

[0091]

[0092] The result shows that the performance score of this network node is 0.731. This relatively high score indicates that even with a certain amount of delay and packet loss, the overall performance of the node is still evaluated as good.

[0093] Please refer to Figure 3 , the steps for obtaining network node performance data information are:

[0094] Based on the network node performance evaluation results, collect the performance scores of the network node over a period of time to obtain node stability correlation data;

[0095] Based on the node stability correlation data, through the formula:

[0096]

[0097] Calculate the stability score of the computing node;

[0098] where R is the stability score of the node, and n R is the total number of data points, and S i is the i-th performance score; is the average of the performance scores, and σ max is the maximum standard deviation threshold;

[0099] Based on the stability score of the node, integrate the performance scores of the network nodes to obtain the network node performance data information.

[0100] Formula:

[0101]

[0102] Detailed explanation of parameters and acquisition method:

[0103] S i : This is the i-th performance score, obtained by calculation through the previous method.

[0104] Average performance score, calculated as the arithmetic mean of all S within a certain period, that is i This parameter provides an overall performance representation of the network nodes within a time period.

[0105] σ max : The preset maximum standard deviation threshold, used to standardize the stability score R. In network operation and maintenance, this parameter is usually determined based on past performance data and experience, reflecting the upper limit of acceptable performance fluctuations.

[0106] Calculation example:

[0107] Set the following data: n R = 3, S1 = 0.90, S2 = 0.92, S3 = 0.88, σ max = 0.05.

[0108] Calculate the average performance score

[0109]

[0110] Calculate the numerator:

[0111]

[0112] Calculate R:

[0113]

[0114] The calculation results show that the stability score is 0.674, and the performance scores of the nodes are relatively stable, indicating that the performance fluctuations are within an acceptable range.

[0115] Please refer to Figure 4 , the method for evaluating the processing capacity of each node is as follows:

[0116] Based on the network node performance data information, extract the performance score, stability score and bandwidth of each node to obtain the processing capacity correlation data;

[0117] Based on the processing capacity correlation data, through the formula:

[0118]

[0119] Calculate the processing capacity index of the node;

[0120] where S is the performance score of the node, R is the stability score of the node, B is the bandwidth of the node, B max is the maximum bandwidth of all nodes in the network, and P is the processing capacity index of the node;

[0121] Based on the processing capacity index of the node, evaluate the processing performance of the node according to the size of the processing capacity index to obtain the processing performance evaluation result.

[0122] Formula:

[0123]

[0124] Detailed explanation of parameters and acquisition methods:

[0125] S: The performance score of the network node, obtained by calculation in the previous step.

[0126] R: The stability score of the node, obtained by calculation in the previous step.

[0127] B: The bandwidth of the node, directly read from the network management system or the configuration file of the device. Bandwidth represents the maximum data traffic that a network node can handle and is a direct physical parameter for evaluating node performance.

[0128] B max : The maximum bandwidth of all nodes in the network, used to standardize the processing performance index. This value is also statistical data obtained from the network system to ensure that the calculated standardized processing performance index is between 0 and 1.

[0129] Calculation example:

[0130] Set the following parameters: performance score S = 0.90, stability score R = 0.80, node bandwidth B = 100 Mbps, and the maximum bandwidth B max in the network = 200 Mbps.

[0131] The process of calculating the processing performance index P is as follows:

[0132]

[0133]

[0134] P = 0.36

[0135] The meaning of the calculation result is that the processing capacity index of the node under the conditions of its performance score, stability score, and bandwidth is 0.36. This indicates that, relative to the possible maximum bandwidth, the actual processing capacity of this node occupies a certain proportion, but it is not the most efficient. This value helps network administrators understand the relative performance of each node in the overall network, so that more reasonable network load distribution and optimization decisions can be made.

[0136] Please refer to Figure 5 for the steps to obtain the node load optimization result:

[0137] Based on the processing performance evaluation results, collect the processing capacity index of each node and the actual load of the current node, and calculate the total processing capacity of all nodes to obtain load distribution correlation data;

[0138] Based on the load distribution correlation data, through the formula:

[0139]

[0140] Calculate the new load distribution value;

[0141] where L new is the new load distribution value, P is the processing capacity index of the node, L total is the total load of the network, and P total is the sum of the processing capacity indexes of all nodes;

[0142] Based on the new load distribution value, redistribute the node load to obtain the node load optimization result.

[0143] The formula is:

[0144]

[0145] Detailed Explanation and Acquisition Method of Parameters

[0146] P: The processing capacity index of the node, obtained by calculation in the previous steps.

[0147] L total : This is the total load of the network, which is the sum of the loads of all current nodes. It is obtained by collecting the current load of each node from the network management system and summing them up.

[0148] P total : This is the sum of the processing capacity indices of all nodes in the network. It is obtained by adding the processing capacity indices of each node.

[0149] Calculation example:

[0150] Suppose a network consists of three nodes, and the processing capacity indices and current loads of each node are as follows: Node 1: P1 = 0.5, L1 = 40 Mbps; Node 2: P2 = 0.3, L2 = 60 Mbps; Node 3: P3 = 0.2, L3 = 50 Mbps.

[0151] Calculate L total and P total :

[0152] L total = 40 + 60 + 50 = 150 Mbps;

[0153] P total = 0.5 + 0.3 + 0.2 = 1.0;

[0154] Calculate the new load allocation value L of each node new :

[0155]

[0156] The total load is redistributed according to the processing capacity ratio of each node. The load of Node 1 increases because its processing capacity index is relatively high; the loads of Node 2 and Node 3 decrease, in line with their lower processing capacities. The allocation method ensures the effective utilization of network resources and avoids problems of resource overload and idleness.

[0157] Please refer to Figure 6 , the method for evaluating the deviation degree of the operation data of network node devices is:

[0158] Based on the optimization result of node load, the CPU usage rate, memory usage rate, and device operation temperature of network node devices are detected through the Internet of Things to obtain real-time operation data;

[0159] Based on the real-time operation data, compare it with the preset health standard, and through the formula:

[0160]

[0161] Calculate the operation parameter deviation index D to obtain the operation deviation evaluation result:

[0162] Among them, C u is the current CPU usage rate, M u is the current memory usage rate, T u is the current device operating temperature, C s is the healthy standard value of the CPU usage rate, M s is the healthy standard value of the memory usage rate, T s is the healthy standard value of the device operating temperature, w C is the weight of the CPU usage rate, w M is the weight of the memory usage rate, w T is the weight of the temperature.

[0163] Formula:

[0164]

[0165] Detailed parameter explanation and acquisition method:

[0166] C u : The current CPU usage rate, which is directly obtained from the operating system of the network node or the hardware monitoring tool. This is a real-time monitoring value, indicating the current workload percentage of the CPU.

[0167] M u : The current memory usage rate, which is also extracted through the operating system or a dedicated hardware monitoring tool. It represents the percentage of the currently used memory in the total memory.

[0168] T u : The device operating temperature, which is usually provided by a temperature sensor built into the hardware and is in degrees Celsius.

[0169] C s 、M s 、T s : Are the preset healthy standard values, which are obtained based on the device manufacturer's specification sheet or the analysis of long-term operation data. They respectively correspond to the ideal or acceptable maximum values of the CPU usage rate, memory usage rate, and temperature.

[0170] w C 、w M 、w T : The weights reflect the importance of each operating parameter to the overall device health. The weight allocation is based on the device's fault sensitivity and historical fault data, and is obtained through expert evaluation or data analysis.

[0171] Calculation example:

[0172] Suppose there are the following real-time data and standard values: C u = 85%, C s = 50%, M u= 70%, M s = 50%, T u = 75°C, T s = 60°C, the weight is set to w C = 0.5, w M = 0.3, w T = 0.2.

[0173] Calculate the absolute value of the deviation of each parameter: |C u - C s | = |0.85 - 0.5| = 0.35, |M u - M s | = |0.7 - 0.5| = 0.2, |T u - T s | = |75 - 60| = 15°C.

[0174] Calculate the operating parameter deviation index D:

[0175]

[0176] The calculation result D = 3.235 indicates that there is a certain deviation between the current operating state of the device and the preset health standard. This value can help the maintenance team prioritize the devices that may face problems such as overloading or overheating. A high value of this index indicates that the device needs to be inspected or maintained to prevent potential failures.

[0177] Please refer to Figure 7 , the steps to obtain the node anomaly assessment result are

[0178] Based on the operating deviation assessment result, according to the maintenance records of the network node devices, extract the time since the last maintenance of the node devices to obtain the anomaly correlation data;

[0179] Based on the anomaly correlation data, through the formula:

[0180]

[0181] Calculate the anomaly index;

[0182] Among them, β E is the adjustment coefficient, θ E is the standard number of days threshold for the device maintenance cycle, T E is the time since the last maintenance, D is the operating parameter deviation index, e is the base of the natural logarithm, and E is the anomaly index;

[0183] Based on the anomaly index, compare it with the preset anomaly threshold to determine whether the network node device is abnormal and obtain the node anomaly assessment result.

[0184] Formula:

[0185]

[0186] Detailed Explanation and Acquisition Method of Parameters:

[0187] D: Operating parameter deviation index, obtained through previous steps.

[0188] T E : Represents the time since the last maintenance, usually obtained from the device's maintenance record database. This time shows the total number of days elapsed since the device last underwent inspection or maintenance.

[0189] β E : Adjustment coefficient, which is an empirical value set according to the device type and expected lifespan, used to adjust the influence intensity of the time factor in anomaly assessment.

[0190] θ E : Set time threshold, representing the standard interval days for device maintenance or inspection. This is usually determined based on the manufacturer's recommendations or historical maintenance data.

[0191] Calculation Example:

[0192] Set the operating parameter deviation index D of the network node to 15, the time T since the last maintenance E = 180 days, the set time sensitivity adjustment coefficient β E = 0.03, and the time threshold θ E = 90 days.

[0193] Calculate the time adjustment factor:

[0194] β E ×(T E -θ E ) = 0.03 × (180 - 90) = 0.03 × 90 = 2.7;

[0195] Calculate the anomaly index E:

[0196]

[0197] The calculated anomaly index E ≈ 14.05 indicates that there is an obvious anomaly in this node. This value is relatively high, indicating that the device has deviated significantly from the normal operating state since the last maintenance.

[0198] Please refer to Figure 8 , and the acquisition steps for the results of network anomaly prevention operations are as follows:

[0199] Based on the node anomaly assessment results, through the formula:

[0200]

[0201] Calculate the bandwidth ratio after current limiting;

[0202] Among them, E is the anomaly index, and E max is the historical maximum anomaly index of similar devices, and α Z is the adjustment coefficient, and Z is the bandwidth ratio after current limiting;

[0203] Based on the bandwidth ratio after current limiting, adjust the network configuration parameters and activate the standby node to maintain the uninterrupted network service, and obtain the network anomaly prevention operation result.

[0204] Formula:

[0205]

[0206] Detailed parameter explanations and acquisition methods:

[0207] E: Anomaly index, obtained by calculation in the previous steps.

[0208] E max : The historical maximum anomaly index of similar devices, usually determined by the performance records of the devices, and used as a benchmark for the degree of anomaly.

[0209] α Z : Adjustment coefficient, used to adjust the influence of the anomaly index in the logarithmic calculation. The setting of the coefficient is based on the fault response sensitivity of the device and historical data analysis.

[0210] Calculation example:

[0211] Set the anomaly index E of the network node to 10, the historical maximum anomaly index E max = 50, and the adjustment coefficient α Z = 9.

[0212] Calculate the ratio of the anomaly index to the maximum anomaly index and multiply by the adjustment coefficient:

[0213]

[0214] Calculate the current limiting ratio Z:

[0215]

[0216] The calculated bandwidth ratio Z after current limiting is approximately 0.5528, indicating that the data processing capacity of the node needs to be significantly reduced to approximately 55.28% of the normal capacity. This adjustment is based on the current abnormal state of the node, aiming to relieve the pressure on the node and prevent further problems or failures that may be caused by overload. In this way, the stability and reliability of the network can be protected, and at the same time, the negative impact on the entire network caused by the abnormal state of a single node can be avoided.

[0217] The above are only the preferred embodiments of the present invention and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A decentralized network node management system based on the Internet of Things, characterized in that: The system comprises: The network performance evaluation module is based on distributed network nodes, collects the delay and packet loss rate data of each network node, evaluates the real-time performance of the network node, and analyzes the performance data of the network node over a period of time. According to the changes in performance data, it evaluates the stability of the node and generates network node performance data information; The node load optimization module evaluates the processing capacity of each node based on the network node performance data information and the bandwidth of each network node, and redistributes the node load according to the current total network load and the load status of each node to obtain the node load optimization result; Based on the node load optimization result, the node status monitoring module detects the CPU usage, memory usage and device operating temperature of the network node device through the Internet of Things, compares the real-time operating data with the preset health standard, evaluates the degree of deviation of the network node device operating data, and evaluates the abnormality of the node in combination with the time since the last maintenance of the node device, and generates a node abnormality evaluation result; The fault prevention operation module evaluates the current limiting ratio of the network node based on the node anomaly evaluation result, adjusts the network configuration parameters, and activates the standby node to maintain the network service without interruption, thereby obtaining the network anomaly prevention operation result.

2. The decentralized network node management system based on the Internet of Things according to claim 1 is characterized in that: The method for evaluating the real-time performance of a network node is: Based on distributed network nodes, the delay and packet loss rate data of each network node are collected to obtain basic network status information; Based on the basic network status information, through the formula: Calculate the performance score of the node, evaluate the real-time performance of the network node according to the size of the performance score, and obtain the network node performance evaluation result; Among them, α S represents the weight of the delay time, β S Represents the weight of the packet loss rate, d S is the node delay time, p S is the packet loss rate of the node, e is the base of the natural logarithm, and S is the performance score of the node.

3. The decentralized network node management system based on the Internet of Things according to claim 2 is characterized in that: The steps for obtaining the network node performance data information are as follows: Based on the network node performance evaluation results, the performance scores of the network nodes over a period of time are collected to obtain node stability related data; Based on the node stability association data, the formula: Calculate the stability score of the node; Among them, R is the stability score of the node, n R is the total number of data points, S i is the i-th performance score; is the mean value of the performance score, σ max is the maximum standard deviation threshold; Based on the stability scores of the nodes, the performance scores of the network nodes are integrated to obtain network node performance data information.

4. The decentralized network node management system based on the Internet of Things according to claim 1, characterized in that: The method for evaluating the processing capability of each node is: Based on the network node performance data information, extract the performance score, stability score and bandwidth of each node to obtain processing capacity related data; Based on the processing capability correlation data, by the formula: The processing power index of the computing node; Among them, S is the performance score of the node, R is the stability score of the node, B is the bandwidth of the node, and B max It is the maximum bandwidth of all nodes in the network, and P is the processing capacity index of the node; Based on the processing capability index of the node, the processing performance of the node is evaluated according to the size of the processing capability index to obtain a processing performance evaluation result.

5. The decentralized network node management system based on the Internet of Things according to claim 4 is characterized in that: The steps for obtaining the node load optimization result are: Based on the processing performance evaluation result, the processing capacity index of each node and the actual load of the current node are collected, and the sum of the processing capacities of all nodes is calculated to obtain load distribution related data; Based on the load distribution association data, by formula: Calculate new load distribution values; Among them, L new is the new load distribution value, P is the processing capacity index of the node, L total is the total load of the network, P total is the sum of the processing power indexes of all nodes; Based on the new load distribution value, the node load is redistributed to obtain a node load optimization result.

6. The decentralized network node management system based on the Internet of Things according to claim 1, characterized in that: The method for evaluating the degree of deviation of network node device operation data is: Based on the node load optimization result, the CPU usage rate, memory usage rate and device operating temperature of the network node device are detected through the Internet of Things to obtain real-time operating data; Based on the real-time operating data, compared with the preset health standards, the formula is: Calculate the operating parameter deviation index D and obtain the operating deviation evaluation result: Among them, C u is the current CPU usage, M u is the current memory usage, T u is the current device operating temperature, C s is the healthy standard value of CPU usage, M s is the healthy standard value of memory usage, T s is the healthy standard value of the equipment operating temperature, w C is the weight of CPU usage, w M is the weight of memory usage, w T is the weight of temperature.

7. The decentralized network node management system based on the Internet of Things according to claim 6, characterized in that: The steps for obtaining the node abnormality evaluation result are as follows: Based on the operation deviation assessment result, according to the maintenance record of the network node device, the time from the last maintenance of the node device is extracted to obtain abnormal correlation data; Based on the abnormal correlation data, the formula: Calculate the anomaly index; Among them, β E is the adjustment coefficient, θ E is the standard number of days threshold for equipment maintenance cycle, T E is the time since the last maintenance, D is the operating parameter deviation index, e is the base of the natural logarithm, and E is the abnormality index; Based on the abnormality index, it is compared with the preset abnormality threshold to determine whether the network node device is abnormal, and obtain the node abnormality evaluation result.

8. The decentralized network node management system based on the Internet of Things according to claim 1, characterized in that: The steps for obtaining the network anomaly prevention operation result are: Based on the node anomaly evaluation result, the formula is: Calculate the bandwidth ratio after current limiting; Among them, E is the abnormal index, E max is the historical maximum abnormal index of similar equipment, α Z is the adjustment coefficient, Z is the bandwidth ratio after current limiting; Based on the bandwidth ratio after the current limiting, the network configuration parameters are adjusted, and the standby nodes are activated to maintain the network service without interruption, thereby obtaining the network anomaly prevention operation result.

Images