A network monitoring method, device and storage medium
By dividing the network into subnets and extracting feature information using principal component analysis method, and combining preset monitoring models for network monitoring, the problems of low network monitoring efficiency and poor accuracy in the existing technology are solved, and more efficient network status recognition is achieved and fault location.
Patent Information
- Application Number
- CN202110993998.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-08-27
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-08-27
AI Technical Summary
During the existing network management process, it is difficult for the data acquisition module to quickly and accurately understand the current status of the network and fault location, resulting in low monitoring efficiency and poor accuracy.
By obtaining the link information of the network to be monitored, the network feature information is extracted from the network situation information and performance data using principal component analysis method, and network monitoring is carried out in combination with preset monitoring models to optimize information to improve monitoring accuracy and efficiency.
It improves the accuracy and efficiency of network monitoring, can quickly identify abnormalities and output alarm information, reduce false alarms, and enhance fault location capabilities.
Smart Images

Figure CN115733772B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of communication technology, and in particular to a network monitoring method, device and storage medium. Background Art
[0002] With the development of network technology, the network scale is growing, network equipment is constantly updated and new services are constantly emerging, which makes the reliability of the network face greater challenges.
[0003] At present, in the general network management process, the data acquisition module directly displays the collected equipment performance data in a graphical manner through the management system. It is difficult to quickly and accurately understand the network status and fault location from the massive amount of monitoring data, resulting in low efficiency and poor accuracy in network monitoring. Summary of the invention
[0004] To solve the above technical problems, the embodiments of the present invention hope to provide a network monitoring method, device and storage medium, which utilize characteristic information characterizing the network performance of the subnet and combine the corresponding monitoring model of the subnet to perform network monitoring, thereby improving the accuracy and efficiency of network monitoring.
[0005] The technical solution of the present invention is achieved in this way:
[0006] The present invention provides a network detection method, the method comprising:
[0007] Acquire a network to be monitored, and divide the network to be monitored into multiple subnets based on link information of the network to be monitored;
[0008] Obtaining network status information and network performance data of a target subnet at a current moment; the target subnet is any one of the multiple subnets;
[0009] Selecting current network characteristic information of the target subnet from the network status information and the network performance data;
[0010] The preset monitoring model corresponding to the target subnet is used to perform network monitoring on the target subnet based on the current network feature information.
[0011] In the above method, before performing network monitoring on the target subnet using the preset monitoring model corresponding to the target subnet based on the current network feature information, the method further includes:
[0012] The current network characteristic information is optimized using a preset optimized network model.
[0013] In the above method, the network status information includes at least one device parameter corresponding to the network device in the target subnet, the network performance data includes at least one performance parameter corresponding to the network device in the target subnet, and selecting the current network characteristic information of the target subnet from the network status information and the network performance data includes:
[0014] Determine the component information corresponding to each parameter of the at least one equipment parameter and the at least one performance parameter according to the principal component analysis method to obtain a plurality of component information, and determine the sum of the plurality of component information as the total component information;
[0015] Determine a ratio of each component information in the plurality of component information to the total component information as a contribution rate, and obtain a plurality of component contribution rates corresponding to the plurality of component information, wherein each component information corresponds to a component contribution rate;
[0016] From the multiple component contribution rates, select component contribution rates in order from large to small, until the sum of at least one selected component contribution rate exceeds a preset contribution rate threshold;
[0017] At least one component information corresponding to at least one selected component contribution rate is obtained from the multiple component information, and at least one parameter corresponding to the at least one component information among the at least one device parameter and the at least one performance parameter is determined as the current network characteristic information.
[0018] In the above method, before selecting the component contribution rates from the multiple component contribution rates in descending order, the method further includes:
[0019] Obtaining the network scenario of the target subnet;
[0020] Based on the network scenario, selecting at least one component contribution rate from the multiple component contribution rates;
[0021] Among the at least one component contribution rate, the component contribution rate of the corresponding component information that plays a positive role in the network scenario is increased, and / or the component contribution rate of the corresponding component information that plays a negative role in the network scenario is reduced.
[0022] In the above method, the using the preset monitoring model to perform network monitoring on the target subnet based on the current network feature information includes:
[0023] Obtaining historical network feature information of the target subnet from a preset time to the current time;
[0024] Determine the network monitoring items corresponding to the target subnet by using the preset monitoring model based on the historical network feature information and the current network feature information;
[0025] By using the preset monitoring model, when the normal deviation of the network monitoring item is greater than the preset normal deviation threshold, the abnormal alarm information corresponding to the target subnet is output.
[0026] In the above method, the method further comprises:
[0027] Obtaining network link information of the target subnet, and generating a target topology map using the network link information;
[0028] The historical network feature information, the current network feature information and the abnormal alarm information are superimposed on the target topology map to obtain a superimposed topology map.
[0029] In the above method, after outputting the abnormal alarm information corresponding to the target subnet, the method further includes:
[0030] The abnormal alarm information is used to optimize the preset monitoring model.
[0031] The present invention provides a network monitoring device, comprising:
[0032] A division module, used for acquiring a network to be monitored, and dividing the network to be monitored into a plurality of subnets based on link information of the network to be monitored;
[0033] An acquisition module, used to acquire network status information and network performance data of a target subnet at a current moment; the target subnet is any one of the multiple subnets;
[0034] A selection module, used to select the current network characteristic information of the target subnet from the network status information and the network performance data;
[0035] The monitoring module is used to perform network monitoring on the target subnet based on the current network feature information by using a preset monitoring model corresponding to the target subnet.
[0036] In the above device, the monitoring module is also used to optimize the current network feature information using a preset optimized network model.
[0037] In the above-mentioned device, the selection module is specifically used to determine the component information corresponding to each parameter in the at least one device parameter and the at least one performance parameter according to the principal component analysis method, obtain multiple component information, and determine the sum of the multiple component information as the total component information; determine the ratio of each component information in the multiple component information to the total component information as a contribution rate, and obtain multiple component contribution rates corresponding to the multiple component information, wherein each component information corresponds to a component contribution rate; from the multiple component contribution rates, select the component contribution rates in order from large to small until the sum of the selected at least one component contribution rate exceeds the preset contribution rate threshold; obtain at least one component information corresponding to the selected at least one component contribution rate from the multiple component information, and determine at least one parameter corresponding to the at least one component information in the at least one device parameter and the at least one performance parameter as the current network characteristic information.
[0038] The above-mentioned device also includes an adjustment module, which is also used to obtain the network scenario of the target subnet; based on the network scenario, select at least one component contribution rate from the multiple component contribution rates; among the at least one component contribution rate, increase the component contribution rate of the corresponding component information that plays a positive role in the network scenario, and / or reduce the component contribution rate of the corresponding component information that plays a negative role in the network scenario.
[0039] In the above-mentioned device, the monitoring module is specifically used to obtain the historical network characteristic information of the target subnet between a preset moment and the current moment; using the preset monitoring model, based on the historical network characteristic information and the current network characteristic information, determine the network monitoring item corresponding to the target subnet; using the preset monitoring model, when the normal deviation of the network monitoring item is greater than a preset normal deviation threshold, output the abnormal alarm information corresponding to the target subnet.
[0040] The above-mentioned device also includes an overlay module, which is used to obtain the network link information of the target subnet and generate a target topology map using the network link information; the historical network feature information, the current network feature information and the abnormal alarm information are overlaid onto the target topology map to obtain a overlaid topology map.
[0041] The above device also includes an optimization module for optimizing the preset monitoring model using the abnormal alarm information.
[0042] The present invention provides a network monitoring device, comprising: a processor, a memory and a communication bus;
[0043] The communication bus is used to realize the communication connection between the processor and the memory;
[0044] The processor is used to execute the network monitoring program stored in the memory to implement the above network monitoring method.
[0045] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned network monitoring method.
[0046] The present invention provides a network monitoring method, device and storage medium, the method comprising: obtaining a network to be monitored, and dividing the network to be monitored into multiple subnets based on the link information of the network to be monitored; obtaining the network status information and network performance data of the target subnet at the current moment; selecting the current network feature information of the target subnet from the network status information and network performance data; using the preset monitoring model corresponding to the target subnet, based on the current network feature information, to perform network monitoring on the target subnet. The technical solution provided by the present invention uses the feature information that characterizes the network performance of the subnet, combined with the corresponding monitoring model of the subnet to perform network monitoring, thereby improving the accuracy and efficiency of network monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 A schematic diagram of a flow chart of a network monitoring method provided by an embodiment of the present invention;
[0048] Figure 2 A schematic diagram of an exemplary visualization interface provided by an embodiment of the present invention;
[0049] Figure 3 A schematic diagram of a network monitoring device provided by an embodiment of the present invention Figure 1 ;
[0050] Figure 4 A schematic diagram of a network monitoring device provided by an embodiment of the present invention Figure 2 . DETAILED DESCRIPTION
[0051] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention. It is to be understood that the specific embodiments described herein are only used to explain the relevant application, rather than to limit the application. It should also be noted that, for ease of description, only the parts related to the relevant application are shown in the accompanying drawings.
[0052] The present invention provides a network monitoring method, which is applied to a network monitoring device. Figure 1 A schematic diagram of a network monitoring method provided by an embodiment of the present invention. Figure 1 As shown, it mainly includes the following steps:
[0053] S101: Acquire a network to be monitored, and divide the network to be monitored into multiple subnets based on link information of the network to be monitored.
[0054] In an embodiment of the present invention, the network monitoring device obtains a network to be monitored, and divides the network to be monitored into a plurality of subnets based on link information of the network to be monitored.
[0055] It should be noted that in the embodiments of the present invention, for two links with different numbers and types of devices, if the network performance is monitored for the entire network according to a single unified measurement standard, it is easy to cause deviations in the alarm information. Therefore, the present invention divides the entire network into multiple subnets according to the link information, and sets corresponding alarm standards for each of the multiple subnets, thereby improving the accuracy of network monitoring.
[0056] S102: Obtain network status information and network performance data of the target subnet at the current moment.
[0057] In the embodiment of the present invention, the network monitoring device can directly obtain the network status information and network performance data of the target subnet at the current moment.
[0058] It should be noted that, in the embodiment of the present invention, the target subnet is any subnet among multiple subnets included in a complete network.
[0059] It should be noted that, in an embodiment of the present invention, network status information may include the number of network devices in the target subnet, the device type of the network device, the device name, the device IP address, the device description, or one or more network device-related information in the device details; network performance data may include the channel utilization rate, inflow rate, delay, jitter, packet loss, port traffic statistics, device availability, other service level agreement (SLA) quality parameters, or CPU utilization of the network device, one or more data characterizing network performance.
[0060] S103: Select current network characteristic information of the target subnet from the network status information and network performance data.
[0061] In an embodiment of the present invention, the network monitoring device selects the current network characteristic information of the target subnet from the network status information and the network performance data.
[0062] Specifically, in an embodiment of the present invention, the network status information includes at least one device parameter corresponding to the network device in the target subnet, the network performance data includes at least one performance parameter corresponding to the network device in the target subnet, and the network monitoring device selects the current network characteristic information of the target subnet from the network status information and the network performance data, including: determining the component information corresponding to each parameter in at least one device parameter and at least one performance parameter according to the principal component analysis method, obtaining multiple component information, and determining the sum of the multiple component information as the total component information; determining the ratio of each component information in the multiple component information to the total component information as a contribution rate, and obtaining multiple component contribution rates corresponding to the multiple component information, wherein each component information corresponds to a component contribution rate; from the multiple component contribution rates, selecting the component contribution rates in order from large to small until the sum of the selected at least one component contribution rate exceeds the preset contribution rate threshold; obtaining at least one component information corresponding to the selected at least one component contribution rate from the multiple component information, and determining at least one parameter corresponding to the at least one component information in the at least one device parameter and the at least one performance parameter as the current network characteristic information.
[0063] It should be noted that, in the embodiment of the present invention, the network monitoring device uses the principal component analysis method to determine the component information corresponding to each parameter in at least one device parameter included in the network status information and at least one performance parameter included in the network performance data, and the specific determination method is:
[0064] The data corresponding to at least one device parameter included in the network status information are combined into a status matrix A1, the data corresponding to at least one performance parameter included in the network performance data are combined into a performance matrix A2, and the status matrix A1 and the performance matrix A2 are combined into a matrix A, where Σ is the eigenvalues λ1, λ2, λ3, ..., λ of the covariance matrix of the matrix A p , and λ1≥λ2≥...≥λ p ≥0, and the orthogonal normalized eigenvectors e1,e2,...,e corresponding to the eigenvalues p , where p is the number of parameters included in the matrix A.
[0065] For example, if the situation matrix A1 includes the number of network devices A 11 and device type A of network devices 12 ; Performance matrix A2 includes the channel utilization of network devices A 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 , packet loss A 25 , and CPU utilization A 26 , then p is the number of parameters included in the matrix A at this time, which is 8.
[0066] Then, the principal component analysis formula can be used to find the component information corresponding to the 8 parameters included in the matrix A. The specific formula is as follows:
[0067]
[0068] Among them, Y represents the principal component information, i is the i-th parameter represented by the matrix A, A1 is the situation matrix, A2 is the performance matrix, e i1 and e i2 is the unit eigenvector.
[0069] It should be noted that, in the embodiments of the present invention, the contribution rate is the ratio of a certain component information to the total component information. In fact, the network monitoring device calculates the ratio of each component information in the multiple component information to the total component information to determine a contribution rate. Each component information corresponds to a component contribution rate. Then, from the multiple component contribution rates, they are selected in turn according to the size of the contribution rate until the sum of at least one selected component contribution rate reaches a preset contribution rate threshold. Then, the selection is stopped to obtain at least one component information corresponding to at least one component contribution rate, and then the parameters corresponding to the at least one component information are obtained from at least one device parameter and at least one performance parameter, and determined as the current network characteristic information.
[0070] It should be noted that, in the embodiments of the present invention, the preset contribution rate threshold may be preset according to actual needs and application scenarios. For example, the preset contribution rate threshold may be 80% or 90%, and the present invention is not limited to this.
[0071] Specifically, in an embodiment of the present invention, before the network monitoring device selects component contribution rates from multiple component contribution rates in order from large to small, it can also perform the following steps: obtain the network scenario of the target subnet; based on the network scenario, select at least one component contribution rate from multiple component contribution rates; increase the component contribution rate of the at least one component contribution rate, whose corresponding component information plays a positive role in the network scenario, and / or reduce the component contribution rate of the corresponding component information that plays a negative role in the network scenario.
[0072] It should be noted that in an embodiment of the present invention, the network monitoring device can obtain information characterizing network scenarios, such as high-precision, high-speed, high-density, or peak-hour residential areas, and determine the specific network scenario based on the corresponding information characterizing the network scenario.
[0073] It should be noted that, in the embodiment of the present invention, the network monitoring device can select at least one component contribution rate according to the network scenario in which the target subnet is located. For example, when the target subnet is in a high-precision network scenario, the packet loss A25 The change of can affect the network performance of the target subnet to a greater extent, that is, in this scenario, the packet loss A 25 Play a positive role, need to lose packet A 25 The component contribution rate is increased to make the current network feature information selected based on the component contribution rate more accurate. If the target subnet is in a high-precision network scenario, the channel utilization rate A is selected. 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 At least one component contribution rate is equal to, because in this scenario, the channel utilization rate A 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 It has a negative effect and needs to be adjusted for channel utilization A 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 The contribution rate of the component is reduced; in the case of a high-speed network scenario where the target subnet is in a high-speed network, the inflow rate A 21 The requirements are relatively high, and the inflow rate A needs to be adjusted 22 The component contribution rate of the channel is used to make the current network feature information selected based on the component contribution rate more accurate and to improve the channel utilization rate A. 21 、Packet loss A 25 , Delay A 23 , Jitter A 24 etc., and at least one component contribution rate is reduced; for different network scenarios, the network monitoring device can select different component contribution rates from multiple component contribution rates.
[0074] For example, in a high-precision network scenario, if the network monitoring device selects packet loss A 25 Component contribution rate, then we need to 25 The component contribution rate is increased, and if the network monitoring device selects the channel utilization rate A 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 If at least one component contribution rate is equal to , then the channel utilization rate A needs to be 21 , inflow rate A 22 , Delay A 23 , Jitter A 24 The contribution rate of at least one component is reduced, and the specific adjustment method is not limited in the present invention.
[0075] S104: Using a preset monitoring model corresponding to the target subnet and based on current network feature information, perform network monitoring on the target subnet.
[0076] In an embodiment of the present invention, the network monitoring device uses a preset monitoring model corresponding to the target subnet and performs network monitoring on the target subnet based on current network feature information.
[0077] It should be noted that, in the embodiment of the present invention, the preset monitoring model may be an autoregressive model.
[0078] Specifically, in an embodiment of the present invention, the network monitoring device uses the preset monitoring model corresponding to the target subnet, and based on the current network feature information, before performing network monitoring on the target subnet, it can also perform the following steps: using the preset optimized network model to optimize the current network feature information.
[0079] It should be noted that, in the embodiment of the present invention, the preset optimization network model may be a model for machine learning based on a BP neural network (Back Propagation Neural Network).
[0080] It should be noted that in an embodiment of the present invention, the network monitoring device can first obtain network feature sample information, and then use the network feature sample information to perform data optimization training on a preset optimized network model, and determine the model obtained after the training as the optimized network model, thereby using the optimized network model to optimize the current network feature information. Correspondingly, using the preset monitoring model corresponding to the target subnet, based on the current network feature information, the current network feature information on which the network monitoring of the target subnet is based is the optimized current network feature information.
[0081] Specifically, in an embodiment of the present invention, the network monitoring device uses a preset monitoring model to perform network monitoring on the target subnet based on current network feature information, including: obtaining historical network feature information of the target subnet from a preset moment to the current moment; using the preset monitoring model to determine the network monitoring item corresponding to the target subnet based on the historical network feature information and the current network feature information; using the preset monitoring model to output abnormal alarm information corresponding to the target subnet when the normal deviation of the network monitoring item is greater than a preset normal deviation threshold.
[0082] It should be noted that, in the embodiments of the present invention, the preset time can be set according to actual needs and application scenarios, for example, it can be 5 minutes before the current time, 10 minutes before the current time, and the present invention does not limit the specific preset time.
[0083] It should be noted that, in the embodiment of the present invention, the network monitoring device obtains the historical network feature information of the target subnet from the preset time to the current time, and the current network feature information, and inputs it into the preset monitoring model. The preset monitoring model is:
[0084] X=ω1X t-1 +ω2X t-2 +…+ω m X t-m +e t (2)
[0085] Among them, X is the current network feature information, t is the current time, t-1 is the time before the current time, and then, t-2, ······, tm and so on, X t-1 ~X t-m is the network feature information before the current moment, X t-1 is the network feature information of the moment before the current moment, X t-m is the network characteristic information at the preset time, ω1, ω2, ..., ω m is the influence strength of the corresponding moment on the current moment, which is preset information, e t This is a network monitoring item.
[0086] It should be noted that, in the embodiment of the present invention, if X is the current optimized feature information, then X t-1 ~X t-m Optimize feature information for the history before the current moment, X t-1 is the optimized feature information of the moment before the current moment, X t-m It is optimized characteristic information at a preset time, and of course, it can also be unoptimized network characteristic information.
[0087] It should be noted that, in the embodiment of the present invention, the network monitoring device monitors the network monitoring items e t To determine whether there is a network anomaly, under normal circumstances, the network monitoring item e t It shows a normal distribution. When the network characteristic information is abnormal, the network monitoring item e t There will be normal deviation.
[0088] It should be noted that, in the embodiment of the present invention, for different subnets, the network monitoring device may allow network monitoring items e t There is a certain deviation, and in the network monitoring item e t An abnormal alarm will only occur when the allowable deviation is exceeded.
[0089] It should be noted that, in the embodiment of the present invention, the preset normal deviation threshold value can be preset by the network monitoring device according to the application scenario and actual needs. t Compare with the preset normal deviation threshold to determine whether to output abnormal alarm information.
[0090] It should be noted that in an embodiment of the present invention, when the monitoring personnel receive abnormal alarm information, they will check the abnormal alarm information. If the normal deviation is greater than the preset normal deviation threshold, and the monitoring personnel believes that the normal deviation is within the deviation range that the subnet can tolerate, then the monitoring personnel will increase the preset normal deviation threshold according to the specific situation, so as to avoid false alarms in similar situations.
[0091] Specifically, in an embodiment of the present invention, after the network monitoring device outputs the abnormal alarm information corresponding to the target subnet, the following steps may be performed: optimizing a preset monitoring model using the abnormal alarm information.
[0092] It should be noted that in an embodiment of the present invention, after outputting abnormal alarm information, the network monitoring device can input the abnormal alarm information and the processing method for the abnormal alarm information into a preset monitoring model to optimize the preset monitoring model and improve the accuracy of network monitoring.
[0093] Specifically, in an embodiment of the present invention, the network monitoring device obtains network link information of the target subnet, and uses the network link information to generate a target topology map; historical network feature information, current network feature information and abnormal alarm information are superimposed on the target topology map to obtain a superimposed topology map.
[0094] It should be noted that in an embodiment of the present invention, the network monitoring device associates historical network feature information, current network feature information and abnormal alarm information with the device identification information of at least one network device included in the target subnet, obtains the link information corresponding to each network device in the at least one network device, historical network feature information, current network feature information, and abnormal alarm information, and superimposes the information corresponding to each network device into the target topology map.
[0095] It should be noted that, in the embodiment of the present invention, the network monitoring device may also superimpose the historical optimization feature information and the current optimization feature information into the target topology map accordingly.
[0096] It should be noted that, in an embodiment of the present invention, the network monitoring device can set different colors for the corresponding network device identification according to the abnormal alarm information output by the preset monitoring model to indicate the status information of the network device. For example, if the status information of a certain network device is seriously abnormal, then the corresponding network device identification can be colored with red to indicate that the status information of the network device is slightly abnormal. Of course, other types of representation methods can also be set at the identification position of the network device. For example, the network monitoring device can add different numbers of five-pointed stars, triangles or other shapes at the position of the network device identification corresponding to the network device, and the number represents the status information of the network device. The specific setting method can be set according to actual needs and application scenarios.
[0097] It should be noted that in the embodiments of the present invention, due to the large number of network events, for recurring abnormal alarm information, time aggregation can be used to record the number of occurrences and the start time. For related abnormal alarm information, aggregation is performed according to the time intervals at which the abnormal alarm information appears, and the aggregated information is used to generate an impact relationship tree, thereby determining the fault relationship diagram of the network device and the impact range of the fault. The impact relationship tree can also be superimposed on the target topology diagram, which can improve the efficiency of fault location.
[0098] Figure 2 A schematic diagram of an exemplary visualization interface provided by an embodiment of the present invention. Figure 2 As shown, the visualization interface displays the superimposed topology map, which is the target topology map corresponding to the target subnet. The target topology map includes network device 1, network device 2 and network device 3, wherein the relevant information of network device 1 is specifically presented, such as: abnormal alarm information corresponding to network device 1, performance visualization map, abnormal alarm identification information of network devices, etc. The network monitoring device can display the corresponding information according to different needs. Figure 2 In the example, the status information of the network device 1 is a network device corresponding to the network device 1 marked with different colors to distinguish the status of the network device 1, such as Figure 2 As shown, the network device identification of network device 1 can be green, yellow, and red. If the color corresponding to the network device identification of network device 1 is green, it means that network device 1 is in a normal state. If the color corresponding to the network device identification of network device 1 is yellow, it means that network device 1 is in a performance warning state. If the color corresponding to the network device identification of network device 1 is red, it means that network device 1 is in a serious abnormal state. Port A of network device 1 can be network device 2, and port B of network device 1 can be network device 3.
[0099] The present invention provides a network monitoring method, the method comprising: obtaining a network to be monitored, and dividing the network to be monitored into multiple subnets based on link information of the network to be monitored; obtaining network status information and network performance data of a target subnet at the current moment; selecting current network feature information of the target subnet from the network status information and network performance data; and using a preset monitoring model corresponding to the target subnet to perform network monitoring on the target subnet based on the current network feature information. The method provided by the present invention uses feature information characterizing the network performance of a subnet and combines the corresponding monitoring model of the subnet to perform network monitoring, thereby improving the accuracy and efficiency of network monitoring.
[0100] The present invention provides a network monitoring device. Figure 3 A schematic diagram of a network monitoring device provided by an embodiment of the present invention Figure 1 .like Figure 3 As shown, including:
[0101] A division module 301 is used to obtain a network to be monitored and divide the network to be monitored into multiple subnets based on link information of the network to be monitored;
[0102] The acquisition module 302 is used to acquire the network status information and network performance data of the target subnet at the current moment; the target subnet is any subnet among the multiple subnets;
[0103] A selection module 303 is used to select the current network characteristic information of the target subnet from the network status information and the network performance data;
[0104] The monitoring module 304 is used to perform network monitoring on the target subnet based on the current network feature information by using a preset monitoring model corresponding to the target subnet.
[0105] Optionally, the monitoring module 304 is further configured to optimize the current network feature information using a preset optimized network model.
[0106] Optionally, the selection module 302 is specifically used to determine the component information corresponding to each parameter in the at least one device parameter and the at least one performance parameter according to the principal component analysis method, obtain multiple component information, and determine the sum of the multiple component information as the total component information; determine the ratio of each component information in the multiple component information to the total component information as a contribution rate, and obtain multiple component contribution rates corresponding to the multiple component information, wherein each component information corresponds to a component contribution rate; from the multiple component contribution rates, select the component contribution rates in order from large to small until the sum of the selected at least one component contribution rate exceeds a preset contribution rate threshold; obtain at least one component information corresponding to the selected at least one component contribution rate from the multiple component information, and determine at least one parameter corresponding to the at least one component information in the at least one device parameter and the at least one performance parameter as the current network characteristic information.
[0107] Optionally, the network detection device also includes an adjustment module (not shown in the figure), which is used to obtain the network scenario of the target subnet; based on the network scenario, select at least one component contribution rate from the multiple component contribution rates; among the at least one component contribution rate, increase the component contribution rate of the corresponding component information that plays a positive role in the network scenario, and / or reduce the component contribution rate of the corresponding component information that plays a negative role in the network scenario.
[0108] Optionally, the monitoring module 304 is specifically used to obtain historical network characteristic information of the target subnet between a preset moment and the current moment; using the preset monitoring model, determine the network monitoring item corresponding to the target subnet based on the historical network characteristic information and the current network characteristic information; using the preset monitoring model, when the normal deviation of the network monitoring item is greater than a preset normal deviation threshold, output abnormal alarm information corresponding to the target subnet.
[0109] Optionally, the network detection device also includes a superposition module (not shown in the figure), which is used to obtain network link information of the target subnet and generate a target topology map using the network link information; superimpose the historical network feature information, the current network feature information and the abnormal alarm information onto the target topology map to obtain a superimposed topology map.
[0110] Optionally, the network detection device further includes an optimization module (not shown in the figure) for optimizing the preset monitoring model using the abnormal alarm information.
[0111] The present invention provides a network monitoring device. Figure 4 A schematic diagram of a network monitoring device provided by an embodiment of the present invention Figure 2 .like Figure 4 As shown, the network monitoring device includes: a processor 401, a memory 402 and a communication bus 403;
[0112] The communication bus 403 is used to realize the communication connection between the processor 401 and the memory 402;
[0113] The processor 401 is used to execute the network monitoring program stored in the memory 402 to implement the above network monitoring method.
[0114] The present invention provides a network monitoring device, which obtains a network to be monitored, and divides the network to be monitored into multiple subnets based on the link information of the network to be monitored; obtains the network status information and network performance data of the target subnet at the current moment; selects the current network feature information of the target subnet from the network status information and the network performance data; and uses the preset monitoring model corresponding to the target subnet to perform network monitoring on the target subnet based on the current network feature information. The device provided by the present invention uses the feature information that characterizes the network performance of the subnet and combines the corresponding monitoring model of the subnet to perform network monitoring, thereby improving the accuracy and efficiency of network monitoring.
[0115] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned network monitoring method. The computer-readable storage medium can be a volatile memory (volatile memory), such as a random access memory (Random-Access Memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (Read-Only Memory, ROM), a flash memory (flash memory), a hard disk (Hard Disk Drive, HDD) or a solid-state drive (Solid-State Drive, SSD); or it can be a respective device including one or any combination of the above-mentioned memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc.
[0116] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.
[0117] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0118] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0120] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in this utility model should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A network monitoring method, characterized in that: The method comprises: Acquire a network to be monitored, and divide the network to be monitored into multiple subnets based on link information of the network to be monitored; Acquire network status information and network performance data of a target subnet at a current moment; the target subnet is any one of the multiple subnets; the network status information includes at least one device parameter corresponding to a network device in the target subnet, and the network performance data includes at least one performance parameter corresponding to a network device in the target subnet; Determine the component information corresponding to each parameter of the at least one equipment parameter and the at least one performance parameter according to the principal component analysis method to obtain a plurality of component information, and determine the sum of the plurality of component information as the total component information; Determine a ratio of each component information in the plurality of component information to the total component information as a contribution rate, and obtain a plurality of component contribution rates corresponding to the plurality of component information, wherein each component information corresponds to a component contribution rate; From the plurality of component contribution rates, selecting component contribution rates in descending order, until the sum of at least one selected component contribution rate exceeds a preset contribution rate threshold; Acquire at least one component information corresponding to the selected at least one component contribution rate from the plurality of component information, and determine at least one parameter corresponding to the at least one component information among the at least one device parameter and the at least one performance parameter as the current network characteristic information; The preset monitoring model corresponding to the target subnet is used to perform network monitoring on the target subnet based on the current network feature information.
2. The method according to claim 1, characterized in that: Before performing network monitoring on the target subnet by using the preset monitoring model corresponding to the target subnet based on the current network feature information, the method further includes: The current network characteristic information is optimized using a preset optimized network model.
3. The method according to claim 1, characterized in that Before selecting component contribution rates from the plurality of component contribution rates in descending order, the method further includes: Obtaining the network scenario of the target subnet; Based on the network scenario, selecting at least one component contribution rate from the multiple component contribution rates; Among the at least one component contribution rate, the component contribution rate of the corresponding component information that plays a positive role in the network scenario is increased, and / or the component contribution rate of the corresponding component information that plays a negative role in the network scenario is reduced.
4. The method according to claim 1, characterized in that The using the preset monitoring model to perform network monitoring on the target subnet based on the current network feature information includes: Obtaining historical network feature information of the target subnet from a preset time to the current time; Determine the network monitoring items corresponding to the target subnet by using the preset monitoring model based on the historical network feature information and the current network feature information; By using the preset monitoring model, when the normal deviation of the network monitoring item is greater than the preset normal deviation threshold, the abnormal alarm information corresponding to the target subnet is output.
5. The method according to claim 4, characterized in that The method further comprises: Obtaining network link information of the target subnet, and generating a target topology map using the network link information; The historical network feature information, the current network feature information and the abnormal alarm information are superimposed on the target topology map to obtain a superimposed topology map.
6. The method according to claim 4, characterized in that After outputting the abnormal alarm information corresponding to the target subnet, the method further includes: The abnormal alarm information is used to optimize the preset monitoring model.
7. A network monitoring device, characterized in that: include: A division module, used for acquiring a network to be monitored, and dividing the network to be monitored into a plurality of subnets based on link information of the network to be monitored; An acquisition module is used to obtain the network status information and network performance data of the target subnet at the current moment; The target subnet is any one of the multiple subnets; the network status information includes at least one device parameter corresponding to the network device in the target subnet, and the network performance data includes at least one performance parameter corresponding to the network device in the target subnet; A selection module, used for determining the component information corresponding to each parameter of the at least one equipment parameter and the at least one performance parameter according to the principal component analysis method, obtaining a plurality of component information, and determining the sum of the plurality of component information as the total component information; Determine a ratio of each component information in the multiple component information to the total component information as a contribution rate, and obtain multiple component contribution rates corresponding to the multiple component information, wherein each component information corresponds to a component contribution rate; select component contribution rates from the multiple component contribution rates in order from large to small until the sum of at least one selected component contribution rate exceeds a preset contribution rate threshold; obtain at least one component information corresponding to at least one selected component contribution rate from the multiple component information, and determine at least one parameter corresponding to the at least one component information among the at least one device parameter and the at least one performance parameter as the current network feature information; The monitoring module is used to perform network monitoring on the target subnet based on the current network feature information by using a preset monitoring model corresponding to the target subnet.
8. A network monitoring device, characterized in that: include: processor, memory, and communication bus; The communication bus is used to realize the communication connection between the processor and the memory; The processor is used to execute the network monitoring program stored in the memory to implement the network monitoring method according to any one of claims 1-6.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the network monitoring method according to any one of claims 1 to 6.
