Network fluctuation monitoring method and system based on big data

Abstract

The invention belongs to the technical field of network fluctuation monitoring, and particularly relates to a network fluctuation monitoring method and system based on big data. According to the network fluctuation monitoring method based on big data, the network state can be monitored in real time, the network fluctuation condition can be accurately evaluated, the normal state and the abnormal state of the network can be effectively distinguished by extracting the reference characteristic parameters and determining the standard evaluation range, and the network fluctuation condition can be accurately evaluated by counting and analyzing the real-time state indexes in the normal state. The standard evaluation range is dynamically optimized and adjusted, so that the real-time performance of the network monitoring process can be improved, the monitoring accuracy and efficiency can be improved in this way, network fluctuation can be found and responded in time through grading interval matching monitoring frequency in an abnormal state, stable operation of the network is ensured, and the network quality is improved. Therefore, powerful technical support is provided for network management.

Description

Technical Field

[0001] The present invention belongs to the technical field of network fluctuation monitoring, and in particular relates to a network fluctuation monitoring method and system based on big data. Background Art

[0002] With the rapid development of Internet technology, the types and number of network services are increasing, and the impact of network fluctuations on user experience and business continuity is becoming increasingly significant. When the network fluctuates greatly, it often leads to a decrease in user access speed or even service interruption, which will affect the user's normal work and life. Therefore, effective monitoring and management of network fluctuations to ensure network stability and reliability has become a key requirement for network services.

[0003] The network fluctuation monitoring method in the prior art usually relies on fixed threshold settings, which are difficult to adapt to the dynamic changes of the network environment, resulting in poor monitoring effect. For example, in the network service process, new network traffic patterns and user behaviors will appear over time. These changes may cause deviations in the original threshold judgment method, and then misjudgment or missed judgment will occur. Based on this, the present invention proposes a network fluctuation monitoring method based on big data to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide a network fluctuation monitoring method and system based on big data, which can adapt to the dynamic changes of the network environment and improve the accuracy and efficiency of network fluctuation monitoring.

[0005] The technical solution adopted by the present invention is as follows: A network fluctuation monitoring method based on big data, comprising: Acquire network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; Extracting features from the network status indicators to obtain benchmark feature parameters reflecting network fluctuations; Determining a standard assessment range of network fluctuations based on the benchmark characteristic parameters; Collecting real-time status indicators of network traffic data, and simultaneously extracting features to obtain real-time feature parameters, and then evaluating the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, wherein the network fluctuation state includes a normal state and an abnormal state; Performing statistical analysis on the real-time status indicators under the normal state, and optimizing and adjusting the standard evaluation range in real time according to the statistical analysis results; A differential analysis is performed on the real-time status indicator under the abnormal state to determine the abnormal differential of the network fluctuation, and the monitoring frequency of the network state is adjusted according to the abnormal differential.

[0006] In a preferred solution, after the network status indicator is output, preprocessing is performed synchronously, and the preprocessing step includes: Obtaining a network status indicator and performing denoising processing to eliminate noise in the network status indicator; Normalizing the denoised network status indicators to unify the magnitudes of the network status indicators; A timestamp is added to the normalized network status indicator to determine the time series characteristics of each of the network status indicators.

[0007] In a preferred solution, the step of extracting features from the network status indicator to obtain a reference feature parameter reflecting network fluctuations includes: Get network status indicators; Obtain demand characteristics that reflect network fluctuations; According to the demand feature, the network status indicator is compared, and a portion consistent with the demand feature is selected from the network status indicator, and recorded as a reference feature; A feature code is added to the reference feature and is synchronously output as a reference feature parameter.

[0008] In a preferred embodiment, the step of determining the standard evaluation range of network fluctuations based on the reference characteristic parameters includes: Acquire the reference characteristic parameters, and sort the reference characteristic parameters according to the acquisition time sequence; Performing difference processing on the sorted and adjacent reference characteristic parameters to obtain fluctuation parameters; Performing an offset process on the fluctuation parameter to obtain a fluctuation assessment interval; Counting the number of fluctuation parameters covered in each fluctuation evaluation interval and recording them as parameters to be evaluated; The parameters to be evaluated are arranged in descending order, and the fluctuation evaluation interval corresponding to the parameter to be evaluated with the highest ranking is recorded as the standard evaluation range.

[0009] In a preferred embodiment, the step of evaluating the real-time characteristic parameters according to the standard evaluation range to determine the network fluctuation state includes: Obtain real-time characteristic parameters of adjacent nodes and standard evaluation range; Performing difference processing on the real-time characteristic parameters of adjacent nodes to obtain real-time fluctuation parameters, and then comparing the real-time fluctuation parameters with the standard evaluation range; If the real-time fluctuation parameter falls within the standard evaluation range, it indicates that the real-time fluctuation parameter meets the preset standard, and the corresponding network fluctuation state is recorded as a normal state; If the real-time fluctuation parameter does not fall within the standard evaluation range, it indicates that the real-time fluctuation parameter does not meet the preset standard, and the corresponding network fluctuation state is recorded as an abnormal state.

[0010] In a preferred embodiment, the step of statistically analyzing the real-time status indicators under the normal state and optimizing and adjusting the standard evaluation range in real time according to the statistical analysis results includes: Obtaining real-time fluctuation parameters corresponding to the real-time status indicators under the normal state, and summarizing them into a subset to be evaluated; Performing average processing on the real-time fluctuation parameters in the subset to be evaluated to obtain an average fluctuation parameter; The standard evaluation range is dynamically adjusted according to the average fluctuation parameter and the span of the standard evaluation range to obtain an updated evaluation range, and the subsequent network fluctuation state is evaluated according to the updated evaluation range.

[0011] In a preferred solution, when the subset to be evaluated is summarized, the number of real-time fluctuation parameters in the subset to be evaluated is collected in real time and recorded as an execution condition parameter; Obtaining a first-level execution threshold, and comparing the execution threshold with an execution condition parameter; If the execution condition parameter is greater than or equal to the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is sufficient, and then the fluctuation range of the real-time fluctuation parameter is synchronously collected; Comparing the fluctuation range of the real-time fluctuation parameter with the standard evaluation range, and outputting the fluctuation deviation difference; Obtaining a secondary execution threshold, and comparing the secondary execution threshold with the fluctuation deviation difference, and when the fluctuation deviation difference exceeds the secondary execution threshold, performing optimization adjustment on the standard evaluation range, otherwise, retaining the standard evaluation range, and not performing optimization adjustment on the standard evaluation range; If the execution condition parameter is less than the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is insufficient, and the optimization adjustment of the standard evaluation range will not be continued.

[0012] In a preferred solution, in the abnormal state, the difference between the real-time fluctuation parameter under the real-time state indicator and the boundary of the standard evaluation range is calculated and recorded as the abnormal difference; Obtaining a grading interval, wherein a plurality of grading intervals are provided, and each grading interval corresponds to a monitoring frequency; The abnormal difference is compared with the classification interval, and the corresponding monitoring frequency is synchronously matched to monitor the network status.

[0013] The present invention also provides a network fluctuation monitoring system based on big data, using the above-mentioned network fluctuation monitoring method based on big data, including: A data acquisition module, the data acquisition module is used to acquire network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; A feature extraction module, the feature extraction module is used to extract features of the network status indicators to obtain benchmark feature parameters reflecting network fluctuations; An evaluation range output module, the evaluation range output module is used to determine a standard evaluation range of network fluctuations according to the benchmark characteristic parameters; A status evaluation module, the status evaluation module is used to collect real-time status indicators of network traffic data, and simultaneously perform feature extraction to obtain real-time feature parameters, and then evaluate the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, wherein the network fluctuation state includes a normal state and an abnormal state; An optimization module, the optimization module is used to perform statistical analysis on the real-time status indicators under the normal state, and to perform real-time optimization and adjustment on the standard evaluation range according to the statistical analysis results; The monitoring and adjustment module is used to perform differential analysis on the real-time status indicators under the abnormal state, determine the abnormal difference of network fluctuations, and adjust the monitoring frequency of the network state according to the abnormal difference.

[0014] And, an electronic device, the electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; Among them, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned network fluctuation monitoring method based on big data.

[0015] The technical effects achieved by the present invention are: The present invention can monitor the network status in real time and accurately evaluate the network fluctuation situation through the network fluctuation monitoring method based on big data. By extracting the benchmark characteristic parameters and determining the standard evaluation range, the normal state and abnormal state of the network can be effectively distinguished. Under the normal state, the standard evaluation range is dynamically optimized and adjusted by statistically analyzing the real-time state indicators, so that the network monitoring process can be improved in real time, thereby improving the accuracy and efficiency of monitoring. Under the abnormal state, the monitoring frequency can be matched by graded intervals, so that the network fluctuation can be discovered and responded to in time to ensure the stable operation of the network, thereby providing strong technical support for network management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 It is a schematic diagram of the system module of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in a preferred embodiment" that appears in different places in this specification does not refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0020] With the rapid development of Internet technology, the types and number of network services are increasing, and the stability and reliability of the network are becoming more and more important. However, network fluctuations are inevitable. They may be caused by a variety of factors, such as hardware failures, software defects, network attacks, etc. Network fluctuations not only affect user experience, but may also cause economic losses and data security issues. Therefore, effective monitoring and management of network fluctuations, and timely detection and handling of network anomalies are crucial to ensuring the stable operation of network services.

[0021] See also Figure 1 As shown, the present invention provides a network fluctuation monitoring method based on big data, comprising: S1. Obtain network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; In step S1, in the process of monitoring the network status, it is first necessary to obtain the network status indicators of the network traffic data. The network status indicators include key indicators such as data delay parameters and packet loss rate. Based on this, the historical operation status of the network can be understood, which is convenient for subsequent corresponding analysis. After the network status indicators are output, preprocessing is performed synchronously. The preprocessing steps include: Obtain network status indicators and perform denoising to eliminate noise in the network status indicators; Normalize the denoised network status indicators to unify the magnitude of the network status indicators; Add timestamps to the normalized network status indicators to determine the timing characteristics of each network status indicator; Specifically, after the network status indicators are output, they need to be preprocessed accordingly. That is, after obtaining these data, denoising is first performed. Due to the complex and changeable network environment, the data related to the network status indicators will inevitably be mixed with some noise. These noises may be caused by equipment failures, network fluctuations or other external interferences. In order to ensure the accuracy of the data, it is necessary to eliminate the noise through methods such as filters and smoothing algorithms to obtain purer network status indicator data. Secondly, the denoised network status indicators need to be normalized. Since the magnitudes and units of different indicators may be different, direct analysis and comparison may cause the impact of some indicators to be magnified or reduced. Therefore, it is necessary to unify the relevant network status indicators to one magnitude for fair comparison and analysis. Finally, it is necessary to add timestamps to the normalized network status indicators. The addition of timestamps is to determine the time series characteristics of each network status indicator. Through the timestamp, it is possible to understand when each indicator data is generated. This can be achieved by inserting specific timestamp information into the data, such as using Unix timestamps or other timestamp formats.

[0022] S2. Extract features from network status indicators to obtain benchmark feature parameters that reflect network fluctuations; In step S2, after the network status indicator is determined, corresponding feature extraction processing is required to obtain benchmark feature parameters reflecting network fluctuations, and the benchmark feature parameters obtained here are the basis for subsequent evaluation and analysis. The step of extracting features from the network status indicator to obtain benchmark feature parameters reflecting network fluctuations includes: Get network status indicators; Obtain demand characteristics that reflect network fluctuations; According to the demand characteristics, compare them with the network status indicators, select the parts that are consistent with the demand characteristics from the network status indicators, and record them as the benchmark characteristics; Add feature codes to the datum features and output them as datum feature parameters synchronously; Specifically, in order to extract features from network status indicators, it is first necessary to obtain specific network status indicators and clearly reflect the demand characteristics of network fluctuations. The demand characteristics may come from the specific needs of users, the special requirements of business scenarios, or the goals of network management. For example, some applications may be very sensitive to latency, while other applications may be more concerned with bandwidth stability. Based on these demand characteristics, corresponding comparative analysis can be performed with previously obtained network status indicators. Through this comparison method, parts that are consistent with or highly correlated with the demand characteristics can be screened out from a large number of network status indicators, and recorded as benchmark features reflecting network fluctuations. Finally, in order to facilitate subsequent processing and analysis, the benchmark features need to be feature encoded. The feature encoding can be a simple numerical encoding or a more complex encoding method, such as one-hot encoding or label encoding. After the feature encoding is completed, the benchmark feature parameters are synchronously output for subsequent use.

[0023] S3. Determine the standard assessment range of network fluctuations based on the benchmark characteristic parameters; In step S3, after the benchmark characteristic parameters are determined, it is also necessary to determine the standard evaluation range of network fluctuations based on the benchmark characteristic parameters. The standard evaluation range will serve as a basis for determining whether the network fluctuates, so as to facilitate the subsequent real-time monitoring of the network status. The step of determining the standard evaluation range of network fluctuations based on the benchmark characteristic parameters includes: Obtaining benchmark characteristic parameters, and sorting the benchmark characteristic parameters according to the acquisition time sequence; Perform difference processing on the sorted and adjacent benchmark characteristic parameters to obtain the fluctuation parameters; The fluctuation parameters are offset to obtain the fluctuation assessment interval; Count the number of fluctuation parameters covered in each fluctuation assessment interval and record them as parameters to be assessed; Arrange the parameters to be evaluated in descending order, and record the fluctuation evaluation interval corresponding to the highest ranked parameter to be evaluated as the standard evaluation range; Specifically, in order to determine the standard evaluation range of network fluctuations, the benchmark feature parameters are first sorted in the order of their collection time, and then the sorted and adjacent benchmark feature parameters are subjected to difference processing, so as to obtain multiple fluctuation parameters, which reflect the performance changes of the network between different time points. After obtaining the fluctuation parameters, offset processing is required. The offset processing method is bidirectional offset, and its purpose is to eliminate possible errors, so as to obtain a more accurate fluctuation evaluation interval. Then, the number of fluctuation parameters covered in each fluctuation evaluation interval needs to be counted and recorded as parameters to be evaluated. The number of parameters to be evaluated reflects the density of fluctuation parameters in each fluctuation evaluation interval, which can be used as a basis for evaluating network stability. Finally, the parameters to be evaluated are arranged in order from large to small. After the arrangement is completed, the fluctuation evaluation interval corresponding to the parameter to be evaluated with the highest ranking will be selected and recorded as the standard evaluation range. The standard evaluation range represents the normal fluctuation range of network fluctuations, that is, the standard evaluation range.

[0024] S4, collecting real-time status indicators of network traffic data, and synchronously extracting features to obtain real-time feature parameters, and then evaluating the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, wherein the network fluctuation state includes a normal state and an abnormal state; In step S4, after the standard evaluation range is determined, the real-time status indicators of the network traffic data can be collected, and feature extraction can be performed simultaneously to obtain real-time feature parameters. By comparing the real-time feature parameters with the standard evaluation range, the network fluctuation state can be evaluated and determined. In this embodiment, the network fluctuation state mainly includes two types: normal state and abnormal state. Among them, the real-time feature parameters are evaluated according to the standard evaluation range to determine the network fluctuation state, including: Obtain real-time characteristic parameters of adjacent nodes and standard evaluation range; Perform difference processing on the real-time characteristic parameters of adjacent nodes to obtain real-time fluctuation parameters, and then compare the real-time fluctuation parameters with the standard evaluation range; If the real-time fluctuation parameter falls within the standard evaluation range, it indicates that the real-time fluctuation parameter meets the preset standard, and the corresponding network fluctuation state is recorded as normal; If the real-time fluctuation parameter does not fall within the standard evaluation range, it indicates that the real-time fluctuation parameter does not meet the preset standard, and the corresponding network fluctuation state is recorded as an abnormal state; Specifically, in order to ensure the stable operation of the network, it is necessary to conduct a detailed evaluation of the real-time characteristic parameters based on the established standard evaluation range, so as to accurately judge the fluctuation state of the network. First, it is necessary to obtain the real-time characteristic parameters from the adjacent nodes in the network, and then perform difference processing on the real-time characteristic parameters provided by the adjacent nodes to calculate the real-time fluctuation parameters. After obtaining the real-time fluctuation parameters, they are compared with the previously determined standard evaluation range. If after comparison, it is found that the real-time fluctuation parameters completely fall within the standard evaluation range, it can be determined that the real-time fluctuation parameters meet the preset standards. In this case, the corresponding network fluctuation state is recorded as a normal state, indicating that the network runs smoothly and there are no abnormal fluctuations. On the contrary, if the real-time fluctuation parameters fail to fall within the standard evaluation range, it means that they exceed the preset standards and do not meet the expected fluctuation range. In this case, the corresponding network fluctuation state will be recorded as an abnormal state, indicating that there may be potential problems or unstable factors in the network, which require further inspection and processing.

[0025] S5. Conduct statistical analysis on the real-time status indicators under normal conditions, and optimize and adjust the standard evaluation scope in real time based on the statistical analysis results; In step S5, for the real-time status indicators in the normal state, corresponding statistical analysis needs to be performed, and according to the results of the statistical analysis, the standard evaluation range is optimized and adjusted in real time to adapt to changes in the network environment and ensure the accuracy of the status evaluation process. The steps of performing statistical analysis on the real-time status indicators in the normal state and optimizing and adjusting the standard evaluation range in real time according to the statistical analysis results include: Obtain the real-time fluctuation parameters corresponding to the real-time status indicators under normal conditions and summarize them into subsets to be evaluated; Perform average processing on the real-time fluctuation parameters in the subset to be evaluated to obtain the average fluctuation parameter; According to the average fluctuation parameter and the span of the standard evaluation range, the standard evaluation range is dynamically adjusted to obtain an updated evaluation range, and the subsequent network fluctuation state is evaluated according to the updated evaluation range; Specifically, when statistically analyzing the real-time status indicators under normal conditions, it is first necessary to obtain the real-time fluctuation parameters corresponding to the real-time status indicators under normal conditions, and at the same time, the real-time fluctuation parameters that have been counted will be summarized to form a subset to be evaluated for further analysis, and then the real-time fluctuation parameters in the subset to be evaluated will be averaged to obtain an average fluctuation parameter, which can better reflect the fluctuation of the network under normal conditions. Finally, the standard evaluation range will be dynamically adjusted based on the average fluctuation parameter and the span of the existing standard evaluation range. Based on this method, an updated evaluation range can be obtained, and this updated evaluation range will be closer to the actual network operation situation. Thereafter, the subsequent network fluctuation state will be evaluated more accurately and effectively based on the updated evaluation range. In addition, when the subset to be evaluated is summarized, the number of real-time fluctuation parameters in the subset to be evaluated is collected in real time and recorded as the execution condition parameter; Obtaining a first-level execution threshold, and comparing the execution threshold with the execution condition parameter; If the execution condition parameter is greater than or equal to the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is sufficient, and then the fluctuation range of the real-time fluctuation parameter is synchronously collected; Compare the fluctuation range of the real-time fluctuation parameter with the standard evaluation range, and output the fluctuation deviation difference; Obtaining a secondary execution threshold, and comparing the secondary execution threshold with the fluctuation deviation difference, and when the fluctuation deviation difference exceeds the secondary execution threshold, performing optimization adjustment on the standard evaluation range, otherwise, retaining the standard evaluation range, and not performing optimization adjustment on the standard evaluation range; If the execution condition parameter is less than the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is insufficient, and the optimization adjustment of the standard evaluation scope will not be continued; In the process of summarizing the subset to be evaluated, it is necessary to collect the number of each real-time fluctuation parameter in the subset to be evaluated in real time and record it as an execution condition parameter. Then it is necessary to obtain the first-level execution threshold. The first-level execution threshold is a pre-set standard value used to compare with the execution condition parameter. Through this comparison, it can be determined whether the amount of data in the subset to be evaluated is sufficient, so as to decide whether further analysis and processing is needed. If the execution condition parameter is greater than or equal to the first-level execution threshold, it means that the amount of data in the subset to be evaluated is sufficient. In this case, the fluctuation range of the real-time fluctuation parameter will be collected synchronously, and then the fluctuation range of the real-time fluctuation parameter will be compared with the pre-set standard evaluation range to determine the fluctuation deviation difference. The fluctuation deviation difference reflects the difference between the actual fluctuation range and the standard evaluation range. In order to further deal with this difference, it is necessary to obtain the secondary execution threshold, which is also a pre-set standard value. By comparing the secondary execution threshold with the fluctuation deviation difference, it is determined whether the standard evaluation range needs to be optimized. If the fluctuation deviation difference exceeds the secondary execution threshold, the optimization adjustment of the standard evaluation range will be performed to ensure the accuracy and reliability of the evaluation. On the contrary, if the fluctuation deviation difference does not exceed the secondary execution threshold, the existing standard evaluation range will be retained and no optimization adjustment will be performed. However, if the execution condition parameter is less than the first-level execution threshold, it means that the amount of data in the subset to be evaluated is insufficient. In this case, the optimization adjustment of the standard evaluation range will no longer be performed because the lack of sufficient data may lead to inaccurate and unreliable evaluation results.

[0026] S6. Perform differential analysis on the real-time status indicators under abnormal conditions to determine the abnormal differential of network fluctuations, and adjust the monitoring frequency of the network status according to the abnormal differential; In step S6, for the real-time status indicator in an abnormal state, a corresponding difference analysis needs to be performed. Through the difference analysis, the abnormal difference of the network fluctuation can be determined. According to the abnormal difference, the monitoring frequency of the network status can be adjusted to ensure that the network fluctuation problem can be discovered and handled in time. In the abnormal state, the difference between the real-time fluctuation parameter under the real-time status indicator and the boundary of the standard evaluation range is calculated and recorded as the abnormal difference; Obtaining a grading interval, wherein a plurality of grading intervals are set, and each grading interval corresponds to a monitoring frequency; The abnormal difference is compared with the classification interval, and the corresponding monitoring frequency is synchronously matched to monitor the network status.

[0027] Specifically, under abnormal conditions, it is necessary to calculate the real-time fluctuation parameters under the real-time status indicators, and compare them with the difference between the boundaries of the standard evaluation range, and record this difference as the abnormal difference. After that, it is necessary to obtain the grading interval. The grading interval is pre-set and usually contains multiple different intervals. Each grading interval corresponds to a specific monitoring frequency. By comparing the abnormal difference with the grading interval, the corresponding monitoring frequency can be synchronously matched, so that the network status can be monitored in real time according to the matched monitoring frequency to ensure the stability and security of the network.

[0028] See also Figure 2 A network fluctuation monitoring system based on big data, using the above-mentioned network fluctuation monitoring method based on big data, includes: A data acquisition module, the data acquisition module is used to obtain network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; The feature extraction module is used to extract features of network status indicators to obtain benchmark feature parameters reflecting network fluctuations; An evaluation range output module, the evaluation range output module is used to determine a standard evaluation range of network fluctuations based on a benchmark characteristic parameter; The status evaluation module is used to collect real-time status indicators of network traffic data, and simultaneously perform feature extraction to obtain real-time feature parameters, and then evaluate the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, where the network fluctuation state includes normal state and abnormal state; Optimization module: The optimization module is used to perform statistical analysis on real-time status indicators under normal conditions, and to optimize and adjust the standard evaluation range in real time based on the statistical analysis results; The monitoring and adjustment module is used to perform differential analysis on real-time status indicators under abnormal conditions, determine the abnormal differential of network fluctuations, and adjust the monitoring frequency of the network status according to the abnormal differential.

[0029] In the above, the system includes a data acquisition module, a feature extraction module, an evaluation range output module, a state evaluation module, an optimization module and a monitoring and adjustment module. The data acquisition module is responsible for collecting network status indicators under network traffic data in real time. The network status indicators cover important parameters such as data delay parameters and packet loss rate. The feature extraction module will perform corresponding analysis and processing on the network status indicators, and extract benchmark feature parameters that can reflect network fluctuations. The evaluation range output module determines the standard evaluation range of network fluctuations based on the benchmark feature parameters. The standard evaluation range provides an important reference standard for subsequent network status evaluation. The state evaluation module is responsible for collecting real-time status indicators of network traffic data in real time, and simultaneously extracts features to obtain real-time status indicators. Characteristic parameters, and then evaluate the real-time characteristic parameters according to the previously determined standard evaluation range to determine the network fluctuation state. The network fluctuation state is mainly divided into normal state and abnormal state. The optimization module focuses on statistical analysis of real-time status indicators under normal state. Based on this, the system can optimize and adjust the standard evaluation range in real time to adapt to changes in the network environment and ensure the accuracy and timeliness of the evaluation. The monitoring and adjustment module will play a role when an abnormal state is detected. It will perform differential analysis on the real-time status indicators under the abnormal state to determine the abnormal differential of the network fluctuation, and according to the abnormal differential, the system can dynamically adjust the monitoring frequency of the network state, so as to more accurately monitor and manage network fluctuations and ensure the stable operation of the network.

[0030] See also Figure 3 , an electronic device, the electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; Among them, the memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned network fluctuation monitoring method based on big data.

[0031] The processor of the above-mentioned electronic device can be a central processing unit (CPU), a graphics processing unit (GPU) or an application-specific integrated circuit (ASIC), and the memory can be a random access memory (RAM), a read-only memory (ROM), flash memory or other forms of non-volatile memory. By executing computer programs, the electronic device can realize real-time monitoring and analysis of network traffic data, thereby effectively managing network fluctuations. In addition, the electronic device can also include an operator, an input device and an output device. The operator can perform various computing tasks, the input device allows the user to interact with the electronic device, and the output device provides a visual display of the processing results. These components of the electronic device work together to ensure that the network fluctuation monitoring method based on big data can be executed efficiently and accurately.

[0032] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0033] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A network fluctuation monitoring method based on big data, characterized by: include: Acquire network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; Extracting features of the network status indicators to obtain benchmark feature parameters reflecting network fluctuations; Determining a standard assessment range of network fluctuations based on the benchmark characteristic parameters; Collecting real-time status indicators of network traffic data, and simultaneously performing feature extraction to obtain real-time feature parameters, and then evaluating the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, wherein the network fluctuation state includes a normal state and an abnormal state; Performing statistical analysis on the real-time status indicators under the normal state, and optimizing and adjusting the standard evaluation range in real time according to the statistical analysis results; A differential analysis is performed on the real-time status indicator under the abnormal state to determine the abnormal differential of the network fluctuation, and the monitoring frequency of the network state is adjusted according to the abnormal differential.

2. According to the big data-based network fluctuation monitoring method of claim 1, it is characterized by: After the network status indicator is output, preprocessing is performed synchronously, and the preprocessing steps include: Obtaining a network status indicator and performing denoising processing to eliminate noise in the network status indicator; Normalizing the denoised network status indicators to unify the magnitudes of the network status indicators; A timestamp is added to the normalized network status indicator to determine the time series characteristics of each of the network status indicators.

3. The network fluctuation monitoring method based on big data according to claim 1 is characterized by: The step of extracting features from the network status indicators to obtain reference feature parameters reflecting network fluctuations includes: Get network status indicators; Obtain demand characteristics that reflect network fluctuations; According to the demand characteristics, the network status indicators are compared, and a portion consistent with the demand characteristics is selected from the network status indicators, and recorded as a benchmark characteristic; A feature code is added to the reference feature and is synchronously output as a reference feature parameter.

4. The network fluctuation monitoring method based on big data according to claim 1 is characterized by: The step of determining the standard evaluation range of network fluctuations according to the benchmark characteristic parameters includes: Acquire the reference characteristic parameters, and sort the reference characteristic parameters according to the acquisition time sequence; Performing difference processing on the sorted and adjacent reference characteristic parameters to obtain fluctuation parameters; Performing an offset process on the fluctuation parameter to obtain a fluctuation assessment interval; Counting the number of fluctuation parameters covered in each fluctuation evaluation interval and recording them as parameters to be evaluated; The parameters to be evaluated are arranged in descending order, and the fluctuation evaluation interval corresponding to the parameter to be evaluated with the highest ranking is recorded as the standard evaluation range.

5. The network fluctuation monitoring method based on big data according to claim 1 is characterized by: The step of evaluating the real-time characteristic parameters according to the standard evaluation range to determine the network fluctuation state includes: Obtain real-time characteristic parameters of adjacent nodes and standard evaluation range; Performing difference processing on the real-time characteristic parameters of adjacent nodes to obtain real-time fluctuation parameters, and then comparing the real-time fluctuation parameters with the standard evaluation range; If the real-time fluctuation parameter falls within the standard evaluation range, it indicates that the real-time fluctuation parameter meets the preset standard, and the corresponding network fluctuation state is recorded as a normal state; If the real-time fluctuation parameter does not fall within the standard evaluation range, it indicates that the real-time fluctuation parameter does not meet the preset standard, and the corresponding network fluctuation state is recorded as an abnormal state.

6. The network fluctuation monitoring method based on big data according to claim 1 is characterized by: The step of performing statistical analysis on the real-time status indicators under the normal state and optimizing and adjusting the standard evaluation range in real time according to the statistical analysis results includes: Obtaining real-time fluctuation parameters corresponding to the real-time status indicators under the normal state, and summarizing them into a subset to be evaluated; Performing average processing on the real-time fluctuation parameters in the subset to be evaluated to obtain an average fluctuation parameter; The standard evaluation range is dynamically adjusted according to the average fluctuation parameter and the span of the standard evaluation range to obtain an updated evaluation range, and the subsequent network fluctuation state is evaluated according to the updated evaluation range.

7. A network fluctuation monitoring method based on big data according to claim 6, characterized in that: When the subset to be evaluated is summarized, the number of real-time fluctuation parameters in the subset to be evaluated is collected in real time and recorded as an execution condition parameter; Obtaining a first-level execution threshold, and comparing the execution threshold with an execution condition parameter; If the execution condition parameter is greater than or equal to the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is sufficient, and then the fluctuation range of the real-time fluctuation parameter is synchronously collected; Comparing the fluctuation range of the real-time fluctuation parameter with the standard evaluation range, and outputting the fluctuation deviation difference; Obtaining a secondary execution threshold, and comparing the secondary execution threshold with the fluctuation deviation difference, and when the fluctuation deviation difference exceeds the secondary execution threshold, performing optimization adjustment on the standard evaluation range, otherwise, retaining the standard evaluation range, and not performing optimization adjustment on the standard evaluation range; If the execution condition parameter is less than the first-level execution threshold, it indicates that the amount of data in the subset to be evaluated is insufficient, and the optimization adjustment of the standard evaluation range will not be continued.

8. The network fluctuation monitoring method based on big data according to claim 1 is characterized by: Under the abnormal state, the difference between the real-time fluctuation parameter under the real-time state indicator and the boundary of the standard evaluation range is calculated and recorded as the abnormal difference; Obtaining a grading interval, wherein a plurality of grading intervals are provided, and each grading interval corresponds to a monitoring frequency; The abnormal difference is compared with the classification interval, and the corresponding monitoring frequency is synchronously matched to monitor the network status.

9. A network fluctuation monitoring system based on big data, characterized by: The network fluctuation monitoring method based on big data according to any one of claims 1 to 8 comprises: A data acquisition module, the data acquisition module is used to acquire network status indicators of network traffic data, wherein the network status indicators include data delay parameters and packet loss rate; A feature extraction module, the feature extraction module is used to extract features of the network status indicators to obtain benchmark feature parameters reflecting network fluctuations; An evaluation range output module, the evaluation range output module is used to determine a standard evaluation range of network fluctuations according to the benchmark characteristic parameters; A status evaluation module, the status evaluation module is used to collect real-time status indicators of network traffic data, and simultaneously perform feature extraction to obtain real-time feature parameters, and then evaluate the real-time feature parameters according to the standard evaluation range to determine the network fluctuation state, wherein the network fluctuation state includes a normal state and an abnormal state; An optimization module, the optimization module is used to perform statistical analysis on the real-time status indicators under the normal state, and to perform real-time optimization and adjustment on the standard evaluation range according to the statistical analysis results; The monitoring and adjustment module is used to perform differential analysis on the real-time status indicators under the abnormal state, determine the abnormal difference of network fluctuations, and adjust the monitoring frequency of the network state according to the abnormal difference.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the network fluctuation monitoring method based on big data as described in any one of claims 1 to 8.

Images