Intelligent cloud problem and fault analysis method and system

Abstract

The invention discloses an intelligent cloud problem fault analysis method and system, and relates to the technical field of cloud computing, and the method comprises the steps: extracting data features, and carrying out the first extraction of fault features; carrying out fault prediction and positioning; and establishing a first alarm mechanism, creating a fault management knowledge base, and providing an automatic processing suggestion. According to the method, deep analysis and processing of fault data are achieved by extracting fault features, the accuracy and efficiency of fault prediction and positioning are improved, guarantee is provided for the stability and reliability of a cloud system, efficient recognition and processing of faults are achieved through fault prediction and positioning, and the fault prediction and positioning efficiency is improved. According to the invention, the method improves the accuracy and efficiency of fault processing, guarantees the stability and reliability of the cloud system, achieves the quick response and effective processing of faults through a fault alarm mechanism and a fault management knowledge base, and improves the operation and maintenance efficiency and the stability and reliability of the cloud system.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to an intelligent cloud problem fault analysis method and system. Background Art

[0002] With the widespread application of cloud computing technology, the stability and reliability of cloud systems have become the core concerns of operation and maintenance management. In recent years, intelligent fault analysis methods have developed rapidly. By using data analysis and machine learning technologies, automatic detection, prediction and location of cloud system faults can be achieved. However, existing fault analysis methods still have certain limitations in real-time, accuracy and degree of automation.

[0003] The shortcomings of existing related technologies are mainly reflected in the following aspects. First, traditional fault analysis methods rely on manual experience and often cannot respond to the rapidly changing cloud environment in a timely manner, resulting in inefficient fault handling. Second, the existing technology is not accurate enough in fault feature extraction and fault prediction, and it is difficult to accurately identify complex fault modes. In addition, the existing methods lack an effective fault knowledge base and automated processing suggestions, resulting in a lack of standardization and intelligence in the fault handling process, making it difficult to achieve rapid recovery from faults. Therefore, how to improve the real-time, accuracy and automation level of fault analysis has become an urgent problem to be solved in the current cloud problem fault analysis methods. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing cloud problem fault analysis method has poor real-time performance, low accuracy, low degree of automation, and the problem of how to achieve rapid fault detection, prediction and positioning.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent cloud-based problem fault analysis method, comprising extracting data features and performing a first extraction of fault features; performing fault prediction and location; establishing a first alarm mechanism, creating a fault management knowledge base, and providing automated processing suggestions.

[0007] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, the extracting of data features includes collecting system operation data and performing preprocessing.

[0008] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, the first extraction of fault features includes extracting fault features and standardizing the feature data.

[0009] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, the fault prediction and location includes identifying the fault mode and type.

[0010] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, the fault prediction and location also includes analyzing historical data to identify normal modes and abnormal modes.

[0011] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, the establishment of the first alarm mechanism includes setting a threshold. If the indicator exceeds the preset range, it is judged as a fault, the system automatically generates an alarm, and notifies the operation and maintenance personnel.

[0012] As a preferred solution of the intelligent cloud-based problem fault analysis method described in the present invention, wherein: the creation of a fault management knowledge base and the provision of automated processing suggestions include recording fault types, processing steps and historical experience, creating a fault management knowledge base, and providing automated processing suggestions by matching current faults with historical data.

[0013] Another object of the present invention is to provide an intelligent cloud problem fault analysis system, which can establish an alarm mechanism, create a fault management knowledge base, provide automated processing suggestions, and continuously optimize system performance, thereby solving the problem that the current cloud problem fault analysis has a low degree of automation.

[0014] As a preferred solution of the intelligent cloud-based problem fault analysis system described in the present invention, it includes: a data processing module, a fault analysis module, and a fault management module; the data processing module is used to collect cloud system operation data and perform preprocessing; the fault analysis module is used to predict and locate faults and analyze historical data; the fault management module is used to establish an alarm mechanism, create a fault management knowledge base, provide automated processing suggestions, and continuously optimize system performance.

[0015] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of an intelligent cloud problem fault analysis method.

[0016] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of an intelligent cloud problem fault analysis method.

[0017] Beneficial effects of the invention: The intelligent cloud problem fault analysis method provided by the invention first extracts fault features, extracts features valuable for fault analysis, and standardizes the feature data, thereby constructing a data set containing multiple dimensions and statistical features, laying a solid foundation for fault prediction and positioning, realizing comprehensive collection and preprocessing of cloud system operation data, improving the quality and reliability of data, and providing accurate data support for subsequent fault prediction and positioning. By extracting fault features, in-depth analysis and processing of fault data are realized, providing effective data support for fault prediction and positioning, achieving the goal of improving the accuracy and efficiency of fault prediction and positioning. The high-efficiency fault detection and fault location algorithm can effectively identify and deal with faults, improve the accuracy and efficiency of fault handling, and ensure the stability and reliability of the cloud system. The high-efficiency fault detection and fault location algorithm can effectively identify and deal with faults, and improve the accuracy and efficiency of fault handling, thereby ensuring the stability and reliability of the cloud system. The high-efficiency fault detection and fault location algorithm can effectively identify and deal with faults, and improve the accuracy and efficiency of fault handling, thereby ensuring the stability and reliability of the cloud system. The high-efficiency fault detection and fault location algorithm can effectively identify and deal with faults, and improve the operation and maintenance efficiency and the stability and reliability of the cloud system. The present invention can achieve better results in terms of real-time performance, accuracy and automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 An overall flow chart of an intelligent cloud problem fault analysis method provided for the first embodiment of the present invention.

[0020] Figure 2 A module diagram of an intelligent cloud-based problem fault analysis system provided for a third embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, 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 drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0022] Example 1, reference Figure 1, is an embodiment of the present invention, and provides an intelligent cloud problem fault analysis method, comprising:

[0023] S1: Extract data features and perform the first extraction of fault features.

[0024] Furthermore, extracting data features includes collecting system operation data. Cloud system operation data can be collected from different sources through various monitoring tools and interfaces. The data can be real-time (such as network traffic, CPU utilization) or historical (such as application logs, system events), and stored in the database in the form of time series for subsequent analysis. The collected data can be pre-processed to ensure the quality and accuracy of the data. Pandas, NumPy and other libraries in Python can be used for data cleaning, and the reliability of the data can be improved through filtering and denoising, filling missing values, detecting and correcting outliers, and other methods.

[0025] It should be noted that the first extraction of fault features includes extracting features that are valuable for fault analysis, such as CPU load, memory usage, etc., and standardizing the feature data. The first extraction of fault features can be performed by a variety of methods, such as Pearson correlation coefficient or principal component analysis.

[0026] As a preferred method, in this embodiment, the Pearson correlation coefficient method is used to extract the most valuable feature data for fault analysis, and the principle steps are as follows:

[0027] The linear relationship between each feature and the fault label is calculated using the Pearson Correlation Coefficient, expressed as:

[0028]

[0029] Among them, r is the Pearson correlation coefficient, x i To evaluate the eigenvalues, y i is the fault characteristic value, and are the means respectively. When |r|>0.5, the evaluation feature is considered to have a strong correlation with the fault feature and is thus extracted as the fault feature.

[0030] It should be noted that in another feasible embodiment, principal component analysis (PCA) is adopted to extract the most valuable feature data for fault analysis. PCA is a statistical method that converts a set of possibly related variables into a set of linearly unrelated variables, called principal components, through orthogonal transformation. PCA reduces the dimension of features and retains the principal components with the most information. The PCA algorithm is applied to transform the preprocessed data. The goal of PCA is to find the principal components by performing eigenvalue decomposition on the feature covariance matrix, project the original data onto the selected principal components, and obtain the reduced-dimensional feature data, which are the extracted fault features.

[0031] It should also be noted that after the fault feature extraction is completed, the feature data is standardized, including encoding and converting the features to meet the input requirements of the subsequent model for fault prediction and location. All features need to be converted into numerical format. For example, categorical features (such as error type) are converted into numerical format through One-Hot Encoding. For time series data, the sliding window method is used to generate time features. Through the sliding window, the statistical features of the time series data, such as the average, maximum, minimum and standard deviation, are extracted to retain the time information. For features with strong time dependence (network delay and response time), they are aggregated within a certain time window to generate statistical features such as average and peak values. By first extracting the fault features and standardizing the feature data, a data set containing multiple dimensions and statistical features is constructed, which lays a solid foundation for fault prediction and location.

[0032] S2: Perform fault prediction and location.

[0033] Furthermore, fault prediction and location include identifying fault modes and types. This embodiment uses a trained XGBoost model to predict whether a fault exists by inputting real-time data. The model achieves efficient fault identification through feature extraction and classification. By learning from historical data, the model can be continuously optimized to improve the accuracy of fault detection. In the fault prediction stage, the trained XGBoost model can perform real-time fault prediction for new samples. The specific steps are as follows:

[0034] (1) Input new data. Input the newly collected system data into the XGBoost model. This data may include device status, operating parameters, alarm information, etc. The model will perform analysis based on the existing historical data.

[0035] (2) Generate prediction results. The XGBoost model uses its ensemble learning characteristics to combine the results of multiple weak learners and output the probability of each fault type. This process is usually efficient and fast and can provide prediction results at the moment the equipment is running.

[0036] (3) Determine the failure risk. Based on the probability output by the model, the operation and maintenance personnel can quickly determine the current failure risk of the system. For example, when the probability of a certain failure type exceeds the preset threshold, the system will automatically issue an alarm to prompt the operation and maintenance personnel to pay attention.

[0037] Fault location is a key step to ensure timely problem resolution. The application of XGBoost model in fault location includes:

[0038] (1) Feature importance evaluation: XGBoost can evaluate the contribution of each feature to model prediction. These features may include equipment parameters, environmental conditions, and operation records. Operation and maintenance personnel can use feature importance ranking to quickly find features that are highly correlated with the fault, thereby focusing on possible causes of the fault. For example, if the weight of a feature is very high and its value is abnormal when the fault occurs, it can be preliminarily determined that the feature may be the root cause of the fault.

[0039] (2) Decision path analysis: The decision tree structure of the XGBoost model allows operation and maintenance personnel to track the predicted path of fault instances. By analyzing the decision path of a specific fault, operation and maintenance personnel can understand how the model makes judgments based on various features. For example, if some feature values ​​in the decision path are not within the normal range, this suggests that these features may be directly related to the fault.

[0040] (3) Fault pattern recognition: By analyzing historical fault data, the XGBoost model can not only predict the fault type, but also identify the similarities and patterns between different faults. This capability helps operation and maintenance personnel better understand the potential fault mechanisms, especially when multiple faults coexist.

[0041] (4) Multi-dimensional data association analysis. Failures are often the result of the combined effect of multiple factors. XGBoost can integrate data from different sources for comprehensive analysis. For example, by combining multi-dimensional information such as device performance, external environment, and user operations, operation and maintenance personnel can fully understand the background of the failure. This multi-dimensional analysis capability greatly improves the accuracy and efficiency of fault location.

[0042] It should be noted that fault prediction and location also include analyzing historical data and identifying normal and abnormal modes. This embodiment uses time series data analysis technology to monitor the changing trends of key performance indicators. By analyzing historical data, the model can identify the difference between normal and abnormal modes and realize early warning of faults. For example, abnormal fluctuations in network delay may indicate the occurrence of network faults. The specific steps include: first, real-time monitoring of key performance indicators (KPIs) through the principle steps of trend monitoring. In the data collection stage, time series data such as response time, transaction volume, and error rate of the collection system are collected. In the data preprocessing stage, these data are time-aligned, missing value processed, and noise reduced to ensure the quality and continuity of the data. Then, the trend component of the data is extracted using time series decomposition technology, and future trends are predicted through prediction models such as ARIMA. Set Prediction intervals and thresholds are used to detect trend deviations. Once abnormal fluctuations in KPIs are monitored, such as when the deviation between the current trend and the predicted trend exceeds the preset threshold, the system will mark potential fault points. At the same time, in the principle step of pattern recognition, feature extraction is performed on historical data, including periodic features and statistical features, to identify normal operating modes. Through cluster analysis and sequence pattern mining, the normal behavior patterns of the system are learned, and real-time data are compared with these patterns, the distance between them is calculated, and abnormal scores are assigned to real-time data points. In this way, when real-time data is significantly different from normal patterns, the system can identify abnormal patterns, thereby achieving early warning of faults. Through this method of integrating trend monitoring and pattern recognition, the fault analysis system of the present application can effectively predict and locate potential faults in cloud systems, and provide timely fault warnings and disposal suggestions for the operation and maintenance team.

[0043] S3: Establish a first alert mechanism, create a fault management knowledge base, and provide automated processing suggestions.

[0044] Furthermore, establishing the first alarm mechanism includes setting a threshold. If the indicator exceeds the preset range, it is judged as a fault. The system automatically generates an alarm and notifies the operation and maintenance personnel. The first alarm mechanism can set a variety of alarm judgment methods, which can be an alarm mechanism based on threshold judgment or an alarm mechanism based on machine learning.

[0045] As a preferred embodiment, a threshold-based alarm mechanism is adopted in this embodiment. First, the specific thresholds of CPU utilization and network delay are determined through quantitative analysis. For example, for CPU utilization, the threshold is set to 75%, which is based on historical data analysis. When the CPU utilization exceeds this level, the system performance may be affected. For network delay, the threshold is set to 200 milliseconds. Exceeding this delay may cause a decline in user experience. The monitoring system collects these indicators at a frequency of once per second. The real-time data is smoothed by the moving average method to reduce false alarms caused by instantaneous fluctuations. When the CPU utilization is detected to exceed 75% or the network delay exceeds 200 milliseconds for three consecutive times, the system triggers an alarm. This follows The "three-times principle" is implemented to avoid false alarms caused by a single anomaly. The alarm information records in detail the abnormal value (such as the actual CPU utilization reaching 80%), the fault type (such as resource overload or network congestion) and the possible impact (such as slow service response or transaction failure). The alarm information is instantly notified to the operation and maintenance personnel in the form of SMS and email through the integrated notification system. At the same time, the alarm information will also be highlighted on the console. After receiving the alarm, the operation and maintenance personnel will perform log review, resource allocation adjustment or network configuration optimization according to the preset fault handling process to restore the normal operation of the system. This mechanism is particularly important in dealing with resource-constrained scenarios caused by sudden traffic, which can ensure rapid response and minimize service interruption time.

[0046] In another feasible embodiment, an alarm mechanism based on machine learning is adopted. First, a large amount of feature data is extracted from the historical fault database, including but not limited to CPU utilization, memory usage, disk I / O, network traffic, etc. These data are processed by feature engineering, such as standardization, normalization and dimension reduction, to enhance the learning effect of the model. Support vector machine (SVM) and deep neural network (DNN) are selected as the main machine learning models, and the model parameters are optimized through cross-validation and grid search technology to achieve the best fault prediction performance. In the real-time monitoring stage, the system inputs the current system operation parameters into the trained model, and the model calculates a fault risk score. The score is based on quantitative analysis and divides the risk into five levels, from "low risk" to "severe risk". When the risk score exceeds the preset threshold (for example, above medium risk), the system will automatically generate an alarm message, including the specific fault risk level (such as "high risk"), the possible fault type based on the model output (such as "database connection timeout"), and the recommended treatment measures based on historical case analysis (such as "restart database service" or "increase server resources"). The alarm information is sent to the operation and maintenance personnel in real time in the form of email, SMS and instant message through the integrated notification system. This mechanism can effectively predict and warn of potential system overload problems in advance, helping the operation and maintenance personnel to take capacity expansion or current limiting measures in advance to ensure that the user experience is not affected. Through this quantitative and logically refined alarm mechanism, the operation and maintenance team can diagnose and handle faults more quickly and accurately, significantly improving the operation and maintenance efficiency.

[0047] It should be noted that creating a fault management knowledge base and providing automated processing suggestions include recording fault types, processing steps and historical experience, creating a fault management knowledge base, and providing automated processing suggestions by matching current faults with historical data.

[0048] It should also be noted that a fault management knowledge base is created and automated processing suggestions are provided. Through quantitative analysis of historical fault data, fault types are refined into categories with clear parameters, including detailed descriptions of faults, quantitative indicators of impact (such as the number of minutes of service interruption and the percentage of affected users), and statistical characteristics of fault occurrence. For processing steps, intelligent algorithms, such as decision trees and logistic regression, are used to define standard processing procedures for each type of fault. These procedures are designed to achieve rapid diagnosis, accurate isolation, and effective recovery of faults. In terms of historical experience, the knowledge base includes fault processing cases that have undergone time series analysis. These cases record in detail the causal logic, processing path, and final effect of the fault. When a new fault occurs, the current fault characteristics are compared with the historical data in the knowledge base, and the most matching processing suggestions are automatically recommended. For example, when the cloud service encounters sudden traffic and causes performance degradation, the system will intelligently recommend expansion plans or optimize configuration parameters based on historical successful cases, thereby ensuring the intelligence and efficiency of cloud problem fault analysis and improving the reliability and user experience of cloud services.

[0049] Example 2 is an embodiment of the present invention, which provides an intelligent cloud problem fault analysis method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0050] First, in order to verify the effectiveness of the fault feature extraction and fault prediction model of the present invention, a cloud-based system monitoring experimental platform was built. The platform adopts a multi-node hybrid architecture and is composed of multiple servers, databases and network devices. Typical system performance indicators, such as CPU load, memory usage, network bandwidth and latency, are selected as feature data in the experiment. In the data collection stage, the above performance indicators are collected in real time by the monitoring agent deployed at each node, and transmitted to the central database through a secure channel. In order to ensure the accuracy of the data, the Pandas and NumPy libraries in Python are used to preprocess the data, including missing value filling and outlier correction. The mean interpolation and standard deviation methods are used for processing. The system event and application log data are organized into time series format for storage for subsequent analysis.

[0051] In the feature extraction stage, the Pearson correlation coefficient between each performance indicator and the fault event label is calculated to identify the features most relevant to the fault event. During this process, it was found that the correlation coefficients of CPU load and network delay reached 0.67 and 0.55 respectively, exceeding the evaluation standard of 0.5, and therefore they were selected as key fault features. Subsequently, these feature data were standardized, and the categorical and numerical data were converted into an input format acceptable to the machine learning model through normalization and One-Hot Encoding.

[0052] Next, the trained XGBoost model is used for fault prediction. The newly collected data is input into the model, including equipment status and alarm information. The model adopts an integrated learning method to output the probability of occurrence of fault categories through the voting mechanism of multiple weak learners. Through the set threshold, the system generates alarms for abnormal events with a probability higher than the set value, and records its feature importance and possible failure modes. The determination and location of the fault type are based on the feature contribution of the model output, as well as the analysis of multi-dimensional information such as decision tree paths and time series changes.

[0053] During the experiment, a fault management knowledge base was also established to record the processing steps and historical experience of various faults. The knowledge base provides automated processing suggestions through decision trees and logistic regression algorithms, and uses the experience accumulated from historical cases to improve the accuracy and efficiency of fault response.

[0054] Refer to Table 1 for a comparative analysis of the experimental data.

[0055] Table 1 Experimental data record table

[0056]

[0057] The experimental results show that the recognition and location of cloud system faults have been effectively improved through the application of fault feature extraction and prediction models. As can be seen from Table 1, the probability of system failure increases significantly during periods of high CPU load and network delay, especially when the network delay exceeds 200ms and the CPU utilization exceeds 75%, the fault prediction accuracy of the model reaches more than 80%, which proves the effectiveness of the feature set in fault analysis. Compared with the traditional fault monitoring method that only relies on a single performance indicator, the multi-feature extraction and analysis scheme provided by the present invention effectively improves the comprehensiveness and flexibility of fault detection.

[0058] The model uses XGBoost's feature importance evaluation function to quickly identify and locate key features related to faults, thereby improving the accuracy of fault prediction. At the same time, through the establishment of a fault management knowledge base and automated processing suggestion functions, it significantly shortens the fault response time and enhances the degree of automation of fault diagnosis and recovery, which are all lacking in traditional fault detection methods. With the continuous accumulation and learning of historical data, the recognition accuracy and response speed of the fault prediction model have also been further improved, providing scientific and powerful support for cloud stability maintenance.

[0059] Compared with the existing technology, the present invention shows unique innovation in feature extraction, fault prediction and automated processing suggestions. Especially when facing such complex and changeable clouds, it provides an efficient and reliable fault management solution, adapts to the needs of modern cloud operation and maintenance, and demonstrates foresight and practical operational value.

[0060] Example 3, reference Figure 2 , is an embodiment of the present invention, and provides an intelligent cloud-based problem fault analysis system, including a data processing module, a fault analysis module, and a fault management module.

[0061] The data processing module is used to collect and pre-process cloud system operation data; the fault analysis module is used to predict and locate faults and analyze historical data; the fault management module is used to establish an alarm mechanism, create a fault management knowledge base, provide automated processing suggestions, and continuously optimize system performance.

[0062] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0063] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0064] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0065] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0066] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent cloud problem fault analysis method, characterized in that: include: Extract data features and perform the first extraction of fault features; Conduct fault prediction and location; Establish a first alert mechanism, create a fault management knowledge base, and provide automated processing suggestions.

2. The intelligent cloud problem fault analysis method according to claim 1, characterized in that: The extracting of data features includes collecting system operation data and performing preprocessing.

3. The intelligent cloud problem fault analysis method according to claim 2, characterized in that: The first extraction of fault features includes extracting fault features and performing standardization processing on feature data.

4. The intelligent cloud problem fault analysis method according to claim 3, characterized in that: The fault prediction and location includes identifying the fault mode and type.

5. The intelligent cloud problem fault analysis method according to claim 4, characterized in that: The fault prediction and location also includes analyzing historical data to identify normal modes and abnormal modes.

6. The intelligent cloud problem fault analysis method according to claim 5, characterized in that: The establishment of the first alarm mechanism includes setting a threshold value. If the indicator exceeds a preset range, it is determined to be a fault, and the system automatically generates an alarm and notifies the operation and maintenance personnel.

7. The intelligent cloud problem fault analysis method according to claim 6, characterized in that: The creating of the fault management knowledge base and providing automated processing suggestions include recording the fault type, processing steps and historical experience, creating the fault management knowledge base, and providing automated processing suggestions by matching the current fault with historical data.

8. A system using the intelligent cloud problem fault analysis method according to any one of claims 1 to 7, characterized in that: Including data processing module, fault analysis module and fault management module; The data processing module is used to collect and pre-process the cloud system operation data; The fault analysis module is used to predict and locate faults and analyze historical data; The fault management module is used to establish an alarm mechanism, create a fault management knowledge base, provide automated processing suggestions, and continuously optimize system performance.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent cloud problem fault analysis method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent cloud problem fault analysis method described in any one of claims 1 to 7 are implemented.

Images