CN117909112A - Automatic diagnosis method, device, equipment and storage medium for application program faults - Google Patents
Automatic diagnosis method, device, equipment and storage medium for application program faults Download PDFInfo
- Publication number
- CN117909112A CN117909112A CN202311789590.6A CN202311789590A CN117909112A CN 117909112 A CN117909112 A CN 117909112A CN 202311789590 A CN202311789590 A CN 202311789590A CN 117909112 A CN117909112 A CN 117909112A
- Authority
- CN
- China
- Prior art keywords
- data
- information
- fault
- application
- processed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 110
- 238000003745 diagnosis Methods 0.000 title claims abstract description 70
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 69
- 238000004458 analytical method Methods 0.000 claims abstract description 61
- 238000013139 quantization Methods 0.000 claims abstract description 57
- 239000011159 matrix material Substances 0.000 claims description 49
- 238000010606 normalization Methods 0.000 claims description 44
- 230000009467 reduction Effects 0.000 claims description 35
- 230000004044 response Effects 0.000 claims description 27
- 238000000605 extraction Methods 0.000 claims description 21
- 238000003066 decision tree Methods 0.000 claims description 20
- 238000012706 support-vector machine Methods 0.000 claims description 17
- 238000007621 cluster analysis Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 238000007637 random forest analysis Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 13
- 238000004140 cleaning Methods 0.000 claims description 12
- 238000012847 principal component analysis method Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 238000002405 diagnostic procedure Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 16
- 238000000513 principal component analysis Methods 0.000 description 11
- 238000007405 data analysis Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000003064 k means clustering Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000002759 z-score normalization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention relates to the field of application fault diagnosis and discloses an automatic application fault diagnosis method, device, equipment and storage medium. The method comprises the following steps: obtaining quantization index information by obtaining application information to be processed and converting the information to be processed; normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information; establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information; and carrying out root cause analysis on the application fault information to obtain a fault source of the application. The invention provides an automatic diagnosis method for faults of an application program, which can rapidly and accurately locate and solve the faults in the application program through quantitative processing and intelligent analysis, and improves the efficiency and the accuracy of the diagnosis of the application faults.
Description
Technical Field
The present invention relates to the field of application fault diagnosis technologies, and in particular, to an automatic application fault diagnosis method, apparatus, device, and storage medium.
Background
Currently, application fault diagnostics rely primarily on manual analysis, including log review, performance monitoring, and user feedback. The method is long in time consumption, high in professional knowledge requirement and easy to make mistakes. In a rapidly iterated software environment, the manual diagnosis method is difficult to adapt to the real-time and accuracy requirements of fault diagnosis. In addition, the conventional method has limitations in handling the faults of the complex system, and is difficult to effectively analyze and locate the problems caused by multi-factor interleaving.
Existing application fault diagnosis techniques rely primarily on experienced technicians to manually analyze application logs, performance metrics, and user feedback. This approach has significant limitations: first, manual analysis is time consuming and inefficient, especially in the face of large-scale and complex systems; secondly, because of relying on personal experience and knowledge, the diagnosis result may have deviation, and consistency and accuracy are difficult to ensure; furthermore, with the increasing complexity of application architecture and rapid iteration of technology, conventional fault diagnosis methods have been difficult to meet the requirements of high efficiency and real time.
Accordingly, the prior art is still in need of improvement and development.
Disclosure of Invention
The invention mainly aims to solve the problems that in the prior art, fault diagnosis is dependent on manual analysis, and the manual analysis requires deep analysis by an analyzer with higher professional ability, and requires longer time consumption, so that the efficiency of fault diagnosis is lower.
The first aspect of the present invention provides an automatic fault diagnosis method for an application program, comprising: acquiring information to be processed of an application, and converting the information to be processed to obtain quantization index information; normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information; establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information; and carrying out root cause analysis on the application fault information to obtain a fault source of the application.
Optionally, in a first implementation manner of the first aspect of the present invention, the information to be processed includes an application log and performance data; the step of obtaining the information to be processed of the application and converting the information to be processed to obtain the quantization index information comprises the following steps: acquiring the application data and the performance data, and performing data cleaning on the application data and the performance data to obtain the processed application data and the processed performance data; and carrying out quantization processing on the processed application data and the performance data to obtain quantization index information.
Optionally, in a second implementation manner of the first aspect of the present invention, the quantization index information includes service response time data and request processing error rate data; the normalization information comprises normalization values corresponding to the service response time data and each item of data of the request processing error rate data; the step of carrying out normalization processing on the quantization index information to obtain normalization information and carrying out feature extraction on the normalization information to obtain key feature information comprises the following steps: respectively calculating a first difference value between an actual value and a minimum value corresponding to each item of data in the service response time data and the request processing error rate data, and a second difference value between a maximum value and the minimum value corresponding to each item of data; taking the product of the first difference value and the second difference value of each item of data as a normalization value corresponding to each item of data of the service response time data and the request processing error rate data; performing data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data; and clustering the dimensionality reduction data by adopting a cluster analysis method to obtain key characteristic information.
Optionally, in a third implementation manner of the first aspect of the present invention, the dimension-reducing data is a dimension-reducing data matrix; the step of performing data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data comprises the following steps: forming an initial data matrix according to all data sets in the normalization information; extracting a feature vector from the obtained covariance matrix to obtain a principal component matrix; and calculating the respective corresponding products of the initial data matrix and the principal component matrix to obtain a reduced-dimension data matrix.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the step of clustering the reduced dimension data by using a cluster analysis method to obtain key feature information includes: selecting K points from the dimension reduction data as initial clustering centers, calculating the distance between each point in the dimension reduction data and each initial clustering center, and distributing each point in the dimension reduction data to the initial clustering center closest to the point to obtain a current cluster of the initial clustering center; and recalculating the cluster center of each current cluster according to the current cluster until the change of the cluster center tends to be stable, and taking the cluster center corresponding to the current cluster as key feature information.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the step of establishing a fault model identification model by using a classification method, inputting the key feature information into the fault model identification model, and outputting application fault information includes: establishing a fault model identification model based on a support vector machine and a random forest method, acquiring historical application fault information as a training set, and training the fault model identification model according to the training set to obtain a fault model identification model meeting type identification conditions; inputting the key characteristic information into the trained fault model identification model, and identifying the trained fault model identification model according to the key characteristic information to output application fault information.
Optionally, in a sixth implementation manner of the first aspect of the present invention, the step of performing root cause analysis on the application fault information to obtain a fault source of the application includes: processing the application fault information by adopting a decision tree method to obtain a fault root cause of the application; and analyzing the fault root cause by adopting an association rule method to obtain an applied fault root cause.
A second aspect of the present invention provides an automatic application program failure diagnosis apparatus, comprising: the information acquisition and quantization module is used for acquiring information to be processed of an application and converting the information to be processed to obtain quantization index information; the feature extraction module is used for carrying out normalization processing on the quantization index information to obtain normalized information, and carrying out feature extraction on the normalized information to obtain key feature information; the fault mode identification module is used for establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model and outputting application fault information; and the root cause analysis module is used for carrying out root cause analysis on the application fault information to obtain a fault source of the application.
Optionally, in a first implementation manner of the second aspect of the present invention, the information acquiring and quantizing module includes: the information acquisition and processing unit is used for acquiring the application data and the performance data, and carrying out data cleaning on the application data and the performance data to obtain the processed application data and the processed performance data; and the index quantization unit is used for carrying out quantization processing on the processed application data and the performance data to obtain quantization index information.
Optionally, in a second implementation manner of the second aspect of the present invention, the feature extraction module includes: a data calculation unit, configured to calculate a first difference value between an actual value and a minimum value, which correspond to each item of data, in the service response time data and the request processing error rate data, and a second difference value between a maximum value and the minimum value, which correspond to each item of data, respectively; a normalized value determining unit, configured to take a product of the first difference value and the second difference value of each item of data as a normalized value corresponding to each item of data of the service response time data and the request processing error rate data; the data dimension reduction unit is used for carrying out data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data; and the cluster analysis unit is used for clustering the dimensionality reduction data by adopting a cluster analysis method to obtain key characteristic information.
Optionally, in a third implementation manner of the second aspect of the present invention, the data dimension reduction unit includes: a data set subunit, configured to form an initial data matrix according to all data sets in the normalization information; the feature vector extraction subunit is used for extracting feature vectors from the obtained covariance matrix to obtain a main component matrix; and the data matrix determining subunit is used for calculating the respective corresponding products of the initial data matrix and the principal component matrix to obtain a reduced-data matrix.
Optionally, in a fourth implementation manner of the second aspect of the present invention, the cluster analysis unit includes: the current cluster determining subunit is used for selecting K points from the dimension reduction data as initial cluster centers, calculating the distance between each point in the dimension reduction data and each initial cluster center, and distributing each point in the dimension reduction data to the initial cluster center nearest to the point to obtain a current cluster of the initial cluster center; and the cluster center determining subunit is used for recalculating the cluster center of each current cluster according to the current cluster until the change of the cluster center tends to be stable, and taking the cluster center corresponding to the current cluster as key characteristic information.
Optionally, in a fifth implementation manner of the second aspect of the present invention, the fault mode identifying module includes: the model construction unit is used for establishing a fault model identification model based on a support vector machine and a random forest method, acquiring historical application fault information as a training set, and training the fault model identification model according to the training set to obtain a fault model identification model meeting type identification conditions; and the application obstacle recognition unit is used for inputting the key characteristic information into the trained fault model recognition model, recognizing the trained fault model recognition model according to the key characteristic information and outputting application fault information.
Optionally, in a sixth implementation manner of the second aspect of the present invention, the root cause analysis module 80 includes: the first root cause analysis unit is used for processing the application fault information by adopting a decision tree method to obtain the fault root cause of the application; and the second root cause analysis unit is used for analyzing the fault root cause by adopting an association rule method to obtain an applied fault root cause.
A third aspect of the present invention provides an application program failure automatic diagnosis apparatus, comprising: a memory and at least one processor, the memory having computer readable instructions stored therein, the memory and the at least one processor being interconnected by a line; the at least one processor invokes the computer-readable instructions in the memory to cause the application fault automatic diagnostic apparatus to perform the steps of the application fault automatic diagnostic method as described above.
A fourth aspect of the present invention provides a computer-readable storage medium having stored therein computer-readable instructions which, when run on a computer, cause the computer to perform the steps of the application fault automatic diagnosis method as described above.
The beneficial effects are that: in the technical scheme of the invention, the information to be processed of the application is obtained, and the information to be processed is converted to obtain quantization index information; normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information; establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information; and carrying out root cause analysis on the application fault information to obtain a fault source of the application. The invention provides an automatic diagnosis method for faults of an application program, which can rapidly and accurately locate and solve the faults in the application program through quantitative processing and intelligent analysis, and improves the efficiency and the accuracy of the diagnosis of the application faults.
Drawings
FIG. 1 is a first flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 2 is a second flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 3 is a third flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 4 is a fourth flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 5 is a fifth flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 6 is a sixth flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 7 is a seventh flowchart of an automatic application fault diagnosis method according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of an automatic application fault diagnosis device according to an embodiment of the present invention;
FIG. 9 is a schematic diagram of another configuration of an automatic application fault diagnosis apparatus according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of an automatic application fault diagnosis device according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides an automatic fault diagnosis method, device and equipment for an application program and a storage medium, wherein the method, device and equipment are used for acquiring information to be processed and converting the information to be processed to obtain quantitative index information; normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information; establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information; and carrying out root cause analysis on the application fault information to obtain a fault source of the application. The invention solves the problem that the fault diagnosis is low in efficiency due to the fact that the application of the fault diagnosis depends on manual analysis, and the manual analysis requires deep analysis by an analyzer with high professional ability and takes long time.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present invention is described below, referring to fig. 1, and a first embodiment of an automatic application fault diagnosis method in an embodiment of the present invention includes:
S101, acquiring information to be processed of an application, and converting the information to be processed to obtain quantization index information;
In this embodiment, an application log, performance metrics, and user feedback are collected; the collected unstructured data is converted into quantization indices such as error rate, response time, etc. The quantization index is an index which can be embodied by specific data, and reflects the state of nature or society by data which are linearly transformed within a certain range; the quantization index may be based on any subject or object, corresponding to a certain property of that subject or object; in data analysis, quantitative indicators are widely used, and their main roles include providing objective data and indicators, supporting data analysis and research. The collected unstructured data, such as application logs, performance metrics, user feedback, etc., need to be converted into quantitative metrics, such as error rates, response times, etc., which can be used to evaluate performance, e.g., the error rates can reflect stability and performance of the application, and the response times can reflect the response speed of the application. In this way, the quantization index converts abstract concepts and subjective things into specific values and indices, making them more objective and comparable.
S102, carrying out normalization processing on the quantization index information to obtain normalization information, and carrying out feature extraction on the normalization information to obtain key feature information;
In this embodiment, data cleaning and normalization processing are applied to improve data quality; the quantitative index is extracted by a machine learning algorithm (such as principal component analysis and cluster analysis), key performance indexes are identified, and the original data is extracted by principal component analysis, cluster analysis and the like. The method realizes automatic fault positioning and diagnosis through index quantification and intelligent analysis technology, adopts technical means of multiple layers of data preprocessing, feature extraction, fault mode identification, root cause analysis and the like, and realizes a full-automatic flow from original unstructured data to root cause analysis.
S103, establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information;
In this embodiment, a classification algorithm (such as a support vector machine and a random forest) is used to establish a fault mode identification model; the training model identifies different types of application failures, such as memory leaks, service breaks. And carrying out fault mode identification by using algorithms such as a support vector machine, a random forest and the like.
S104, performing root cause analysis on the application fault information to obtain a fault source of the application;
In the embodiment, algorithms such as decision trees, association rules and the like are applied to carry out root cause analysis; and analyzing the correlation among the quantitative indexes to determine possible sources of faults. The method has the advantages that the automatic reasoning of the fault root causes is realized by means of decision trees, association rules and the like, the fault diagnosis and analysis of complex and large-scale systems can be realized, and the method has strong universality.
According to the embodiment of the application, through combining the steps of data preprocessing, feature extraction, fault mode identification and root cause analysis, a high-efficiency intelligent analysis method for the application root cause is provided, and the application fault root cause can be automatically, accurately and rapidly positioned. Firstly, in a data preprocessing stage, the quality and consistency of data are ensured through data cleaning and normalization processing, and a foundation is laid for subsequent analysis; then, the feature extraction is carried out by adopting machine learning technologies such as Principal Component Analysis (PCA) and cluster analysis, so that key performance indexes are effectively identified from a large amount of data, and the complexity of the data is greatly reduced; in the aspect of fault mode identification, the technical scheme uses classification algorithms such as a Support Vector Machine (SVM) and a random forest, and the algorithms not only improve the accuracy of fault type judgment, but also enhance the adaptability of the model to a new fault mode; in addition, the root cause analysis by the decision tree and the association rule algorithm can intuitively display the decision path and the potential cause of the fault occurrence, and provides an effective solution for the quick positioning problem. In general, the embodiment of the application has the greatest advantages of automation and intelligence, can process a large-scale data set, accurately identify and locate application faults, and remarkably improve the fault processing efficiency. The method based on data driving not only improves the accuracy of fault analysis, but also reduces the dependence on manual expertise, so that the system is more robust and reliable. Therefore, it has great applicability and value to rapidly developing and increasingly complex modern software environments.
According to the embodiment of the application, by introducing index quantization and intelligent algorithm, automatic root cause analysis of the application fault is realized, unstructured data in the fault diagnosis process is converted into quantifiable indexes by index quantization, and the intelligent algorithm utilizes the quantified data to carry out deep analysis. After the application root cause intelligent analysis system based on index quantification is introduced, the fault processing efficiency is obviously improved, the system can rapidly identify and locate fault causes, and the fault diagnosis time is greatly shortened. Meanwhile, the application of the intelligent algorithm reduces the dependence on manual expertise and improves the accuracy of fault processing. In addition, the system can process a large amount of data, is suitable for a complex and changeable modern software environment, and improves the stability and reliability of the system.
Referring to fig. 2, a second embodiment of the automatic application fault diagnosis method according to the embodiment of the present invention includes:
S201, acquiring the application data and the performance data, and performing data cleaning on the application data and the performance data to obtain the processed application data and the processed performance data;
S202, carrying out quantization processing on the processed application data and the performance data to obtain quantization index information.
In this embodiment, the information to be processed includes application log and performance data, and data cleaning and normalization processing are applied to improve data quality; and (3) performing feature extraction on the quantization indexes by using a machine learning algorithm (such as principal component analysis and cluster analysis), and identifying key performance indexes. Specifically, the step of data cleaning removes invalid or erroneous data by detecting and removing incomplete, erroneous or irrelevant data, in order to improve the quality and accuracy of the data set. Unstructured data such as application logs, performance metrics, user feedback, etc. are collected and are raw and unprocessed and therefore cannot be directly used for subsequent analysis or decision making, and in order to convert the unstructured data into a format that can be subjected to deep analysis, it is necessary to quantize the data, that is, to convert the unstructured data into quantifiable metrics, for example, the error rate, response time, etc. are the result of quantifying the raw data, and these quantified metrics can better reflect the performance and stability of the application.
The data cleaning mainly performs normalized processing on original data, such as unstructured data including program logs, performance data and the like, and common methods include repeated value processing, null value processing, abnormal value processing and the like so as to improve the data quality. For example, after unstructured data such as application logs, performance indexes and user feedback are collected, some noise data or abnormal data may be encountered, and then the invalid information needs to be removed through data cleaning to ensure the accuracy of subsequent analysis; specifically, for repeated data, invalid samples can be eliminated by a case-by-case process; for both the null and outliers, corresponding processing methods may be employed to fill or correct these missing or outlier portions. The step of performing a data cleaning process to improve the data quality further strengthens the importance of the quantization index, since the quantization index accurately reflects the real situation of the application only if the data quality is guaranteed.
Referring to fig. 3, a third embodiment of an automatic application fault diagnosis method according to an embodiment of the present invention includes:
S301, respectively calculating a first difference value between an actual value and a minimum value, which correspond to each item of data, in the service response time data and the request processing error rate data, and a second difference value between a maximum value and the minimum value, which correspond to each item of data;
S302, taking the product of the first difference value and the second difference value of each item of data as a normalization value corresponding to each item of data of the service response time data and the request processing error rate data;
s303, performing data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data;
S304, clustering the reduced data by adopting a cluster analysis method to obtain key characteristic information.
In this embodiment, the quantization index information includes service response time data and request processing error rate data; the normalization information comprises normalization values corresponding to the service response time data and each item of data of the request processing error rate. Normalization processing, namely data normalization, wherein the formula is as follows:
"(\text { normalized value = \frac { \text { actual value } -, text { minimum value } - { \text { maximum } -, text { minimum value } -; parameter interpretation: normalization processes convert the data to a range of 0 to 1, helping to process data of different metrics. That is, the formula is:
Normalized value= (actual value-minimum value)/(maximum value-minimum value); the two metrics, service response time and request handling error rate, are typically calculated for each individual request or task. For example, for a service response time, it will be different for each independent request, so for a plurality of different response time data, normalization processing is finally performed respectively, to form respective corresponding normalized values. For the request processing error rate, processing is also performed for each specific request to obtain respective error rate data, and then normalization processing is performed on the error rate data. That is, the data normalization process for both of these two indices is performed on a large number of specific data separately.
The normalization processing is mainly one-step processing of the quantized indexes after the quantized indexes are collected and quantized, and the data normalization/normalization processing is used for eliminating the dimension influence among all evaluation indexes and solving the comparability problem among the data indexes; different evaluation indexes often have different dimensions and dimension units, the situation can influence the result of data analysis, and in order to eliminate the dimension influence among indexes, normalization or standardization processing is needed for the quantized indexes, so that all indexes are in the same order of magnitude, and the comparability problem among the data indexes is solved; the normalized/standardized data, each index being in the same order of magnitude, is more suitable for comprehensive comparison and evaluation, and each index being limited to a certain range (e.g., [0,1] or [ -1,1 ]), which helps to eliminate adverse effects caused by singular sample data (i.e., sample vectors with feature vectors that are particularly large or particularly small relative to other input samples). For example, for error rate, response time equating indicators, different units or dimensions may lead to deviations in the analysis results, and therefore normalization processing is required to eliminate such effects. In addition, through feature extraction, key performance indexes can be identified from the quantization indexes, and a basis is provided for subsequent fault mode identification and root cause analysis. Data normalization is an important step of eliminating the dimension and dimension unit difference of each evaluation index, and aims to solve the problem of incomparability between quantization indexes caused by different measurement units or scales. For example, the two indexes of service response time and request processing error rate may be respectively expressed in seconds and percentages, and the numerical ranges are also greatly different, and if normalization processing is not performed, deviation occurs in data analysis. Specifically, the normalization process is to scale the original index data, remove the unit limitation of the data, and convert the data into dimensionless pure values. After such processing, all the metrics are limited to a range (e.g., [0,1] or [ -1,1 ]) such that they can be compared and weighted. Common normalization methods are min-max normalization, z-score normalization, etc.
Referring to fig. 4, a fourth embodiment of the automatic application fault diagnosis method according to the embodiment of the present invention includes:
s401, forming an initial data matrix according to all data sets in the normalization information;
s402, extracting a feature vector from the obtained covariance matrix to obtain a principal component matrix;
S403, calculating respective corresponding products of the initial data matrix and the principal component matrix to obtain a data-reduction matrix.
In this embodiment, the dimension-reduction data is a dimension-reduction data matrix; principal Component Analysis (PCA), formula \text { Y } = \text { X } \times\text { P }; parameter interpretation: x is an original data matrix, P is a main component extracted from a covariance matrix, and Y is dimension-reduced data; the purpose is to reduce the data dimension, highlighting important features. That is, the formula is: y=x×p.
In Principal Component Analysis (PCA), the raw data matrix X is a collection of data to be subjected to a dimension reduction process, each element representing a particular feature or variable. Covariance matrices are calculated from the original data matrix, which reveals the correlation between the variables in the data. The principal components are key components extracted from the covariance matrix, representing the main direction of change of the original data. In practice, the principal components are eigenvectors of the covariance matrix, and eigenvalues corresponding to these eigenvectors represent the variance of the data in the direction of the respective principal component. In other words, the principal component is a representation of the original data in a new spatial coordinate system whose basis is constituted by eigenvectors of the covariance matrix of the original data. The reduced-dimension data matrix Y can be obtained by multiplying the original data matrix X by the principal component matrix P. The process projects the original data into a new space represented by a principal component, the reduced-dimension data retains as much variation information of the original data as possible, and the dimension of the data is reduced, so that the data can be better understood and interpreted, and important characteristics are highlighted.
It will be appreciated that covariance matrices are an important tool in multivariate statistical analysis, which is used to measure the correlation between variables in a dataset. The definition of covariance is the overall error of two random variables. If this concept is extended to multiple variables we get a covariance matrix. For a dataset with n features, the covariance matrix is a table of n x n, each element being the covariance between the corresponding two features. Specifically, if there are two features x and y, the covariance between them is the sum of the products of the two features' respective degrees of deviation from the mean divided by the number of samples minus one. In the covariance matrix, the elements on the diagonal are variances of the individual features, describing the extent to which the data deviates from its mean. The elements at other locations are covariances between different features reflecting the relationship between the features. Thus, by analyzing the covariance matrix, the degree of correlation of the individual features in the dataset and their contribution to the overall variance can be known.
Referring to fig. 5, a fifth embodiment of the automatic application fault diagnosis method according to the embodiment of the present invention includes:
S501, selecting K points from the dimension reduction data as initial clustering centers, calculating the distance between each point in the dimension reduction data and each initial clustering center, and distributing each point in the dimension reduction data to the initial clustering center closest to the point to obtain a current cluster of the initial clustering center;
S502, recalculating the clustering center of each current cluster according to the current cluster until the change of the clustering center tends to be stable, and taking the clustering center corresponding to the current cluster as key feature information.
In the embodiment, the clustering analysis, namely K-means clustering, comprises the steps of randomly selecting K points as initial clustering centers; assigning each point to the nearest cluster center; re-computing the center of each cluster; the two steps are repeated until the clusters are no longer changing.
K-means clustering is an unsupervised learning algorithm, mainly used for computing data aggregation. The main objective of the algorithm is to find the best K Cluster (Cluster) partitioning schemes through continuous iteration, so that the loss function corresponding to the clustering result is minimum. In performing K-means clustering, the algorithm will first randomly select K points as the initial cluster centers. Then, for each point in the dataset, the algorithm will calculate its distance from the respective cluster center and assign each point to the cluster represented by the cluster center closest to it. Then, the cluster center of each cluster is recalculated according to the newly formed cluster. This process is repeated until the cluster center changes tend to stabilize, i.e., local optima are reached.
It should be noted that Principal Component Analysis (PCA) is a dimension reduction technique aimed at reducing the dimensionality of the data while preserving maximum variability. K-means clustering can be performed on the dimension-reduced data to achieve a better clustering effect.
Referring to fig. 6, a sixth embodiment of an automatic application fault diagnosis method according to an embodiment of the present invention includes:
S601, building a fault model identification model based on a support vector machine and a random forest method, acquiring historical application fault information as a training set, and training the fault model identification model according to the training set to obtain a fault model identification model meeting type identification conditions;
S602, inputting the key feature information into the trained fault model identification model, and identifying the trained fault model identification model according to the key feature information to output application fault information.
In the embodiment, a Support Vector Machine (SVM) is provided with a formula of \text { w } \ cdot \text { x } +b=0\); parameter interpretation: w is a weight vector; x is an input feature vector; b is a bias term; the objective is for classification tasks, the classification boundaries being determined by maximizing the separation between classes. Namely the formula:
w·x+b=0. The random forest comprises the following steps: 1. randomly selecting a plurality of sample subsets from the original dataset; 2. constructing a decision tree for each subset; 3. each decision tree predicts independently; 4. majority voting determines the final classification result; the method is used for solving the problem of classification of the fault modes and improving the accuracy and the robustness of the model.
The fault pattern recognition step truly comprises two refinement steps of a Support Vector Machine (SVM) and a random forest. The Support Vector Machine (SVM) is suitable for processing classification tasks of small and medium-sized complex data sets, and the basic formula is as follows: (\text { w } \ cdot \text { x } +b=0), where w is a weight vector, x is an input feature vector, b is a bias term, and the main objective of the SVM is to determine classification boundaries by maximizing the spacing between classes, so the SVM is a nonlinear classifier that can deal with nonlinear problems. Random forests can handle both continuous and discrete variables, the basic idea of which is to combine multiple decision trees together and average them. The processing flow of the random forest comprises the steps of randomly selecting a plurality of sample subsets from an original data set, constructing decision trees for each subset, independently predicting each decision tree, and determining a final classification result by majority voting, so that the random forest is suitable for processing high-dimensional data and large sample problems.
Referring to fig. 7, a seventh embodiment of an automatic application fault diagnosis method according to an embodiment of the present invention includes:
s701, processing the application fault information by adopting a decision tree method to obtain a fault root cause of the application;
S702, analyzing the fault root cause by adopting an association rule method to obtain an applied fault root cause.
In this embodiment, the decision tree comprises the following steps: 1. selecting optimal characteristics to perform node splitting; 2. repeating step 1 for each branch until a termination condition is reached; 3. each leaf node represents a classification decision; the aim is to visualize complex decision paths through decision trees, which is helpful for understanding the root cause of the fault. The formula of the association rule: "(\text { support = \frac { \text { number of times rule occurs } { \text { total number of samples }); parameter interpretation: the support degree is the frequency of occurrence of the rule in the data set, and the confidence degree is the probability of simultaneous occurrence of the condition and the result; the purpose is to find out the correlation between quantization indexes and reveal the potential failure cause.
Root cause analysis is an important method for locating and solving the root cause of the problem, and the embodiment of the application mainly uses two algorithms of decision trees and association rules for analysis. Firstly, the decision tree performs node splitting by circularly selecting an optimal feature until bifurcation cannot be continued, branches in the decision tree represent decision rules (IF THEN rules), and leaf nodes represent results of the IF THEN rules, so that the decision tree has strong interpretation and is helpful for understanding the root cause of faults. The association rules are then used to discover correlations between the quantified indicators, revealing potential causes of failure, the main parameters of the association rules including a degree of support, which is the frequency with which the rules appear in the dataset, and a degree of confidence, which is the probability that the conditions and results occur simultaneously. Decision trees may help understand the possible causes of the failure, and association rules may further reveal the correlation between these causes. By combining the two methods, root cause analysis can be performed more comprehensively and deeply, so that the real root cause of the problem can be found. The steps together form a systematic fault diagnosis method, which can automatically identify and analyze faults of the application program and provide support for quick and effective problem solving.
The following is a description of the rhyme express case:
k1, collecting application logs and performance data in a rhyme express system;
k2, quantization treatment;
such as conversion service response time, request processing error rate, etc.
K3, determining key performance indexes by utilizing a feature extraction method;
K4, analyzing the fault type through a fault mode identification model;
And K5, determining the cause of the fault by using a root cause analysis algorithm, such as database access delay, resource competition and the like.
After determining the cause of the fault, the following optimization and iteration description are performed:
k6, optimizing the system according to the analysis result;
and K7, periodically reviewing the fault mode, and updating and optimizing a fault mode identification model.
The embodiment of the application discloses an automatic fault diagnosis method based on a data driving and intelligent algorithm. The method can rapidly and accurately locate and solve the faults in the application program through quantitative processing and intelligent analysis, and remarkably improves the efficiency and accuracy of fault processing. The advantages of this approach are particularly pronounced, especially when dealing with complex systems and large amounts of data.
The method comprises the steps of obtaining information to be processed, and converting the information to be processed to obtain quantitative index information; normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information; establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information; and carrying out root cause analysis on the application fault information to obtain a fault source of the application. The invention provides an automatic diagnosis method for faults of an application program, which can rapidly and accurately locate and solve the faults in the application program through quantitative processing and intelligent analysis, and improves the efficiency and the accuracy of the diagnosis of the application faults.
The method for automatically diagnosing an application program failure in the embodiment of the present invention is described above, and the apparatus for automatically diagnosing an application program failure in the embodiment of the present invention is described below, referring to fig. 8, and one embodiment of the apparatus for automatically diagnosing an application program failure in the embodiment of the present invention includes:
the information acquisition and quantization module 50 is configured to acquire information to be processed of an application, and convert the information to be processed to obtain quantization index information;
the feature extraction module 60 is configured to normalize the quantization index information to obtain normalized information, and perform feature extraction on the normalized information to obtain key feature information;
The fault mode recognition module 70 is configured to establish a fault model recognition model by using a classification method, input the key feature information into the fault model recognition model, and output application fault information;
and the root cause analysis module 80 is used for performing root cause analysis on the application fault information to obtain a fault source of the application.
In the embodiment, through quantitative processing and intelligent analysis, the faults in the application program can be rapidly and accurately positioned and solved, and the efficiency and the accuracy of the application fault diagnosis are improved.
Referring to fig. 9, another embodiment of an automatic application fault diagnosis apparatus according to an embodiment of the present invention includes:
the information acquisition and quantization module 50 is configured to acquire information to be processed of an application, and convert the information to be processed to obtain quantization index information;
the feature extraction module 60 is configured to normalize the quantization index information to obtain normalized information, and perform feature extraction on the normalized information to obtain key feature information;
The fault mode recognition module 70 is configured to establish a fault model recognition model by using a classification method, input the key feature information into the fault model recognition model, and output application fault information;
and the root cause analysis module 80 is used for performing root cause analysis on the application fault information to obtain a fault source of the application.
In this embodiment, the information acquisition and quantization module 50 includes:
An information acquisition and processing unit 501, configured to acquire the application data and the performance data, and perform data cleaning on the application data and the performance data to obtain the application data and the performance data after processing;
and the index quantization unit 502 is configured to perform quantization processing on the processed application data and the performance data to obtain quantization index information.
In this embodiment, the feature extraction module 60 includes:
a data calculation unit 601, configured to calculate a first difference value between an actual value and a minimum value corresponding to each data in the service response time data and the request processing error rate data, and a second difference value between a maximum value and the minimum value corresponding to each data;
A normalized value determining unit 602, configured to take a product of the first difference value and the second difference value of each item of data as a normalized value corresponding to each item of data of the service response time data and the request processing error rate data;
the data dimension reduction unit 603 is configured to perform data dimension reduction on the data set formed by all the normalized values by using a principal component analysis method, so as to obtain dimension reduction data;
and the cluster analysis unit 604 is used for clustering the dimensionality reduction data by adopting a cluster analysis method to obtain key characteristic information.
In this embodiment, the data dimension reduction unit 603 includes:
a data set subunit 6031, configured to form an initial data matrix according to all data sets in the normalization information;
A feature vector extraction subunit 6032, configured to extract a feature vector from the obtained covariance matrix, so as to obtain a principal component matrix;
And a data matrix determining subunit 6033, configured to calculate respective corresponding products of the initial data matrix and the principal component matrix, to obtain a reduced-dimension data matrix.
In this embodiment, the cluster analysis unit 604 includes:
A current cluster determining subunit 6041, configured to select K points from the dimension-reduced data as initial cluster centers, calculate a distance between each point in the dimension-reduced data and each initial cluster center, and allocate each point in the dimension-reduced data to the initial cluster center closest to the point, so as to obtain a current cluster of the initial cluster center;
And a cluster center determining subunit 6042, configured to recalculate, according to the current clusters, a cluster center of each current cluster until a change of the cluster center tends to be stable, and use a cluster center corresponding to the current cluster as key feature information.
In this embodiment, the failure mode identifying module 70 includes:
The model construction unit 701 is configured to establish a fault model identification model based on a support vector machine and a random forest method, obtain historical application fault information as a training set, and train the fault model identification model according to the training set to obtain a fault model identification model that meets a type identification condition;
The application obstacle recognition unit 702 is configured to input the key feature information to the trained fault model recognition model, and the trained fault model recognition model recognizes according to the key feature information and outputs application fault information.
In this embodiment, the root cause analysis module 80 includes:
a first root cause analysis unit 801, configured to process the application fault information by using a decision tree method, to obtain a fault root cause of the application;
And a second root cause analysis unit 802, configured to analyze the root cause of the fault by using an association rule method, so as to obtain an applied root cause of the fault.
The invention provides an automatic diagnosis method for faults of an application program, which can rapidly and accurately locate and solve the faults in the application program through quantitative processing and intelligent analysis, and improves the efficiency and the accuracy of the diagnosis of the application faults.
The application fault automatic diagnosis apparatus in the embodiment of the present invention is described in detail above in terms of modularized functional entities in fig. 8 and 9, and the application fault automatic diagnosis device in the embodiment of the present invention is described in detail below in terms of hardware processing.
fig. 10 is a schematic structural diagram of an automatic application fault diagnosis apparatus according to an embodiment of the present invention, where the automatic application fault diagnosis apparatus 10 may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 11 (e.g., one or more processors) and a memory 12, and one or more storage mediums 13 (e.g., one or more mass storage devices) storing an application 133 or data 132. Wherein the memory 12 and the storage medium 13 may be transitory or persistent storage. The program stored in the storage medium 13 may include one or more modules (not shown), each of which may include a series of instruction operations in the automatic application program failure diagnosis apparatus 10. Still further, the processor 11 may be arranged to communicate with the storage medium 13, and execute a series of instruction operations in the storage medium 13 on the application fault automatic diagnosis apparatus 10.
The application fault automatic diagnostic apparatus 10 may also include one or more power supplies 14, one or more wired or wireless network interfaces 15, one or more input output interfaces 16, and/or one or more operating systems 131, such as Windows Serve, mac OS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the device configuration shown in fig. 10 is not limiting of the automatic application fault diagnosis device 10 and may include more or fewer components than shown, or may combine certain components, or may have a different arrangement of components.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, in which instructions are stored which, when executed on a computer, cause the computer to perform the steps of the method for automatically diagnosing an application failure.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the system or apparatus and unit described above may refer to the corresponding process in the foregoing method embodiment, which is not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An automatic application program failure diagnosis method, characterized in that the automatic application program failure diagnosis method comprises the following steps:
Acquiring information to be processed of an application, and converting the information to be processed to obtain quantization index information;
Normalizing the quantization index information to obtain normalized information, and extracting features of the normalized information to obtain key feature information;
establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model, and outputting application fault information;
And carrying out root cause analysis on the application fault information to obtain a fault source of the application.
2. The automatic application program fault diagnosis method according to claim 1, wherein the information to be processed includes an application log and performance data;
The step of obtaining the information to be processed of the application and converting the information to be processed to obtain the quantization index information comprises the following steps:
acquiring the application data and the performance data, and performing data cleaning on the application data and the performance data to obtain the processed application data and the processed performance data;
And carrying out quantization processing on the processed application data and the performance data to obtain quantization index information.
3. The automatic application program failure diagnosis method according to claim 1, wherein the quantization index information includes service response time data and request processing error rate data; the normalization information comprises normalization values corresponding to the service response time data and each item of data of the request processing error rate data;
The step of carrying out normalization processing on the quantization index information to obtain normalization information and carrying out feature extraction on the normalization information to obtain key feature information comprises the following steps:
respectively calculating a first difference value between an actual value and a minimum value corresponding to each item of data in the service response time data and the request processing error rate data, and a second difference value between a maximum value and the minimum value corresponding to each item of data;
Taking the product of the first difference value and the second difference value of each item of data as a normalization value corresponding to each item of data of the service response time data and the request processing error rate data;
performing data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data;
and clustering the dimensionality reduction data by adopting a cluster analysis method to obtain key characteristic information.
4. The automatic application program fault diagnosis method according to claim 3, wherein the dimension-reduction data is a dimension-reduction data matrix;
The step of performing data dimension reduction on the data set formed by all the normalized values by adopting a principal component analysis method to obtain dimension reduction data comprises the following steps:
forming an initial data matrix according to all data sets in the normalization information;
extracting a feature vector from the obtained covariance matrix to obtain a principal component matrix;
and calculating the respective corresponding products of the initial data matrix and the principal component matrix to obtain a reduced-dimension data matrix.
5. The method for automatically diagnosing an application program failure according to claim 3, wherein said step of clustering said reduced dimension data using a cluster analysis method to obtain key feature information comprises:
Selecting K points from the dimension reduction data as initial clustering centers, calculating the distance between each point in the dimension reduction data and each initial clustering center, and distributing each point in the dimension reduction data to the initial clustering center closest to the point to obtain a current cluster of the initial clustering center;
and recalculating the cluster center of each current cluster according to the current cluster until the change of the cluster center tends to be stable, and taking the cluster center corresponding to the current cluster as key feature information.
6. The automatic application program fault diagnosis method according to claim 1, wherein the step of establishing a fault model identification model using a classification method, inputting the key feature information into the fault model identification model, and outputting application fault information comprises:
Establishing a fault model identification model based on a support vector machine and a random forest method, acquiring historical application fault information as a training set, and training the fault model identification model according to the training set to obtain a fault model identification model meeting type identification conditions;
inputting the key characteristic information into the trained fault model identification model, and identifying the trained fault model identification model according to the key characteristic information to output application fault information.
7. The automatic application program fault diagnosis method according to claim 1, wherein the step of performing root cause analysis on the application fault information to obtain a fault source of the application comprises:
Processing the application fault information by adopting a decision tree method to obtain a fault root cause of the application;
And analyzing the fault root cause by adopting an association rule method to obtain an applied fault root cause.
8. An automatic application program failure diagnosis apparatus, comprising:
the information acquisition and quantization module is used for acquiring information to be processed of an application and converting the information to be processed to obtain quantization index information;
The feature extraction module is used for carrying out normalization processing on the quantization index information to obtain normalized information, and carrying out feature extraction on the normalized information to obtain key feature information;
the fault mode identification module is used for establishing a fault model identification model by adopting a classification method, inputting the key characteristic information into the fault model identification model and outputting application fault information;
and the root cause analysis module is used for carrying out root cause analysis on the application fault information to obtain a fault source of the application.
9. An automatic application fault diagnosis device comprising a memory and at least one processor, said memory having computer readable instructions stored therein;
The at least one processor invokes the computer readable instructions in the memory to perform the steps of the application fault automatic diagnostic method of any one of claims 1-7.
10. A computer readable storage medium having computer readable instructions stored thereon, which when executed by a processor, implement the steps of the automatic application fault diagnosis method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311789590.6A CN117909112A (en) | 2023-12-22 | 2023-12-22 | Automatic diagnosis method, device, equipment and storage medium for application program faults |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311789590.6A CN117909112A (en) | 2023-12-22 | 2023-12-22 | Automatic diagnosis method, device, equipment and storage medium for application program faults |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117909112A true CN117909112A (en) | 2024-04-19 |
Family
ID=90695417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311789590.6A Pending CN117909112A (en) | 2023-12-22 | 2023-12-22 | Automatic diagnosis method, device, equipment and storage medium for application program faults |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909112A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118245832A (en) * | 2024-05-28 | 2024-06-25 | 国网山东省电力公司阳信县供电公司 | Fault power failure information generation method, system, electronic equipment and storage medium |
-
2023
- 2023-12-22 CN CN202311789590.6A patent/CN117909112A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118245832A (en) * | 2024-05-28 | 2024-06-25 | 国网山东省电力公司阳信县供电公司 | Fault power failure information generation method, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112418277A (en) | Method, system, medium, and apparatus for predicting remaining life of rotating machine component | |
| CN117236656B (en) | Informationized management method and system for engineering project | |
| CN118428775B (en) | Service data processing method and system based on supply chain management | |
| CN109934301B (en) | Power load cluster analysis method, device and equipment | |
| CN113408548A (en) | Transformer abnormal data detection method and device, computer equipment and storage medium | |
| CN114861788A (en) | A load anomaly detection method and system based on DBSCAN clustering | |
| CN111104242A (en) | Method and device for processing abnormal logs of operating system based on deep learning | |
| CN116483015B (en) | Workshop equipment monitoring method, device, equipment and storage medium | |
| CN117170915A (en) | Data center equipment fault prediction method and device and computer equipment | |
| CN116466672B (en) | Data center machine room parameter regulation and control method based on artificial intelligence and related device | |
| CN117421994A (en) | Edge application health monitoring method and system | |
| CN117909112A (en) | Automatic diagnosis method, device, equipment and storage medium for application program faults | |
| CN119728397B (en) | Network fault prediction method and system | |
| CN113610225A (en) | Quality evaluation model training method and device, electronic equipment and storage medium | |
| CN119359049B (en) | A method and system for risk management of engineering projects | |
| CN119720001A (en) | Data monitoring method and device | |
| CN119168611A (en) | A method and system for intelligent operation and maintenance of equipment based on knowledge mining | |
| CN118245264A (en) | Server fault processing method and device, electronic equipment and medium | |
| CN114596011B (en) | Enterprise data processing method based on artificial intelligence and related device | |
| CN117909173A (en) | Cloud application health degree analysis method and device based on big data cloud platform | |
| CN114399407B (en) | Power dispatching monitoring data anomaly detection method based on dynamic and static selection integration | |
| CN116030955A (en) | Medical equipment state monitoring method and related device based on Internet of things | |
| CN116149899A (en) | Method for acquiring platform fault cause and automatic platform fault checking and disposing method | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| CN114610234A (en) | Storage system parameter recommendation method and related device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |