Method, device, processor and storage medium for recommending root cause processing of multi-component software systems based on indicator correlation

Abstract

The present invention relates to a method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation, comprising the following steps: constructing a topology map of the multi-component software system; generating a system state model using historical indicator data according to the topology map; and generating a root cause recommendation list according to the system state model using indicator data within an abnormal time period under abnormal circumstances. The present invention also relates to a method, device, processor and computer-readable storage medium for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation. The method, device, processor and computer-readable storage medium for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation of the present invention are more reliable than simply judging the abnormality of the indicator's own changes. The method has a feedback mechanism, and the indicator weight will be dynamically adjusted according to the user's feedback to gradually improve its own accuracy, and has the advantages of high real-time performance and high accuracy.

Description

Technical Field

[0001] The present invention relates to the field of multi-component systems, and in particular to the field of automated fault root cause recommendation, and specifically refers to a method, device, processor and computer-readable storage medium thereof for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation. Background Art

[0002] A multi-component software system is a software system composed of multiple components. Each component is responsible for only part of the functions. Components are combined with each other through calling relationships to jointly implement complex and diverse business logic. When such software systems fail, due to the complexity of the internal structure, operation and maintenance personnel need to spend a lot of time to locate the fault and find the root cause of the failure. In this context, implementing automated root cause recommendations can greatly narrow the scope of investigation for operation and maintenance personnel, reduce the time required for operation and maintenance personnel to troubleshoot, and improve the availability of the software system.

[0003] The two major practical challenges facing automated root cause recommendations are the real-time and accuracy of the output. Common input information includes collected indicator data, collected log data, and alarms generated by the monitoring system. In terms of real-time, compared with indicator data and alarm data, the amount of log data is much larger, and the real-time recommendation of root cause based on logs is difficult to guarantee. In terms of accuracy, the accuracy of the output depends greatly on the reliability of the input information. Indicator data is the most objective; the objectivity of log data depends on the log printing rules specified by the software developer, which has a certain degree of uncontrollability; and the alarms generated by the monitoring system are highly dependent on the manually configured alarm rules, which are very likely to have false positives and missed positives.

[0004] Therefore, it is necessary to propose a new solution to overcome the existing challenges. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and to provide a method, device, processor and computer-readable storage medium thereof that meet the requirements of high real-time performance, high accuracy and wide application range for implementing fault root cause recommendation processing for multi-component software systems based on indicator correlation.

[0006] In order to achieve the above-mentioned object, the method, device, processor and computer-readable storage medium thereof for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation of the present invention are as follows:

[0007] The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation has the following main features:

[0008] (1) Construct a topology diagram of a multi-component software system;

[0009] (2) generating a system status model using historical indicator data according to the topology diagram;

[0010] (3) In abnormal situations, a root cause recommendation list is generated based on the system status model using the indicator data within the abnormal time period.

[0011] Preferably, the system status model of step (2) is composed of indicator correlation models of all component nodes in the topology graph, and the indicator correlation model includes a throughput indicator correlation model and a status indicator correlation model.

[0012] Preferably, the step (2) further includes a step of generating a throughput indicator correlation model, which specifically includes the following processing steps:

[0013] (1-2.1) The throughput index of a component node is defined as a dependent variable index, and the throughput indexes of all other component nodes that call the component node to which the throughput belongs are defined as a set of independent variable indexes;

[0014] (1-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time;

[0015] (1-2.3) Use historical data to perform multivariate linear fitting on the mapping relationship between independent variable indicators and dependent variable indicators, obtain the fitting function, and calculate the goodness of fit;

[0016] (1-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data;

[0017] (1-2.5) taking the absolute value of the obtained deviation data, performing multivariate linear fitting on the deviation data after taking the absolute value, and obtaining the deviation fitting function;

[0018] (1-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function as the throughput index correlation model.

[0019] Preferably, the step (2) further includes a step of generating a state indicator correlation model, which specifically includes the following processing steps:

[0020] (2-2.1) defining the status indicator of the component node as a dependent variable indicator, and defining the throughput indicator of the component node to which the status indicator belongs as an independent variable indicator;

[0021] (2-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time;

[0022] (2-2.3) Use historical data to perform univariate nonlinear fitting on the mapping relationship between the independent variable index and the dependent variable index, obtain the fitting function, and calculate the goodness of fit;

[0023] (2-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data;

[0024] (2-2.5) taking the absolute value of the obtained deviation data, performing a univariate nonlinear fitting on the deviation data after taking the absolute value, and obtaining a deviation fitting function;

[0025] (2-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function as a state index correlation model.

[0026] Preferably, the step (3) specifically comprises the following steps:

[0027] (3.1) Calculate the correlation anomaly of each indicator;

[0028] (3.2) Calculate the abnormality of each component node;

[0029] (3.3) Arrange all components in descending order of abnormality to obtain a component recommendation list;

[0030] (3.4) For each component, the indicators of the component are arranged in reverse order according to the degree of relevance anomaly to obtain a recommended indicator list for the component.

[0031] Preferably, the step (3.1) specifically comprises the following steps:

[0032] (3.1.1) Use the fitting function in the indicator correlation model to infer the theoretical data of the dependent variable indicator based on the actual data of the independent variable indicator during the abnormal time period;

[0033] (3.1.2) Using the deviation fitting function in the indicator correlation model, the theoretical deviation data of the dependent variable indicator is calculated based on the actual data of the independent variable indicator during the abnormal time period;

[0034] (3.1.3) Subtract the theoretical data of the dependent variable indicator from the actual data of the dependent variable indicator to obtain the absolute deviation data;

[0035] (3.1.4) Divide the absolute deviation data by the theoretical deviation data of the dependent variable indicator and then multiply it by the goodness of fit to obtain the correlation abnormality of the indicator.

[0036] Preferably, the method further comprises the following steps:

[0037] (4) Feedback is provided to the root cause recommendation table, and the system state model is updated according to the user feedback.

[0038] Preferably, the step of providing feedback on the root cause recommendation table in step (4) specifically includes the following processing steps:

[0039] Confirm the root cause components and abnormal indicators of the software system, expand the indicator list of the root cause components in the root cause recommendation list, and select the abnormal indicators in the list to feed back to the positioning system.

[0040] Preferably, the topology diagram of step (1) includes component nodes, call relationships between component nodes, a unique throughput indicator of each component, and a status indicator of each component.

[0041] Preferably, the root cause recommendation list of step (3) is a component recommendation list arranged in descending order of abnormality, wherein each component can be further expanded into an indicator recommendation list arranged in descending order of relevance abnormality.

[0042] The main feature of the device for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation is that the device comprises:

[0043] a processor configured to execute computer executable instructions;

[0044] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the method for recommending root cause processing of faults for a multi-component software system based on indicator correlation are implemented.

[0045] The processor for implementing recommended root cause processing for a multi-component software system based on indicator correlation has the main feature that the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the various steps of the method for implementing recommended root cause processing for a multi-component software system based on indicator correlation are implemented.

[0046] The computer-readable storage medium has the main feature of storing a computer program thereon, which can be executed by a processor to implement the various steps of the above-mentioned method for recommending root cause processing for a multi-component software system based on indicator correlation.

[0047] The method, device, processor and computer-readable storage medium thereof for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation of the present invention are adopted. The root cause recommendation scheme based on component topology and indicator correlation uses component topology and indicator data as input information, the data volume is relatively small, and the human influence factor is low. The present invention uses the correlation between indicators as the basis for abnormality judgment, which is more reliable than simply judging the abnormality of the indicator's own changes. The present invention has a feedback mechanism, and the indicator weight will be dynamically adjusted according to the user's feedback, gradually improving its own accuracy. Therefore, the present invention has the advantages of high real-time performance and high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The present invention is a flow chart of a method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation. DETAILED DESCRIPTION

[0049] In order to more clearly describe the technical content of the present invention, further description is given below in conjunction with specific embodiments.

[0050] The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation of the present invention comprises the following steps:

[0051] The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation has the following main features:

[0052] (1) Construct a topology diagram of a multi-component software system;

[0053] (2) generating a system status model using historical indicator data according to the topology diagram;

[0054] (3) In abnormal situations, a root cause recommendation list is generated based on the system status model using the indicator data within the abnormal time period.

[0055] As a preferred embodiment of the present invention, the system state model of step (2) is composed of indicator correlation models of all component nodes in the topology graph, and the indicator correlation model includes a throughput indicator correlation model and a state indicator correlation model.

[0056] As a preferred embodiment of the present invention, the step (2) further includes a step of generating a throughput indicator correlation model, which specifically includes the following processing steps:

[0057] (1-2.1) The throughput index of a component node is defined as a dependent variable index, and the throughput indexes of all other component nodes that call the component node to which the throughput belongs are defined as a set of independent variable indexes;

[0058] (1-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time;

[0059] (1-2.3) Use historical data to perform multivariate linear fitting on the mapping relationship between independent variable indicators and dependent variable indicators, obtain the fitting function, and calculate the goodness of fit;

[0060] (1-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data;

[0061] (1-2.5) taking the absolute value of the obtained deviation data, performing multivariate linear fitting on the deviation data after taking the absolute value, and obtaining the deviation fitting function;

[0062] (1-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function as the throughput index correlation model.

[0063] As a preferred embodiment of the present invention, the step (2) further includes a step of generating a state indicator correlation model, which specifically includes the following processing steps:

[0064] (2-2.1) defining the status indicator of the component node as a dependent variable indicator, and defining the throughput indicator of the component node to which the status indicator belongs as an independent variable indicator;

[0065] (2-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time;

[0066] (2-2.3) Use historical data to perform univariate nonlinear fitting on the mapping relationship between the independent variable index and the dependent variable index, obtain the fitting function, and calculate the goodness of fit;

[0067] (2-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data;

[0068] (2-2.5) taking the absolute value of the obtained deviation data, performing a univariate nonlinear fitting on the deviation data after taking the absolute value, and obtaining a deviation fitting function;

[0069] (2-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function as a state index correlation model.

[0070] As a preferred embodiment of the present invention, the step (3) specifically comprises the following steps:

[0071] (3.1) Calculate the correlation anomaly of each indicator;

[0072] (3.2) Calculate the abnormality of each component node;

[0073] (3.3) Arrange all components in descending order of abnormality to obtain a component recommendation list;

[0074] (3.4) For each component, the indicators of the component are arranged in reverse order according to the degree of relevance anomaly to obtain a recommended indicator list for the component.

[0075] As a preferred embodiment of the present invention, the step (3.1) specifically includes the following steps:

[0076] (3.1.1) Use the fitting function in the indicator correlation model to infer the theoretical data of the dependent variable indicator based on the actual data of the independent variable indicator during the abnormal time period;

[0077] (3.1.2) Using the deviation fitting function in the indicator correlation model, the theoretical deviation data of the dependent variable indicator is calculated based on the actual data of the independent variable indicator during the abnormal time period;

[0078] (3.1.3) Subtract the theoretical data of the dependent variable indicator from the actual data of the dependent variable indicator to obtain the absolute deviation data;

[0079] (3.1.4) Divide the absolute deviation data by the theoretical deviation data of the dependent variable indicator and then multiply it by the goodness of fit to obtain the correlation abnormality of the indicator.

[0080] As a preferred embodiment of the present invention, the method further comprises the following steps:

[0081] (4) Feedback is provided to the root cause recommendation table, and the system state model is updated according to the user feedback.

[0082] As a preferred embodiment of the present invention, the step (4) of providing feedback on the root cause recommendation table specifically includes the following processing:

[0083] Confirm the root cause components and abnormal indicators of the software system, expand the indicator list of the root cause components in the root cause recommendation list, and select the abnormal indicators in the list to feed back to the positioning system.

[0084] As a preferred embodiment of the present invention, the topology diagram of step (1) includes component nodes, call relationships between component nodes, a unique throughput indicator of each component, and a status indicator of each component.

[0085] As a preferred embodiment of the present invention, the root cause recommendation list of step (3) is a component recommendation list arranged in descending order of abnormality, wherein each component can be further expanded into an indicator recommendation list arranged in descending order of correlation abnormality.

[0086] The device for implementing a recommended processing of a root cause of a fault for a multi-component software system based on indicator correlation of the present invention comprises:

[0087] a processor configured to execute computer executable instructions;

[0088] The memory stores one or more computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the method for recommending root cause processing of faults for a multi-component software system based on indicator correlation are implemented.

[0089] The processor of the present invention is used to implement fault root cause recommendation processing for a multi-component software system based on indicator correlation, wherein the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the various steps of the above-mentioned method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation are implemented.

[0090] The computer-readable storage medium of the present invention stores a computer program thereon, and the computer program can be executed by a processor to implement the various steps of the above-mentioned method for recommending root cause processing for a multi-component software system based on indicator correlation.

[0091] In a specific embodiment of the present invention, Figure 1 The figure shows a specific implementation of a method for recommending root causes of faults in a multi-component software system based on indicator correlation according to the present invention. The method for recommending root causes of faults in a multi-component software system based on indicator correlation according to the present embodiment includes the following steps:

[0092] S1. Construct a topology diagram of a multi-component software system; the topology diagram includes component nodes, call relationships between component nodes, a unique throughput indicator of each component, and a status indicator of each component;

[0093] S2. Based on the topology map, generate a system state model using historical indicator data; the system state model is composed of indicator correlation models of all component nodes in the topology map; the indicator correlation model includes a throughput indicator correlation model and a state indicator correlation model;

[0094] S3. After an abnormality occurs in the system, based on the system status model, a root cause recommendation list is generated using the indicator data within the abnormal time period; the root cause recommendation list is a component recommendation list arranged in descending order of abnormality, wherein each component can be further expanded into an indicator recommendation list arranged in descending order of correlation abnormality.

[0095] Among them, in the above step S2, the step of generating the throughput indicator correlation model is:

[0096] S21, defining a throughput index of a component node as a dependent variable index, and defining throughput indexes of all other component nodes that call the component node to which the throughput belongs as a set of independent variable indexes;

[0097] S22, obtaining historical data of dependent variable indicators and independent variable indicators in the past period of time;

[0098] S23, using historical data to perform multivariate linear fitting on the mapping relationship between the independent variable indicators and the dependent variable indicators, obtain a fitting function, and calculate the goodness of fit (R square);

[0099] S24, using the obtained fitting function to verify the historical data to obtain deviation data;

[0100] S25, taking absolute values ​​of the obtained deviation data, performing multivariate linear fitting on the deviation data after taking the absolute values, and obtaining a deviation fitting function;

[0101] S26. Save the independent variable index, the dependent variable index, the fitting function, the goodness of fit, and the deviation fitting function as a throughput index correlation model.

[0102] In step S2, the steps of generating the above-mentioned state indicator correlation model are:

[0103] S2a, defining a state indicator of a component node as a dependent variable indicator, and defining a throughput indicator of the component node to which the state indicator belongs as an independent variable indicator;

[0104] S2b, obtain the historical data of dependent variable indicators and independent variable indicators in the past period of time;

[0105] S2c, use historical data to perform univariate nonlinear fitting on the mapping relationship between the independent variable index and the dependent variable index, obtain the fitting function, and calculate the goodness of fit (R square);

[0106] S2d, using the obtained fitting function to verify the historical data to obtain deviation data;

[0107] S2e, taking the absolute value of the obtained deviation data, performing a univariate nonlinear fitting on the deviation data after taking the absolute value, and obtaining a deviation fitting function;

[0108] S2f. Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function as a state index correlation model.

[0109] In the above step S3, the steps of generating the root cause recommendation list are:

[0110] S31, calculating the correlation abnormality of each indicator;

[0111] S32, calculating the abnormality of each component node, the abnormality of the component node is the weighted average of the correlation abnormality of the indicators owned by the component, and the indicator weight used in the weighted average is 1 by default;

[0112] S33, arranging all components in descending order of abnormality to obtain a component recommendation list;

[0113] S34. For each component, the indicators of the component are arranged in reverse order according to the degree of abnormality of correlation to obtain a recommended indicator list of the component.

[0114] Among them, in the above step S31, the calculation step of the correlation abnormality of the indicator is:

[0115] S311, using the fitting function in the indicator correlation model, inferring the theoretical data of the dependent variable indicator according to the actual data of the independent variable indicator in the abnormal time period;

[0116] S312, using the deviation fitting function in the indicator correlation model, inferring the theoretical deviation data of the dependent variable indicator according to the actual data of the independent variable indicator in the abnormal time period;

[0117] S313, taking the difference between the theoretical data of the dependent variable indicator and the actual data of the dependent variable indicator to obtain the absolute deviation data;

[0118] S314. Divide the absolute deviation data by the theoretical deviation data of the dependent variable indicator and then multiply by the goodness of fit to obtain the correlation abnormality of the indicator.

[0119] Wherein, after the above step S3, the following steps are also included: S4, the user provides feedback on the root cause recommendation table, and the system state model is updated according to the user feedback.

[0120] Among them, in the present embodiment, a method for recommending root causes of multi-component software system failures based on indicator correlation, the "feedback on the root cause recommendation list" in the above step S4 is specifically that after the user confirms the root cause components and abnormal indicators of the software system, the user expands the indicator list of the root cause components in the root cause recommendation list, and selects the abnormal indicators in the list to feed back to the positioning system.

[0121] In the above step S4, "updating the system status model according to user feedback" specifically means that the weight of the abnormal indicator fed back by the user to the positioning system in the step S32 will be improved, and the updated weight will be used in the next positioning process.

[0122] The specific implementation scheme of this embodiment can refer to the relevant description in the above embodiment, which will not be repeated here.

[0123] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0124] It should be noted that, in the description of the present invention, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "plurality" refers to at least two.

[0125] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0126] 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, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device. 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.

[0127] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the corresponding program may be stored in a computer-readable storage medium, which, when executed, includes one of the steps of the method embodiment or a combination thereof.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0129] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0130] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0131] The method, device, processor and computer-readable storage medium thereof for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation of the present invention are adopted. The root cause recommendation scheme based on component topology and indicator correlation uses component topology and indicator data as input information, the data volume is relatively small, and the human influence factor is low. The present invention uses the correlation between indicators as the basis for abnormality judgment, which is more reliable than simply judging the abnormality of the indicator's own changes. The present invention has a feedback mechanism, and the indicator weight will be dynamically adjusted according to the user's feedback, gradually improving its own accuracy. Therefore, the present invention has the advantages of high real-time performance and high accuracy.

[0132] In this specification, the present invention has been described with reference to specific embodiments thereof. However, it is apparent that various modifications and variations may be made without departing from the spirit and scope of the present invention. Therefore, the specification and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation, characterized in that: The method comprises the following steps: (1) Construct a topology diagram of a multi-component software system; (2) generating a system status model using historical indicator data according to the topology diagram; (3) In the case of an abnormality, a root cause recommendation list is generated based on the system status model using the indicator data within the abnormal time period; The system state model of step (2) is composed of indicator correlation models of all component nodes in the topology graph, and the indicator correlation model includes a throughput indicator correlation model and a state indicator correlation model; The step (2) further includes a step of generating a throughput indicator correlation model, which specifically includes the following processing steps: (1-2.1) The throughput index of a component node is defined as a dependent variable index, and the throughput indexes of all other component nodes that call the component node to which the throughput belongs are defined as a set of independent variable indexes; (1-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time; (1-2.3) Use historical data to perform multivariate linear fitting on the mapping relationship between independent variable indicators and dependent variable indicators, obtain the fitting function, and calculate the goodness of fit; (1-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data; (1-2.5) taking the absolute value of the obtained deviation data, performing multivariate linear fitting on the deviation data after taking the absolute value, and obtaining the deviation fitting function; (1-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function in (1-2.1) to (1-2.5) as a throughput index correlation model; The step (2) further includes a step of generating a state indicator correlation model, which specifically includes the following processing steps: (2-2.1) defining the status indicator of the component node as a dependent variable indicator, and defining the throughput indicator of the component node to which the status indicator belongs as an independent variable indicator; (2-2.2) Obtain historical data of dependent variable indicators and independent variable indicators over a period of time; (2-2.3) Use historical data to perform univariate nonlinear fitting on the mapping relationship between the independent variable index and the dependent variable index, obtain the fitting function, and calculate the goodness of fit; (2-2.4) Use the obtained fitting function to verify the historical data and obtain the deviation data; (2-2.5) taking the absolute value of the obtained deviation data, performing a univariate nonlinear fitting on the deviation data after taking the absolute value, and obtaining a deviation fitting function; (2-2.6) Save the independent variable index, dependent variable index, fitting function, goodness of fit, and deviation fitting function in (2-2.1) to (2-2.5) as a state index correlation model; The step (3) specifically comprises the following steps: (3.1) Calculate the correlation anomaly of each indicator; (3.2) Calculate the abnormality of each component node; (3.3) Arrange all components in descending order of abnormality to obtain a component recommendation list; (3.4) For each component, the indicators of the component are arranged in reverse order according to the degree of relevance anomaly to obtain a recommended indicator list for the component.

2. The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation according to claim 1, characterized in that: The step (3.1) specifically comprises the following steps: (3.1.1) Use the fitting function in the indicator correlation model to infer the theoretical data of the dependent variable indicator based on the actual data of the independent variable indicator during the abnormal time period; (3.1.2) Using the deviation fitting function in the indicator correlation model, the theoretical deviation data of the dependent variable indicator is calculated based on the actual data of the independent variable indicator during the abnormal time period; (3.1.3) Subtract the theoretical data of the dependent variable indicator from the actual data of the dependent variable indicator to obtain the absolute deviation data; (3.1.4) Divide the absolute deviation data by the theoretical deviation data of the dependent variable indicator and then multiply it by the goodness of fit to obtain the correlation abnormality of the indicator.

3. The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation according to claim 1, characterized in that: The method further comprises the following steps: (4) Feedback is provided to the root cause recommendation table, and the system state model is updated according to the user feedback.

4. The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation according to claim 3, characterized in that: The step (4) of providing feedback on the root cause recommendation table specifically includes the following processing steps: Confirm the root cause components and abnormal indicators of the software system, expand the indicator list of the root cause components in the root cause recommendation list, and select the abnormal indicators in the list to feed back to the positioning system.

5. The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation according to claim 1, characterized in that: The topology diagram of step (1) includes component nodes, call relationships between component nodes, a unique throughput indicator of each component, and a status indicator of each component.

6. The method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation according to claim 1, characterized in that: The root cause recommendation list of step (3) is a component recommendation list arranged in descending order of abnormality, wherein each component can be further expanded into an indicator recommendation list arranged in descending order of relevance abnormality.

7. A device for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation, characterized in that: The device comprises: a processor configured to execute computer executable instructions; A memory storing one or more computer executable instructions, wherein when the computer executable instructions are executed by the processor, the steps of the method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation as described in any one of claims 1 to 6 are implemented.

8. A processor for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation, characterized in that: The processor is configured to execute computer executable instructions. When the computer executable instructions are executed by the processor, the various steps of the method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program can be executed by a processor to implement the various steps of the method for implementing fault root cause recommendation processing for a multi-component software system based on indicator correlation as described in any one of claims 1 to 6.

Images