A method, device, equipment and readable storage medium for determining system health
By calculating the relationship similarity between the target system and the knowledge graph entity in the Internet of Things platform and using generalized linear models, combining functional data to calculate health, and updating the knowledge graph, the insufficient dimensions of health evaluation data and dependence on the health evaluation data in the existing technology are solved. The problem of experience in the development of the experience improves the speed of troubleshooting and evaluation accuracy.
Patent Information
- Application Number
- CN202011164827.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-10-27
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2040-10-27
AI Technical Summary
The health evaluation data dimensions of the existing IoT platforms are not rich enough, they rely too much on the fault diagnosis experience of operation and maintenance personnel, and lack a prior data update mechanism, resulting in the problem detection not being intelligent and timely enough.
By obtaining the current status information of the target system, compute the relationship similarity with the knowledge graph entity, query the fault category and level data, calculate the health level using a generalized linear model in combination with functional data, and update the knowledge graph through the health level results.
It provides multi-dimensional data support, improves the accuracy of fault information retrieval matching and health evaluation, and improves the speed of troubleshooting of IoT cloud platform and the effectiveness of health evaluation.
Smart Images

Figure CN114491052B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of Internet of Things, and in particular to a method, device, equipment and readable storage medium for determining system health. Background Art
[0002] At present, with the continuous upgrading and expansion of services and functions of the IoT platform, the traditional platform resource and performance indicator alarm and monitoring are not smart and timely enough for troubleshooting, and the overall health risk factor of the platform is relatively high. The health evaluation of existing IoT platforms mainly uses limited background fault knowledge to locate, and combines the performance indicator data of fewer dimensions for health evaluation.
[0003] In the existing technical solutions, the platform health evaluation only obtains the platform health by weighted calculation based on the weights of the indicator values of basic data such as storage resources, computing resources, and task resources. The evaluation method is relatively simple and not conducive to troubleshooting faults; the data dimensions of the platform health evaluation are not rich enough, and the data value in the massive logs is not reasonably utilized; the platform health evaluation relies too much on the fault diagnosis experience and judgment of professional operation and maintenance personnel, and lacks a mechanism to update prior data. Summary of the invention
[0004] The embodiments of the present invention provide a system health determination method, apparatus, device and readable storage medium to solve the problems in the prior art that the data dimensions of platform health evaluation are not rich enough, overly rely on the fault diagnosis experience and judgment of operation and maintenance personnel, and lack an updating mechanism for prior data.
[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:
[0006] An embodiment of the present invention provides a method for determining system health, including:
[0007] Get the current status information of the target system;
[0008] Calculate the similarity between the current state information of the target system and the relationship between entities in the knowledge graph;
[0009] According to the relationship similarity, query the corresponding fault category and fault level data in the fault knowledge meta-base;
[0010] The health of the target system is calculated according to the fault category data, the fault level data and the function data of the target system.
[0011] Optionally, the current status information of the target system includes:
[0012] The functional status and product status of entities in the target system.
[0013] Optionally, after obtaining the current status information of the target system, the following steps are included:
[0014] Extracting entity ternary information according to current state information of the target system;
[0015] Establishing a functional knowledge base according to the entity ternary information;
[0016] The entities include services and products.
[0017] Optionally, after obtaining the current status information of the target system, the method further includes:
[0018] Obtain fault source data of the target system;
[0019] Analyze and process the fault source data to obtain fault information;
[0020] Establishing a fault knowledge meta-base according to the fault information;
[0021] The fault source data includes operation and maintenance log data and fault processing data.
[0022] Optionally, the parsing and processing the fault source data to obtain fault information includes:
[0023] Processing the fault source data by at least one of the following: data preprocessing, rule matching, and data organization;
[0024] The fault information includes at least one of labeling information, time series data information and risk fault information.
[0025] Optionally, before calculating the relationship similarity between the current state information of the target system and entities in the knowledge graph, the method further comprises:
[0026] The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph;
[0027] The knowledge fusion of the functional knowledge entity and the fault knowledge entity includes at least one of the following: ontology alignment; entity alignment; and conflict resolution.
[0028] Optionally, the calculating the relationship similarity between the current state information of the target system and entities in the knowledge graph includes:
[0029] Knowledge reasoning is performed on the current state data of the target system, and the relationship similarity between the current state information and the entities in the knowledge graph is calculated based on the knowledge reasoning result.
[0030] Optionally, the current status data of the target system includes:
[0031] Basic monitoring data, application process data, and abnormal alarm data.
[0032] Optionally, calculating the health of the target system includes:
[0033] The health of the target system is calculated using a generalized linear model according to the fault category data, the fault level data and the function data of the target system.
[0034] Optionally, the functional data of the target system includes:
[0035] Performance data of system function modules, user experience data, and connection status data of the target system;
[0036] The connection status data of the target system includes product data, equipment data, service data, data flow and data point.
[0037] Optionally, the method further comprises:
[0038] According to the health of the target system, the functional knowledge entity and the fault knowledge entity are added to the knowledge graph.
[0039] The embodiment of the present invention further provides a system health determination device, comprising:
[0040] The acquisition module is used to obtain the current status information of the target system;
[0041] A first calculation module is used to calculate the similarity between the current state information of the target system and the relationship between entities in the knowledge graph;
[0042] A query module, used to query the corresponding fault category and fault level data in the fault knowledge base according to the relationship similarity;
[0043] The second calculation module is used to calculate the health of the target system according to the fault category data, the fault level data and the function data of the target system.
[0044] Optionally, the device further comprises:
[0045] An updating module is used to add the functional knowledge entity and the fault knowledge entity to the knowledge graph according to the health of the target system.
[0046] An embodiment of the present invention also provides a system health determination device, including: a transceiver and a processor;
[0047] The transceiver is used to obtain current status information of the target system;
[0048] The processor is used to calculate the similarity between the current state information of the target system and the entities in the knowledge graph;
[0049] According to the relationship similarity, query the corresponding fault category and fault level data in the fault knowledge meta-base;
[0050] The health of the target system is calculated according to the fault category data, the fault level data and the function data of the target system.
[0051] An embodiment of the present invention further provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the system health determination method described above are implemented.
[0052] The beneficial effects of the present invention are:
[0053] The system health determination method of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. The troubleshooting and retrieval speed of the Internet of Things cloud platform faults is improved, and the effectiveness and accuracy of health evaluation are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 A schematic diagram showing the steps of a method for determining system health according to an embodiment of the present invention;
[0055] Figure 2 A schematic diagram showing a method for determining system health according to an embodiment of the present invention;
[0056] Figure 3 A schematic diagram showing a module of a system health determination device according to an embodiment of the present invention;
[0057] Figure 4 A schematic diagram showing the structure of a device for determining system health according to an embodiment of the present invention. DETAILED DESCRIPTION
[0058] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0059] The present invention aims to solve the problems in the prior art that the data dimensions of platform health evaluation are not rich enough, overly rely on the fault diagnosis experience and judgment of operation and maintenance personnel, and lack an updating mechanism for prior data. A system health determination method, device, equipment and readable storage medium are provided.
[0060] like Figure 1 As shown, an embodiment of the present invention provides a method for determining system health, including:
[0061] Step 101, obtaining current status information of the target system;
[0062] Step 102, calculating the relationship similarity between the current state information of the target system and the entities in the knowledge graph;
[0063] Step 103, according to the relationship similarity, query the corresponding fault category and fault level data in the fault knowledge base;
[0064] Step 104: Calculate the health of the target system according to the fault category data, the fault level data and the function data of the target system.
[0065] The system health determination method of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. The troubleshooting and retrieval speed of the Internet of Things cloud platform faults is improved, and the effectiveness and accuracy of health evaluation are improved.
[0066] Optionally, the current status information of the target system includes:
[0067] The functional status and product status of entities in the target system.
[0068] Here, the current status information of the target system includes functional status and product status information of different industries and scenarios, services and products in the target system, and corresponding entity triplet information is extracted through the current status information of the target system.
[0069] Optionally, after obtaining the current status information of the target system, the following steps are included:
[0070] Extracting entity ternary information according to current state information of the target system;
[0071] Establishing a functional knowledge base according to the entity ternary information;
[0072] The entities include services and products.
[0073] Here, the functional knowledge base includes a functional knowledge base between services, a functional knowledge base between products, and a functional knowledge base between services and products.
[0074] Optionally, after obtaining the current status information of the target system, the method further includes:
[0075] Obtain fault source data of the target system;
[0076] Analyze and process the fault source data to obtain fault information;
[0077] Establishing a fault knowledge meta-base according to the fault information;
[0078] The fault source data includes operation and maintenance log data and fault processing data.
[0079] Here, the fault knowledge base includes a fault knowledge base between services, a fault knowledge base between products, and a fault knowledge base between services and products. The knowledge base contains multiple subgraphs, each of which includes entities, relationships, and attributes.
[0080] Optionally, the parsing and processing the fault source data to obtain fault information includes:
[0081] Processing the fault source data by at least one of the following: data preprocessing, rule matching, and data organization;
[0082] The fault information includes at least one of labeling information, time series data information and risk fault information.
[0083] In one embodiment of the present invention, data preprocessing may include steps such as data cleaning, data integration, data transformation, and data reduction. Data cleaning mainly removes noise and irrelevant data; data integration combines data from multiple data sources and stores them in a consistent data storage; data transformation converts the original data into a form suitable for data mining; the main methods of data reduction include data cube aggregation, dimensionality reduction, data compression, numerical reduction, etc.
[0084] Rule matching is to classify and identify key fault data according to fault analysis template rules to form classified fault information of the physical network cloud platform.
[0085] Data organization is based on the data preprocessed and matched with rules by the IoT cloud platform, forming a unified data model in accordance with business requirements and system specifications, providing basic data preparation for building a knowledge graph for health evaluation of the IoT cloud platform.
[0086] Through the analysis and processing of fault source data, the fault source data is converted into the format required for establishing the fault knowledge base, providing rich and convenient data support for fault information retrieval and matching and health evaluation. The knowledge base contains multiple subgraphs, each of which includes entities, relationships and attributes.
[0087] Optionally, before calculating the relationship similarity between the current state information of the target system and entities in the knowledge graph, the method further comprises:
[0088] The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph;
[0089] The knowledge fusion of the functional knowledge entity and the fault knowledge entity includes at least one of the following: ontology alignment; entity alignment; and conflict resolution.
[0090] In one embodiment of the present invention, ontology alignment is used to solve the heterogeneity between knowledge systems. The knowledge framework mainly includes types, attributes, relationships and constraints. The ontology alignment includes element-level matching and structure-level matching.
[0091] Entity alignment is used to determine whether two entities in the same or different knowledge bases represent the same physical object. Knowledge fusion is achieved by aligning and merging the same entities, including pairwise entity alignment, collaborative alignment, representation learning alignment, and other methods.
[0092] Conflict handling resolves conflicts between different instances based on ontology alignment and entity alignment, including conflict ignoring, conflict avoiding, and conflict resolving.
[0093] Optionally, the calculating the relationship similarity between the current state information of the target system and entities in the knowledge graph includes:
[0094] Knowledge reasoning is performed on the current state data of the target system, and the relationship similarity between the current state information and the entities in the knowledge graph is calculated based on the knowledge reasoning result.
[0095] By performing knowledge reasoning on the current state data of the target system and based on the reasoning control strategy, the formalized knowledge is used to calculate the similarity between the current state data of the target system and the entities in the knowledge graph.
[0096] Optionally, the current status data of the target system includes:
[0097] Basic monitoring data, application process data, and abnormal alarm data.
[0098] Through the existing monitoring data of R&D and operation, as well as the relationship similarity with the entities in the knowledge graph, the fault status of the target system is judged, and the fault category and fault level data corresponding to the fault status are retrieved from the fault knowledge meta-repository.
[0099] Optionally, calculating the health of the target system includes:
[0100] The health of the target system is calculated using a generalized linear model according to the fault category data, the fault level data and the function data of the target system.
[0101] Combining multidimensional data such as fault category data, fault level data and functional data of the target system, the health of the target system is calculated through a generalized linear model. Here, the generalized linear model calculates the significance of fixed factors and random factors, and through reasonable control of the random effects of interactions, effectively explores the network relationship between variables to obtain the evaluation results of multivariate problems.
[0102] Optionally, the functional data of the target system includes:
[0103] Performance data of system function modules, user experience data, and connection status data of the target system;
[0104] The connection status data of the target system includes product data, equipment data, service data, data flow and data point.
[0105] The evaluation of the health of the target system is not only based on the index values of basic data such as storage resources, computing resources, and task resources, but also analyzes and evaluates multi-dimensional data such as resource status indicators and performance indicators.
[0106] Optionally, the method further comprises:
[0107] According to the health of the target system, the functional knowledge entity and the fault knowledge entity are added to the knowledge graph.
[0108] Based on the output results of health, the interactive custom knowledge meta-base method is used to supplement the health knowledge graph of the target system with a new meta-base, and the knowledge graph is updated by combining the extraction of entity information feature engineering and machine learning prediction methods. This is more conducive to the troubleshooting and retrieval of system faults and improves the effectiveness and accuracy of health evaluation.
[0109] The system health determination method of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. The troubleshooting and retrieval speed of the Internet of Things cloud platform faults is improved, and the effectiveness and accuracy of health evaluation are improved.
[0110] like Figure 2 As shown, the system health determination method of the embodiment of the present invention determines the health of the detected system by establishing a functional knowledge base and a fault knowledge base, combined with a health evaluation knowledge graph obtained through knowledge fusion and knowledge calculation; and timely supplements the knowledge graph with the health determination result to update the knowledge graph. The troubleshooting and retrieval speed of the IoT cloud platform is improved, and the effectiveness and accuracy of health evaluation are improved.
[0111] like Figure 3 As shown, the embodiment of the present invention further provides a system health determination device 300, comprising:
[0112] The acquisition module 301 is used to acquire the current status information of the target system;
[0113] The first calculation module 302 is used to calculate the similarity between the current state information of the target system and the relationship between entities in the knowledge graph;
[0114] A query module 303 is used to query the corresponding fault category and fault level data in the fault knowledge base according to the relationship similarity;
[0115] The second calculation module 304 is used to calculate the health of the target system according to the fault category data, the fault level data and the function data of the target system.
[0116] The system health determination device of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. It improves the troubleshooting and retrieval speed of the Internet of Things cloud platform faults and improves the effectiveness and accuracy of health evaluation.
[0117] Optionally, the device further comprises:
[0118] An updating module is used to add the functional knowledge entity and the fault knowledge entity to the knowledge graph according to the health of the target system.
[0119] Based on the output results of health, the interactive custom knowledge meta-base method is used to supplement the health knowledge graph of the target system with a new meta-base, and the knowledge graph is updated by combining the extraction of entity information feature engineering and machine learning prediction methods. This is more conducive to the troubleshooting and retrieval of system faults and improves the effectiveness and accuracy of health evaluation.
[0120] The system health determination device of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. It improves the troubleshooting and retrieval speed of the Internet of Things cloud platform faults and improves the effectiveness and accuracy of health evaluation.
[0121] like Figure 4 As shown, the embodiment of the present invention further provides a system health determination device 400, including: a transceiver 401 and a processor 402;
[0122] The transceiver 401 is used to obtain the current status information of the target system;
[0123] The processor 402 is used to calculate the similarity between the current state information of the target system and the entities in the knowledge graph;
[0124] According to the relationship similarity, query the corresponding fault category and fault level data in the fault knowledge meta-base;
[0125] The health of the target system is calculated according to the fault category data, the fault level data and the function data of the target system.
[0126] Optionally, the processor is further configured to extract entity ternary information according to current state information of the target system;
[0127] Establishing a functional knowledge base according to the entity ternary information;
[0128] The entities include services and products.
[0129] Optionally, the transceiver is also used to obtain fault source data of the target system.
[0130] Optionally, the processor is further used to parse the fault source data to obtain fault information;
[0131] Establishing a fault knowledge meta-base according to the fault information;
[0132] Wherein, the fault source data includes operation and maintenance log data and fault processing data;
[0133] Processing the fault source data by at least one of the following: data preprocessing, rule matching, and data organization;
[0134] The fault information includes: at least one of labeling information, time series data information and risk fault information;
[0135] The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph;
[0136] Wherein, the knowledge fusion of the functional knowledge entity and the fault knowledge entity includes at least one of the following: ontology alignment; entity alignment; and conflict handling;
[0137] Performing knowledge reasoning on the current state data of the target system, and calculating the relationship similarity between the current state information and entities in the knowledge graph according to the knowledge reasoning result;
[0138] The health of the target system is calculated using a generalized linear model according to the fault category data, the fault level data and the function data of the target system.
[0139] The system health determination device of the embodiment of the present invention, in the evaluation of the health of the Internet of Things cloud platform, constructs a knowledge graph of health evaluation based on the multidimensional data of the state indicators and performance indicators of the evaluated system, providing rich and convenient data support for fault information retrieval and matching and health evaluation; analyzes and calculates the health value through a generalized linear model, effectively reveals the health relationship implied in the data, and solves the highly nonlinear and non-monotonic relationship between the response variable and the prediction factor; updates the knowledge graph through the health determination result. It improves the troubleshooting and retrieval speed of the Internet of Things cloud platform faults and improves the effectiveness and accuracy of health evaluation.
[0140] An embodiment of the present invention further provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps in the system health determination method described above are implemented.
[0141] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary personnel in this technical field, several improvements and modifications can be made without departing from the principles described in the present invention. These improvements and modifications are also within the scope of protection of the present invention.
Claims
1. A method for determining system health, characterized in that: include: Get the current status information of the target system; Extracting entity ternary information according to current state information of the target system; Establishing a functional knowledge base according to the entity ternary information; Obtain fault source data of the target system; Analyze and process the fault source data to obtain fault information; Establishing a fault knowledge meta-base according to the fault information; The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph; Calculate the relationship similarity between the current state information of the target system and the entities in the knowledge graph; According to the relationship similarity, querying the corresponding fault category and fault level data in the fault knowledge meta-base; Calculating the health of the target system according to the fault category, the fault level data and the functional data of the target system; The functional knowledge base includes: a functional knowledge base between services, a functional knowledge base between products, and a functional knowledge base between services and products; The fault knowledge base includes a fault knowledge base between services, a fault knowledge base between products, and a fault knowledge base between services and products; The current status information of the target system includes: functional status and product status of entities in the target system; Said entities include services and products; The fault source data includes operation and maintenance log data and fault processing data.
2. The method for determining system health according to claim 1, characterized in that: The step of parsing the fault source data to obtain fault information includes: Processing the fault source data by at least one of the following: data preprocessing, rule matching, and data organization; The fault information includes at least one of labeling information, time series data information and risk fault information.
3. The method for determining system health according to claim 1, characterized in that: The knowledge fusion of the functional knowledge entity and the fault knowledge entity includes at least one of the following: ontology alignment; Entity alignment; and, conflict management.
4. The method for determining system health according to claim 1 or 3, characterized in that: The calculating the relationship similarity between the current state information of the target system and entities in the knowledge graph includes: Knowledge reasoning is performed on the current state data of the target system, and the relationship similarity between the current state information and the entities in the knowledge graph is calculated based on the knowledge reasoning result.
5. The method for determining system health according to claim 4, characterized in that: The current status data of the target system includes: Basic monitoring data, application process data, and abnormal alarm data.
6. The method for determining system health according to claim 1, characterized in that: The calculating the health of the target system includes: The health of the target system is calculated using a generalized linear model according to the fault category, the fault level data and the functional data of the target system.
7. The method for determining system health according to claim 1 or 6, characterized in that: Functional data of the target system, including: Performance data of system function modules, user experience data, and connection status data of the target system; The connection status data of the target system includes product data, equipment data, service data, data flow and data points.
8. The method for determining system health according to claim 1, characterized in that: The method further comprises: According to the health of the target system, the functional knowledge entity and the fault knowledge entity are added to the knowledge graph.
9. A system health determination device, characterized in that: include: The acquisition module is used to obtain the current status information of the target system; Extracting entity ternary information according to current state information of the target system; Establishing a functional knowledge base according to the entity ternary information; Obtain fault source data of the target system; Analyze and process the fault source data to obtain fault information; Establishing a fault knowledge meta-base according to the fault information; The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph; A first calculation module is used to calculate the similarity between the current state information of the target system and the entities in the knowledge graph; A query module, used for querying corresponding fault category and fault level data in the fault knowledge meta-base according to the relationship similarity; A second calculation module, used for calculating the health of the target system according to the fault category, the fault level data and the function data of the target system; The functional knowledge base includes: a functional knowledge base between services, a functional knowledge base between products, and a functional knowledge base between services and products; The fault knowledge base includes a fault knowledge base between services, a fault knowledge base between products, and a fault knowledge base between services and products; The current status information of the target system includes: functional status and product status of entities in the target system; Said entities include services and products; The fault source data includes operation and maintenance log data and fault processing data.
10. The system health determination device according to claim 9, characterized in that: The device also includes: An updating module is used to add functional knowledge entities and fault knowledge entities to the knowledge graph according to the health of the target system.
11. A system health determination device, characterized in that: include: transceivers and processors; The transceiver is used to obtain current status information of the target system; Extracting entity ternary information according to current state information of the target system; Establishing a functional knowledge base according to the entity ternary information; The transceiver is also used to obtain fault source data of the target system; The processor is used to parse and process the fault source data to obtain fault information; Establishing a fault knowledge meta-base according to the fault information; The functional knowledge entity and the fault knowledge entity are integrated to obtain a knowledge graph; The processor is used to calculate the similarity between the current state information of the target system and the entities in the knowledge graph; According to the relationship similarity, query the corresponding fault category and fault level data in the fault knowledge meta-base; Calculating the health of the target system according to the fault category, the fault level data and the functional data of the target system; The functional knowledge base includes: a functional knowledge base between services, a functional knowledge base between products, and a functional knowledge base between services and products; The fault knowledge base includes a fault knowledge base between services, a fault knowledge base between products, and a fault knowledge base between services and products; The current status information of the target system includes: functional status and product status of entities in the target system; Said entities include services and products; The fault source data includes operation and maintenance log data and fault processing data.
12. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by the processor, the steps in the system health determination method as described in any one of claims 1-8 are implemented.
