ICT system maintenance strategy generation method and device, equipment and medium

Abstract

The invention discloses an ICT system maintenance strategy generation method and device, equipment and a medium, and the method comprises the steps: recognizing service information, equipment information and fault type parameters associated with fault risk operation data under the condition that the existence of the fault risk operation data in an ICT system is predicted, determining fault complexity corresponding to the fault risk operation data; under the condition that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, determining a plurality of candidate maintenance strategies; and determining a plurality of target maintenance strategies corresponding to the fault reasons according to the strategy conditions in the candidate maintenance strategies and the service constraint conditions in the service information, and generating an ICT system maintenance strategy, so that the timeliness and efficiency of system maintenance can be improved, and the success rate of ICT system maintenance is ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of equipment inspection, and particularly relates to an ICT system maintenance strategy generation method and device, equipment and a medium. BACKGROUND

[0002] An ICT (Information and Communication Technology) system is a whole composed of hardware, software, network and data, integrates information processing and communication transmission functions, and is widely used in various scenes such as enterprises and industries. With the deepening of digital transformation, the scale of data center servers gradually increases, the number of sensors and edge devices in industrial scenes increases exponentially, and the demand for cloud and local system cooperation surges. The stable operation of the ICT system directly determines the continuity of enterprise business and service quality. Therefore, the maintenance of the ICT system has become very important.

[0003] In the prior art, the maintenance of the ICT system is mainly to formulate a fixed maintenance inspection cycle according to the static information such as the equipment model and the running time of the ICT system, and when the maintenance inspection cycle is reached, it is determined whether the system operation has a fault by comparing the running parameters in the system operation process with the preset threshold. If there is a fault, the fault parameter is reported to the system maintenance terminal. However, the prior art can only determine the fault based on the fixed threshold when the system has a running fault, and can only report the fault parameter to the system maintenance terminal for maintenance. The use of the prior art for ICT system maintenance has the problems of untimely system maintenance and low maintenance efficiency. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide an ICT system maintenance strategy generation method, device, equipment and medium, which solves the problems of untimely system maintenance and low maintenance efficiency in the prior art. By determining and predicting the fault risk running data corresponding to the fault complexity, in the case that the fault complexity is greater than or equal to the preset multi-strategy maintenance fault complexity threshold, a plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy conditions in each candidate maintenance strategy and the business constraint conditions in the business information, and the plurality of target maintenance strategies are integrated to generate an ICT system maintenance strategy. The purpose of generating the ICT system maintenance strategy in advance before the system failure can be achieved, the timeliness and efficiency of system maintenance are improved, and the generation of the ICT system maintenance strategy can be combined with the business constraint conditions of the running data to ensure the success rate of the ICT system maintenance.

[0005] In a first aspect, the embodiments of the present application provide an ICT system maintenance strategy generation method, which comprises: In the case that the fault risk running data of the ICT system is predicted, the business information, the equipment information and the fault type parameter associated with the fault risk running data are identified, and the fault complexity corresponding to the fault risk running data is determined according to the business information, the equipment information and the fault type parameter; In the case that the fault complexity is greater than or equal to the preset multi-policy maintenance fault complexity threshold, the fault cause corresponding to the fault risk running data in the preset ICT system knowledge base is determined, and a plurality of candidate maintenance strategies corresponding to the fault cause are determined; The plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information, and the ICT system maintenance strategy is generated by integrating the plurality of target maintenance strategies.

[0006] Further, the plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information, including: The consistency of the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information is checked; In the case that the strategy condition and the business constraint condition are consistent, the candidate maintenance strategy corresponding to the strategy condition is filtered out to obtain the plurality of target maintenance strategies corresponding to the fault cause.

[0007] Further, the business constraint condition includes a maintenance time constraint, an available resource constraint and a maintenance risk dimension constraint; The consistency of the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information is checked, including: The execution time constraint, the strategy resource dependency constraint and the strategy risk dimension of each candidate maintenance strategy are determined according to the strategy condition in each candidate maintenance strategy, and the strategy demand set corresponding to each candidate maintenance strategy is generated according to the execution time constraint, the strategy resource dependency constraint and the strategy risk dimension; The maintenance constraint set corresponding to the fault risk running data is generated according to the maintenance time constraint, the available resource constraint and the maintenance risk dimension constraint, and the intersection of the strategy demand set and the maintenance constraint set is calculated; It is judged whether the intersection is empty, and in the case that the intersection is not empty, it is determined that the strategy condition and the business constraint condition are consistent.

[0008] Further, the business information includes business basic information, business constraint conditions and business running conditions; the equipment information includes fault equipment information and dependent equipment information; The fault complexity corresponding to the fault risk running data is determined according to the business information, the equipment information and the fault type parameter, including: determine a proportion of the business constraint condition in the business operation condition according to a first number of the business constraint condition and a second number of the business operation condition, to obtain a business constraint strength corresponding to the fault risk operation data; determine a business operation logic according to the business basic information, determine a device dependency degree according to the business operation logic, a first device function in the fault device information and a second device function in the dependent device information, to obtain a device association degree corresponding to the fault risk operation data; determine a plurality of basic fault type parameters corresponding to the business operation logic, calculate a coincidence degree of each basic fault type parameter and the fault type parameter, to obtain a fault impact range corresponding to the fault risk operation data, and integrate the business constraint strength, the device association degree and the fault impact range to obtain a fault complexity corresponding to the fault risk operation data.

[0009] Further, the number of the dependent device information is a plurality; determine a device dependency degree according to the business operation logic, a first device function in the fault device information and a second device function in the dependent device information, comprising: match the first device function in the fault device information and the second device function in each dependent device information with a business stage function corresponding to the business operation logic respectively, to determine a first business stage corresponding to the first device function, and a second business stage corresponding to each second device function; determine a business processing time sequence relationship between the first business stage and each second business stage, determine an impact function number of the first device function according to the time sequence relationship, and determine an associated function number of the first device function according to the number of the plurality of second device functions; determine the device dependency degree based on a preset impact weight, a preset association weight, the impact function number and the associated function number.

[0010] Further, the construction process of the preset ICT system knowledge base comprises: obtain a plurality of source ICT system knowledge data, perform semantic recognition on entity names of a plurality of entities in the plurality of source ICT system knowledge data, perform naming uniform processing on entity names with the same semantics according to a semantic recognition result, and construct a basic ICT system knowledge base based on a preset knowledge base construction rule and a naming uniform processing result, the plurality of entities comprising a plurality of fault entities and a plurality of device entities; obtain a plurality of historical operation data of the ICT system, determine a plurality of implicit associated fault entities under the same fault occurrence time according to fault occurrence times of each fault entity in the plurality of historical operation data, and determine an implicit associated device entity corresponding to each implicit associated fault entity; update a basic association relationship in the basic ICT system knowledge base based on the implicit associated fault entity and the implicit associated device entity, to obtain a final ICT system knowledge base.

[0011] Further, the process of predicting whether there is a fault risk running data in the ICT system comprises: obtaining current running data and a plurality of historical running data of the ICT system, determining a running data change rule of the ICT system based on the plurality of historical running data; determining a running timestamp of the current running data, and predicting a theoretical running data corresponding to the running timestamp according to the running data change rule; calculating a data deviation between the theoretical running data and the current running data, and determining that there is a fault risk running data in the ICT system in a case where the data deviation is greater than a preset data deviation threshold.

[0012] In a second aspect, an embodiment of the present application provides an ICT system maintenance strategy generation device, the device comprising: a fault complexity determination module configured to, in a case where it is predicted that there is a fault risk running data in the ICT system, identify business information, device information and fault type parameters associated with the fault risk running data, and determine a fault complexity corresponding to the fault risk running data according to the business information, the device information and the fault type parameters; a candidate strategy determination module configured to, in a case where the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, determine a fault cause corresponding to the fault risk running data in a preset ICT system knowledge base, and a plurality of candidate maintenance strategies corresponding to the fault cause; a maintenance strategy generation module configured to determine a plurality of target maintenance strategies corresponding to the fault cause according to a strategy condition in each candidate maintenance strategy and a business constraint condition in the business information, and integrate the plurality of target maintenance strategies to generate an ICT system maintenance strategy.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, which comprises a processor, a memory, and a program or instructions stored in the memory and executable on the processor, and the program or instructions are executed by the processor to implement the steps of the method according to the first aspect.

[0014] In a fourth aspect, an embodiment of the present application provides a readable storage medium, which stores a program or instructions, and the program or instructions are executed by a processor to implement the steps of the method according to the first aspect.

[0015] In a fifth aspect, an embodiment of the present application further provides a computer program product, which comprises a computer program stored in a computer readable storage medium, and at least one processor of a device reads and executes the computer program from the computer readable storage medium, so that the device executes the method according to the first aspect.

[0016] In the embodiment of the present application, in the case that the fault risk operation data in the ICT system is predicted, the business information, the equipment information and the fault type parameter associated with the fault risk operation data are identified, and the fault complexity corresponding to the fault risk operation data is determined according to the business information, the equipment information and the fault type parameter; in the case that the fault complexity is greater than or equal to the preset multi-strategy maintenance fault complexity threshold, the fault cause corresponding to the fault risk operation data in the preset ICT system knowledge base is determined, and a plurality of candidate maintenance strategies corresponding to the fault cause are determined; the plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information, and the ICT system maintenance strategy is generated by integrating the plurality of target maintenance strategies. Through the above-mentioned ICT system maintenance strategy generation method, the problems of system maintenance not in time and low maintenance efficiency in the prior art are solved. By determining the fault complexity corresponding to the predicted fault risk operation data, in the case that the fault complexity is greater than or equal to the preset multi-strategy maintenance fault complexity threshold, the plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information, and the ICT system maintenance strategy is generated by integrating the plurality of target maintenance strategies, the purpose of generating the ICT system maintenance strategy in advance before the system fault occurs can be achieved, the timeliness and efficiency of system maintenance are improved, and the generation of the ICT system maintenance strategy can be combined with the business constraint condition of the operation data, so that the success rate of the ICT system maintenance is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flowchart of an ICT system maintenance strategy generation method provided by the embodiment of the present application; Figure 2 is a flowchart of determining a target maintenance strategy provided by the embodiment of the present application; Figure 3 is a flowchart of determining a fault complexity provided by the embodiment of the present application; Figure 4 is a structural diagram of an ICT system maintenance strategy generation system provided by the embodiment of the present application; Figure 5 is a structural block diagram of an ICT system maintenance strategy generation apparatus provided by the embodiment of the present application; Figure 6 is a structural block diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purposes, technical solutions and advantages of the present application clearer, the following will describe the optional detailed description of the embodiments of the present application in conjunction with the drawings. It can be understood that the specific embodiments described here are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, but not all the contents. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The processes can be terminated when the operations are completed, but can also have additional steps not included in the drawings. The processes can correspond to methods, functions, procedures, subroutines, etc.

[0019] The technical solutions in the embodiments of the present application will be described clearly in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art belong to the scope of protection of the present application.

[0020] The terms "first", "second" and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be exchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second" and the like are usually a kind, and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / ", generally means that the front and rear associated objects are in an "or" relationship.

[0021] Firstly, the use scenario of the scheme can be a scenario of performing inspection and operation and maintenance on an ICT system, especially a scenario of performing inspection and maintenance on running data and device transmission data in the ICT system to ensure stable operation of the ICT system. By determining the fault complexity corresponding to the predicted fault risk running data, in the case that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, a plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy conditions in each candidate maintenance strategy and the business constraint conditions in the business information, and the plurality of target maintenance strategies are integrated to generate an ICT system maintenance strategy, which can achieve the purpose of generating the ICT system maintenance strategy in advance before the system failure occurs, improve the timeliness and efficiency of system maintenance, and enable the ICT system maintenance strategy to be generated in combination with the business constraint conditions of the running data, thereby ensuring the success rate of the ICT system maintenance. Based on the above use scenario, it can be understood that the execution subject of each step in the scheme can be a computer device, which refers to any electronic device with data calculation, processing and storage capabilities, such as a mobile phone, a PC (Personal Computer), a tablet computer and the like, or a server and the like, and the embodiments of the present application are not limited thereto.

[0022] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and specific examples and application scenarios.

[0023] Figure 1 is a flowchart of an ICT system maintenance strategy generation method provided by the embodiments of the present application. As shown in Figure 1 , the specific steps include the following steps: S101, in the case that the ICT system is predicted to have fault risk running data, identifying the business information, device information and fault type parameters associated with the fault risk running data, and determining the fault complexity corresponding to the fault risk running data according to the business information, device information and fault type parameters.

[0024] The ICT (Information and Communications Technology) system is a comprehensive infrastructure system integrating computer hardware, network communication equipment, software application, data storage facilities, etc., covering data center servers, network equipment, sensors, operation and maintenance platforms, etc., and is the core carrier of digital business operation. The fault risk operation data can be operation data deviating from the normal operation state and possibly existing fault risks in the operation process of the ICT system. For example, CPU usage rate exceeding 95%, network traffic sudden drop, Redis cache hit rate sudden drop, and device vibration value exceeding the standard, etc. The business information can be information for describing the business characteristics corresponding to the fault risk operation data. The business information can include business attributes and business constraint information. The device information can be information identifying the device outputting or transmitting the fault risk operation data. The fault type parameter can be a parameter describing the fault characteristics, attributes, and categories corresponding to the fault risk operation data. The fault type parameter can be used to distinguish the nature and severity of the fault. The fault complexity can be data for quantitatively measuring the processing difficulty of the ICT system fault. For simple faults, the fault solution strategy can be directly found based on the preset ICT system knowledge base, which is used for ICT system maintenance in the case of faults corresponding to the fault risk operation data. For complex faults, the fault solution strategy needs to be generated based on the analysis of the fault cause and the constraint condition for system maintenance.

[0025] In one embodiment, the real-time operation data of the ICT system can be continuously monitored and abnormally detected by a preset time series prediction model. The preset time series prediction model can be a LinSeer TS model, and the real-time operation data can include CPU usage rate, network traffic, device temperature, and Redis cache hit rate, etc. In the case of detecting that there is fault risk operation data in the ICT system, the business information and device information associated with the fault risk operation data can be determined according to the business identifier and transmission path carried in the fault risk operation data, and the fault type parameter associated with the fault risk operation data can be determined according to the corresponding relationship between the historical fault parameters of the ICT system and the business stage, the business stage corresponding to the business information, and the operation target corresponding to the fault risk operation data. The identified business information, device information, and fault type parameter are respectively extracted and quantified. For example, the business priority corresponding to the business information is quantified according to the importance of the business stage corresponding to the business information, the device impact degree corresponding to the device information is quantified according to the number of associated devices in the device information, and the fault severity corresponding to the fault type parameter is quantified according to the fault category corresponding to the fault type parameter. The normalized sum of the weighted and normalized multi-feature quantification results is obtained according to the preset weight, and the fault complexity corresponding to the fault risk operation data is obtained.

[0026] In one embodiment, the process of predicting whether there is fault risk running data in the ICT system comprises: obtaining current running data and a plurality of historical running data of the ICT system, determining running data variation law of the ICT system based on the plurality of historical running data; determining a running timestamp of the current running data, and predicting theoretical running data corresponding to the running timestamp according to the running data variation law; calculating data deviation of the theoretical running data and the current running data, and determining that there is fault risk running data in the ICT system in the case that the data deviation is greater than a preset data deviation threshold.

[0027] The current running data can be data used to describe the running state of devices, hardware and software in the ICT system at the current time. The historical running data can be running data in which no running fault occurs within a preset time length in the historical running process of the ICT system. The running timestamp can be the time at which each item of data in the current running data is collected. The theoretical running data can be standard running data of the system at the time corresponding to the running timestamp in the case that no fault occurs in the ICT system.

[0028] In one embodiment, the current running data and the plurality of historical running data of the ICT system can be obtained in real time through a plurality of data interfaces in the ICT system, and the variation law of each running data in the ICT system over time can be determined based on the plurality of historical running data. The running timestamp corresponding to the current running data is read, the theoretical running data of each running item in the ICT system at the time corresponding to the running timestamp is predicted according to the running data variation law, and the data deviation of the theoretical running data and the current running data of the same running item is calculated. Whether the data deviation is greater than a preset data deviation threshold is compared, and in the case that the data deviation is greater than the preset data deviation threshold, it is determined that there is fault risk running data in the ICT system.

[0029] The present scheme determines the running data variation law of the ICT system through a plurality of historical running data, predicts the theoretical running data according to the running data variation law, calculates the data deviation of the theoretical running data and the current running data, and determines that there is fault risk running data in the ICT system, which can improve the accuracy of the judgment result of the fault risk running data.

[0030] S102, in the case that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, the fault cause corresponding to the fault risk running data in the preset ICT system knowledge base is determined, and a plurality of candidate maintenance strategies corresponding to the fault cause are determined.

[0031] The preset multi-policy maintenance fault complexity threshold can be a preset minimum value of the ICT system fault complexity in the case of multi-policy system maintenance. The preset ICT system knowledge base can be a collection of pre-constructed structured, searchable, and system operation knowledge including ICT system operation data range, corresponding fault causes, and corresponding maintenance strategies. The fault cause can be a possible cause of the ICT system failure risk operation data. The candidate maintenance strategy can be a plurality of system maintenance schemes associated with the fault cause retrieved from the preset ICT system knowledge base.

[0032] In one embodiment, the fault complexity can be compared with the size of the preset multi-policy maintenance fault complexity threshold, and in the case that the fault complexity is greater than or equal to the preset multi-policy maintenance fault complexity threshold, all possible fault causes corresponding to the fault risk operation data and one or more candidate maintenance strategies corresponding to each fault cause are retrieved based on the preset ICT system knowledge base.

[0033] In one embodiment, the construction process of the preset ICT system knowledge base includes: acquiring multi-source ICT system knowledge data, performing semantic recognition on entity names of a plurality of entities in the multi-source ICT system knowledge data, performing naming uniform processing on entity names with the same semantics according to the semantic recognition result, and constructing a basic ICT system knowledge base based on a preset knowledge base construction rule and the naming uniform processing result, the plurality of entities including a plurality of fault entities and a plurality of device entities; acquiring a plurality of historical operation data of the ICT system, determining a plurality of implicitly associated fault entities under the same fault occurrence time according to the fault occurrence time corresponding to each fault entity in the plurality of historical operation data, and determining an implicitly associated device entity corresponding to each implicitly associated fault entity; updating the basic association relationship in the basic ICT system knowledge base based on the implicitly associated fault entity and the implicitly associated device entity to obtain a final ICT system knowledge base.

[0034] The multi-source ICT system knowledge data can be a set of system operation knowledge information of different sources and different formats related to the ICT system. The sources include manuals provided by equipment manufacturers, historical fault handling reports, operation expert experience records, operation instruction books, and real-time sensor collected fault associated data, etc. The basic ICT system knowledge base can be a knowledge base including explicit associations and basic coverage between entities in the ICT system. The fault entity can be a structured information unit capable of representing the fault type or fault phenomenon of the ICT system. The device entity can be a structured information unit capable of representing the hardware / software device of the ICT system. For example: servers, switches, sensors, databases, application programs, and operating systems, etc. The implicitly associated fault entity can be a fault entity that appears frequently in the same time window as a certain fault entity in historical operation data, which may have a potential cause or linkage relationship. The implicitly associated device entity can be a device entity corresponding to the implicitly associated fault entity, which has a physical or logical association with the main fault device.

[0035] In one embodiment, a plurality of sources of heterogeneous knowledge data related to the ICT system can be collected by calling an API interface, and redundant information can be removed and data format can be corrected by a data cleaning algorithm to ensure data readability and consistency. The entities in the preprocessed knowledge data are semantically analyzed and extracted by a semantic recognition technology to extract fault entities and device entities, a semantic similarity calculation model is called to match entities with the same semantics but different names, and these entities are uniformly named as standard names. According to a preset rule (such as “fault entity-impact-device entity”, “device entity-occurrence-fault entity”, etc.), the association relationship between entities is defined, the entities are taken as nodes (V) and the relationship is taken as edges (E), a directed graph structure G=(V, E) is constructed, and it is stored in a knowledge graph database (such as Neo4j). At the same time, the unstructured knowledge is associated to the corresponding entity node through RAG technology to form a basic knowledge base combining “structured relationship + unstructured text”. A plurality of historical operation data of the ICT system is extracted from the historical data, records containing fault occurrence time, fault entity, and device entity are screened, and a structured data set is formed. Taking the fault occurrence time as the key dimension, the fault entity combination appearing at the same time in the same time window is counted, the support and confidence are calculated by the Apriori algorithm, the fault combination satisfying the threshold is screened, and the implicitly associated fault entity is obtained. Based on the above implicitly associated fault entity, the device entity corresponding to the fault entity in the historical data is retrieved, and the implicitly associated device entity is obtained. The implicitly associated fault entity and the implicitly associated device entity are added to the basic ICT system knowledge base to update the basic association relationship, and the final ICT system knowledge base is obtained.

[0036] The scheme can improve the comprehensiveness of the construction of the ICT system knowledge base, and is beneficial to the efficiency of the subsequent complex fault maintenance strategy generation result in the ICT system.

[0037] In S103, a plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the business information, and the plurality of target maintenance strategies are integrated to generate an ICT system maintenance strategy.

[0038] The strategy condition can be a prerequisite, resource requirement, and applicable range of each candidate maintenance strategy, and the like. The strategy condition limits whether the strategy can be executed. The business constraint condition can be a limitation condition such as available resources and corresponding executable time of the ICT system that can be used to execute each business.

[0039] In an embodiment, it is judged whether the strategy condition in each candidate maintenance strategy conflicts with the business constraint condition in the business information. For example, the candidate maintenance strategy is shutdown inspection, and the business constraint condition is that the running period cannot be shut down, so the strategy condition in the candidate maintenance strategy conflicts with the business constraint condition in the business information. The candidate maintenance strategy with the condition conflict is filtered out to obtain a plurality of target maintenance strategies corresponding to the fault cause, and the operations in the plurality of target maintenance strategies are combined and de-duplicated according to a preset strategy logic to obtain an ICT system maintenance strategy. The preset strategy logic can be a basic execution sequence logic in the strategy execution process. For example, a standby thread is activated first and then a load is shared by using a thread.

[0040] The technical scheme provided in the embodiments of the present application, in the case that the fault risk operation data in the ICT system is predicted, identifies the service information, the device information and the fault type parameter associated with the fault risk operation data, and determines the fault complexity corresponding to the fault risk operation data according to the service information, the device information and the fault type parameter; in the case that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, determines the fault cause corresponding to the fault risk operation data in the preset ICT system knowledge base, and a plurality of candidate maintenance strategies corresponding to the fault cause; determines a plurality of target maintenance strategies corresponding to the fault cause according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the service information, and integrates the plurality of target maintenance strategies to generate an ICT system maintenance strategy. Through the above ICT system maintenance strategy generation method, the problems of system maintenance not being timely and low maintenance efficiency in the prior art are solved. By determining the fault complexity corresponding to the predicted fault risk operation data, in the case that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, a plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy condition in each candidate maintenance strategy and the business constraint condition in the service information, and the plurality of target maintenance strategies are integrated to generate an ICT system maintenance strategy, the purpose of generating an ICT system maintenance strategy in advance before a system fault occurs can be achieved, the timeliness and efficiency of system maintenance are improved, and the generation of the ICT system maintenance strategy can be combined with the business constraint condition of the operation data, ensuring the success rate of the ICT system maintenance.

[0041] Figure 2 The flowchart of determining the target maintenance strategy provided in the embodiments of the present application is shown in FIG. 1. As shown in FIG. 1, the specific steps include the following steps: Figure 2 S201, consistency check is performed on the strategy condition in each candidate maintenance strategy and the business constraint condition in the service information.

[0042] In an embodiment, the strategy condition in the candidate maintenance strategy can be compared with the business constraint condition in the service information, and whether there is a same condition in the strategy condition and the business constraint condition is compared.

[0043] ​In an embodiment, the business constraints include maintenance time constraints, available resource constraints, and maintenance risk dimension constraints; the consistency check on the policy conditions in each candidate maintenance strategy and the business constraints in the business information includes: determining the execution time constraints, the policy resource dependency constraints, and the policy risk dimensions of each candidate maintenance strategy according to the policy conditions in each candidate maintenance strategy, and generating a strategy demand set corresponding to each candidate maintenance strategy according to the execution time constraints, the policy resource dependency constraints, and the policy risk dimensions; generating a maintenance constraint set corresponding to the fault risk operation data according to the maintenance time constraints, the available resource constraints, and the maintenance risk dimension constraints, and calculating the intersection of the strategy demand set and the maintenance constraint set; judging whether the intersection is empty, and determining that the policy conditions are consistent with the business constraints in the case that the intersection is not empty.

[0044] The maintenance time constraints can be conditions for limiting the implementation time of maintenance operations by each business in the ICT system. The maintenance time constraints can avoid the impact of maintenance operations on the operation of core businesses. The available resource constraints can be resources available for the operation of each business or used in the maintenance process in the ICT system. The available resource constraints can include hardware constraints, software constraints, and the number of maintainable personnel constraints in the ICT system, etc. The maintenance risk dimension constraints can be conditions for limiting the risks that may be caused by maintenance operations. The maintenance risk dimension constraints can ensure that the maintenance operations do not exceed the business risk tolerance. The maintenance risk dimension constraints can include prohibited risk types such as business interruption, data loss, and compliance violation; allowed risk levels such as low level, medium level, and high level; and risk response requirements such as the need to develop a rollback plan in advance for high-risk maintenance operations, etc. The execution time constraints can be the time prerequisites for the execution of the candidate maintenance strategy. The policy resource dependency constraints can be the resource prerequisites required for the execution of the candidate maintenance strategy. The policy risk dimensions can be the risk types and risk levels that may be caused by the execution of the candidate maintenance strategy.

[0045] In an embodiment, the constraints prerequisites that each candidate maintenance strategy can execute can be determined according to the policy conditions in each candidate maintenance strategy, including the execution time constraints, the policy resource dependency constraints, and the policy risk dimensions. The execution time constraints, the policy resource dependency constraints, and the policy risk dimensions are integrated to generate a strategy demand set corresponding to each candidate maintenance strategy. The maintenance time constraints, the available resource constraints, and the maintenance risk dimension constraints are integrated according to the dimensions of the strategy demand set to generate a maintenance constraint set corresponding to the fault risk operation data, and the intersection of the strategy demand set and the maintenance constraint set is calculated. It is judged whether there are same constraints in the intersection. In the case that the intersection is not an empty set, i.e., there are same constraints, it is indicated that there are candidate maintenance strategies that are needed but are limited by the conditions of the ICT system, and at this time it can be determined that the policy conditions are consistent with the business constraints.

[0046] The scheme determines the execution time constraint, the policy resource dependency constraint and the policy risk dimension of each candidate maintenance strategy through the policy condition in each candidate maintenance strategy, generates a policy requirement set, generates a maintenance constraint set according to the maintenance time constraint, the available resource constraint and the maintenance risk dimension constraint, and calculates the intersection of the set to determine whether the policy condition is consistent with the business constraint condition, which can improve the comprehensiveness of the consistency check of the policy condition and the business constraint condition, and further improve the accuracy of subsequent screening of target maintenance strategies.

[0047] S202, in the case that the policy condition is consistent with the business constraint condition, the candidate maintenance strategy corresponding to the policy condition is filtered out to obtain a plurality of target maintenance strategies corresponding to the fault cause.

[0048] In one embodiment, in the case that the policy condition is consistent with the business constraint condition, it is indicated that there is a candidate maintenance strategy requirement, but the condition is limited by the ICT system, at this time, the candidate maintenance strategy needs to be filtered out to avoid failure of the maintenance strategy execution, to obtain a plurality of target maintenance strategies corresponding to the fault cause.

[0049] The technical means provided by the embodiments of the present application can avoid the problem that the maintenance strategy cannot be executed on site, and improve the accuracy of screening of the target maintenance strategy, by performing consistency check on the policy condition in each candidate maintenance strategy and the business constraint condition in the business information, filtering out the candidate maintenance strategy corresponding to the policy condition in the case that the policy condition is consistent with the business constraint condition, and obtaining a plurality of target maintenance strategies corresponding to the fault cause.

[0050] Figure 3 is a flowchart for determining the fault complexity provided by the embodiments of the present application. As shown in Figure 3 The business information includes business basic information, business constraint conditions and business running conditions; the device information includes fault device information and dependent device information, and specifically includes the following steps: S301, determining the proportion of the business constraint condition in the business running condition according to the first number of the business constraint condition and the second number of the business running condition, to obtain the business constraint strength corresponding to the fault risk running data.

[0051] The business basic information can be information describing business attributes and business characteristics such as business name, priority, department to which the business belongs and business function. The business constraint condition can be operation limit such as running time and continuity in the business running process. The business running condition can be a limit for supporting the normal running of the business. The dependent device information can be information of other devices that have physical or logical association with the fault device and support the running of the same business. The business constraint strength can be data quantifying the size of the business maintenance limit.

[0052] In one embodiment, a ratio of a first number of the business constraint conditions to a second number of the business operation conditions is calculated to obtain a proportion of the business constraint conditions in the business operation conditions, and the proportion is taken as the business constraint strength corresponding to the fault risk operation data.

[0053] In S302, the business operation logic is determined according to the business basic information, the device dependency is determined according to the business operation logic, the first device function in the fault device information and the second device function in the dependent device information, and the device association degree corresponding to the fault risk operation data is obtained.

[0054] The business operation logic can be a structured framework describing the cooperation relationship between the business core process and the device. The first device function can be a specific function of the fault device in the business operation logic. The second device function can be a specific function of the dependent device in the business operation logic. The device dependency can be an index quantifying the association relationship between the dependent device and the fault device. The device association degree can be a quantification of the potential influence range of the fault on the device cluster.

[0055] In one embodiment, the business execution sequence in the business basic information can be read, the business operation logic can be determined according to the business execution sequence, the device functions required by each operation stage can be determined according to the business operation logic, and whether the first device function in the fault device information and the second device function in the dependent device information are in a dependent relationship and the device dependency can be determined according to the device functions required by each operation stage, so as to obtain the device association degree corresponding to the fault risk operation data.

[0056] In one embodiment, the number of the dependent device information is multiple; the device dependency is determined according to the business operation logic, the first device function in the fault device information and the second device function in the dependent device information, including: respectively matching the first device function in the fault device information and the second device function in each dependent device information with the business stage function corresponding to the business operation logic, to determine the first business stage corresponding to the first device function, and the second business stage corresponding to each second device function; determining the business processing time sequence relationship between the first business stage and each second business stage, determining the influence function number of the first device function according to the time sequence relationship, and determining the associated function number of the first device function according to the number of the multiple second device functions; determining the device dependency based on the preset influence weight, the preset association weight, the influence function number and the associated function number.

[0057] In one embodiment, the first device function in the fault device information and the second device function in each dependent device information can be respectively compared with the service stage function corresponding to the service running logic, to determine the function in the service stage function that is the same as the first device function and the service stage to which the function belongs, to obtain the first service stage, and similarly obtain the second service stage corresponding to each second device function. The service processing time sequence relationship between the first service stage and each second service stage is determined according to the sequence of the service stages, the number of device functions executed after the first device is determined according to the time sequence relationship, the influence function number of the first device function is obtained, and the number of multiple second device functions is taken as the associated function number of the first device function, and the device dependency is calculated according to the preset influence weight, the preset association weight, the influence function number and the associated function number.

[0058] The scheme can match the device function with the service stage function corresponding to the service running logic to determine the service stage, determine the influence function number and the associated function number of the first device function according to the service stage, and then calculate the device dependency, so as to accurately quantify the device dependency in combination with the service running logic and the device function, and improve the accuracy of the device dependency calculation result.

[0059] S303, a plurality of basic fault type parameters corresponding to the service running logic are determined, the coincidence degree of each basic fault type parameter and the fault type parameter is calculated, the fault influence range corresponding to the fault risk running data is obtained, and the fault complexity corresponding to the fault risk running data is obtained by integrating the business constraint strength, the device association degree and the fault influence range.

[0060] The basic fault type parameter can be a set of core characteristic parameters of a historical common fault in the business scene that is strongly associated with the service running logic. The fault influence range can be a business link and a device cluster range that can be affected by the current fault.

[0061] In one embodiment, a plurality of basic fault type parameters corresponding to the service running logic can be determined according to the common fault type of the preset service running stage and a plurality of service running stages corresponding to the service running logic, the similarity of each basic fault type parameter and the fault type parameter is calculated to obtain the coincidence degree of each basic fault type parameter and the fault type parameter, the basic fault type with a coincidence degree greater than a preset coincidence degree threshold is taken as a fault type that can be caused by the fault risk running data, the number of fault types that can be caused by the fault risk running data is taken as the fault influence range corresponding to the fault risk running data, the business constraint strength, the device association degree and the fault influence range are normalized, and the normalized sum of each parameter is calculated to obtain the fault complexity corresponding to the fault risk running data.

[0062] The technical scheme provided by the embodiment of the application can obtain a service constraint strength by calculating the proportion of a first quantity of service constraint conditions and a second quantity of service running conditions, determine a service running logic and a device function according to service basic information, determine a device correlation degree, calculate the coincidence degree of each basic fault type parameter and a fault type parameter, obtain a fault influence range, and integrate the service constraint strength, the device correlation degree and the fault influence range to obtain a fault complexity corresponding to the fault risk running data, so that the purpose of combining service characteristics, device characteristics and fault types for fault complexity evaluation is achieved, and the accuracy of fault complexity evaluation is further improved.

[0063] Figure 4 is a structural diagram of an ICT system maintenance strategy generation system provided by the embodiment of the application. As shown in Figure 4 , specifically includes the following: The resource inspection module 401 is configured to start intelligent inspection and maintenance processes of the ICT system.

[0064] The data collection module 402 is configured to collect running data, document knowledge and sensor data from multiple source systems, construct an ICT system knowledge base, and further configured to collect sensor or system interface data in real time for intelligent early warning.

[0065] The knowledge base construction module 403 is configured to integrate multiple source data, establish a unified data asset directory, construct a device-fault-maintenance semantic network by using knowledge graph technology, store vectors to achieve millisecond-level retrieval, and use a dynamic updating mechanism driven by reinforcement learning.

[0066] The knowledge base construction module 403 includes a multiple source data integration unit 4031, a knowledge graph construction unit 4032 and a vector storage unit 4033. The multiple source data integration unit 4031 is configured to integrate the multiple source data collected by the data collection module 402, and integrate running data, standard operating procedures, fault maintenance manuals and other heterogeneous data from dispersed systems (such as production management, sensors and document libraries) to establish a unified data asset directory. The knowledge graph construction unit 4032 is configured to model the entity relationships of devices, faults and maintenance by using knowledge graph technology as a semantic network, and formalize it as a directed graph (G=(V,E)), where (V) represents entity nodes (such as device ID, fault type), and (E) represents relationship edges (such as "cause" and "dependence"). For example, the complex relationship between devices, faults and maintenance in a power distribution station system is captured by knowledge graph to support intelligent reasoning. The vector storage unit 4033 adopts a three-layer architecture of "large model + vector database + knowledge plug-in". The model layer integrates DeepSeek, Qianwen and other LLMs to understand the problem semantics through an attention mechanism; the vector storage layer uses Milvus and other databases to embed text blocks into high-dimensional vectors , build millisecond-level retrieval index; the knowledge plug-in layer converts user queries into vectors , calculates cosine similarity , recalls relevant text blocks, and generates natural language answers by LLM integration.

[0067] The intelligent early warning module 404 is used for anomaly detection by the LinSeer TS time series prediction model, alarm by multi-algorithm fusion, root cause analysis by Bayesian reasoning, and complex reasoning by the LinSeer O1 deep thinking model.

[0068] The intelligent early warning module 404 includes a time series prediction analysis unit 4041, a multi-algorithm fusion alarm unit 4042, a deep thinking reasoning unit 4043, and a root cause analysis unit 4044. The time series prediction analysis unit 4041 uses the LinSeer TS time series prediction model to detect anomalies in device state data (such as CPU usage, network traffic). Let the time series data be , the model uses a long short-term memory network (LSTM) to capture long-term dependencies, and its gating mechanism formula is:

[0069] where are the forget gate, input gate, and output gate, respectively, is the neuron state, is the hidden state. The model output prediction value is compared with the threshold value to achieve early warning 24-72 hours in advance.

[0070] The multi-algorithm fusion alarm unit 4042 integrates alarm strategies based on threshold values, statistical analysis, and machine learning. For fault root cause analysis, Bayesian reasoning is used to calculate the hypothesis probability , where is the fault hypothesis, is the evidence (such as sensor data). For example, the Redis cache hit rate drops sharply, relevant documents are recalled by vector retrieval, and a solution is generated by LLM.

[0071] The deep thinking reasoning unit 4043 uses the LinSeer O1 model to improve complex reasoning ability through multi-dimensional reinforcement learning technology through step-by-step weighted reward algorithm. Define the value function , where is the discount factor, and the decision path is optimized by the policy gradient method.

[0072] The natural language interaction module 405 is used to receive query languages from different terminal sources.

[0073] The application layer function module 406 is configured to perform semantic recognition and intelligent query on the natural language received by the natural language interaction module 405.

[0074] The application layer function module 406 includes an intelligent question and answer unit 4061, a visualization unit 4062, and a multi-terminal collaboration unit. The intelligent question and answer unit 4061 relies on the powerful Chinese understanding capability of the DeepSeek-R1 model to analyze user natural language queries. For example, inputting “power load calculation method”, the system automatically disassembles it into sub-tasks such as the required coefficient method and the unit area method, and the instruction analysis accuracy rate is above 95%. The multi-terminal collaboration unit supports seamless switching between mobile terminals, tablets, and PC terminals, and improves response efficiency through asynchronous mechanism and batch processing combined calculation. The request processing model adopts queue theory, sets the request arrival rate as (λ) and the service rate as (μ), and then the system utilization rate (ρ = λ / μ) is obtained, and (ρ < 1) is ensured through load balancing.

[0075] The policy execution module 407 is configured to generate a maintenance strategy according to the alarm information and the fault root cause output by the intelligent early warning module 404.

[0076] The policy execution module 407 includes a fault complexity analysis unit 4071, a maintenance strategy generation unit 4072, and a feedback and optimization unit 4073. The fault complexity analysis unit 4071 is configured to determine the fault complexity and the generation mode of the maintenance strategy according to the alarm information. The maintenance strategy generation unit 4072 is configured to generate a corresponding maintenance scheme in combination with the fault root cause and the alarm information. The feedback and optimization unit 4073 is configured to evaluate the effect and update the system maintenance according to the maintenance strategy to improve the effect of the ICT system maintenance.

[0077] The vectorization storage unit 4031 and the adaptive learning sub-unit 40732 are configured to dynamically update the ICT system knowledge base, introduce reinforcement learning (RL) to optimize the knowledge base, and adjust the blocking strategy and the embedding model parameters in real time according to the operation and maintenance feedback. The reward function is defined as wherein represents the state (such as the knowledge base coverage rate), represents the action (such as the adjustment of the blocking size), and the policy is updated through the Q learning algorithm . For example, the tape machine device reduces the fault handling time through continuous iteration.

[0078] The technical scheme provided by the embodiment of the application solves the pain points of knowledge islands, response lag and inefficient decision-making of traditional operation and maintenance, provides an efficient, reliable and economical solution for ICT resource operation and maintenance, greatly improves system operation and maintenance efficiency, timeliness, system knowledge retrieval efficiency, system operation and maintenance safety risk, and the self-adaptive learning mechanism makes the system more intelligent, supports hybrid cloud deployment and multi-terminal collaboration, seamlessly integrates existing enterprise tools, and improves response efficiency and stability.

[0079] Figure 5 is a structural block diagram of an ICT system maintenance strategy generation device provided by the embodiment of the application. As shown in Figure 5 , specifically includes the following: The fault complexity determination module 501 is configured to, in the case that the fault risk operation data in the ICT system is predicted, identify the business information, device information and fault type parameters associated with the fault risk operation data, and determine the fault complexity corresponding to the fault risk operation data according to the business information, device information and fault type parameters. The candidate strategy determination module 502 is configured to, in the case that the fault complexity is greater than or equal to the preset multi-strategy maintenance fault complexity threshold, determine the fault cause corresponding to the fault risk operation data in the preset ICT system knowledge base, and determine a plurality of candidate maintenance strategies corresponding to the fault cause. The maintenance strategy generation module 503 is configured to determine a plurality of target maintenance strategies corresponding to the fault cause according to the strategy conditions in each candidate maintenance strategy and the business constraint conditions in the business information, and integrate the plurality of target maintenance strategies to generate an ICT system maintenance strategy.

[0080] Further, the maintenance strategy generation module 503 is specifically configured to: perform consistency checking on the strategy conditions in each candidate maintenance strategy and the business constraint conditions in the business information; in the case that the strategy conditions and the business constraint conditions are consistent, filter out the candidate maintenance strategy corresponding to the strategy conditions to obtain a plurality of target maintenance strategies corresponding to the fault cause.

[0081] Further, the business constraint conditions include maintenance time constraints, available resource constraints and maintenance risk dimension constraints. The maintenance strategy generation module 503 is specifically configured to: determine the execution time constraints, strategy resource dependency constraints and strategy risk dimensions of each candidate maintenance strategy according to the strategy conditions in each candidate maintenance strategy, and generate a strategy requirement set corresponding to each candidate maintenance strategy according to the execution time constraints, strategy resource dependency constraints and strategy risk dimensions. generate a maintenance constraint set corresponding to the fault risk running data according to the maintenance time constraint, the available resource constraint and the maintenance risk dimension constraint, and calculate an intersection of the strategy requirement set and the maintenance constraint set; determine whether the intersection is empty, and in a case where the intersection is not empty, determine that the strategy condition is consistent with the business constraint condition.

[0082] Further, the business information includes business basic information, business constraint conditions and business running conditions; the equipment information includes fault equipment information and dependent equipment information. The fault complexity determination module 501 is specifically configured to: determine a proportion of the business constraint conditions in the business running conditions according to a first number of the business constraint conditions and a second number of the business running conditions, to obtain a business constraint strength corresponding to the fault risk running data; determine a business running logic according to the business basic information, determine an equipment dependency degree according to the business running logic, a first equipment function in the fault equipment information and a second equipment function in the dependent equipment information, to obtain an equipment association degree corresponding to the fault risk running data; determine a plurality of basic fault type parameters corresponding to the business running logic, calculate a coincidence degree of each basic fault type parameter and a fault type parameter, to obtain a fault influence range corresponding to the fault risk running data, and integrate the business constraint strength, the equipment association degree and the fault influence range to obtain the fault complexity corresponding to the fault risk running data.

[0083] Further, the number of the dependent equipment information is a plurality. The fault complexity determination module 501 is specifically configured to: match the first equipment function in the fault equipment information and the second equipment function in each dependent equipment information with a business stage function corresponding to the business running logic, to determine a first business stage corresponding to the first equipment function, and a second business stage corresponding to each second equipment function; determine a business processing time sequence relationship between the first business stage and each second business stage, determine an influence function number of the first equipment function according to the time sequence relationship, and determine an associated function number of the first equipment function according to the number of the plurality of second equipment functions; determine the equipment dependency degree based on a preset influence weight, a preset association weight, the influence function number and the associated function number.

[0084] Further, the candidate strategy determination module 502 is specifically configured to: The multi-source ICT system knowledge data is acquired, semantic recognition is performed on entity names of a plurality of entities in the multi-source ICT system knowledge data, entity names of the same semantics are uniformly processed according to a semantic recognition result, and a basic ICT system knowledge base is constructed based on a preset knowledge base construction rule and a naming uniform processing result. The plurality of entities include a plurality of fault entities and a plurality of device entities. A plurality of historical operation data of the ICT system are acquired, a plurality of implicit associated fault entities under the same fault occurrence time are determined according to fault occurrence times corresponding to the fault entities in the plurality of historical operation data, and an implicit associated device entity corresponding to each implicit associated fault entity is determined. Based on the implicit associated fault entity and the implicit associated device entity, a basic associated relationship in the basic ICT system knowledge base is updated to obtain a final ICT system knowledge base.

[0085] Further, the fault complexity determination module 501 is specifically configured to: The current operation data and the plurality of historical operation data of the ICT system are acquired, and an operation data change rule of the ICT system is determined based on the plurality of historical operation data. The running timestamp of the current operation data is determined, and the theoretical operation data corresponding to the running timestamp is predicted according to the operation data change rule. The data deviation between the theoretical operation data and the current operation data is calculated, and in the case that the data deviation is greater than a preset data deviation threshold, it is determined that there is fault risk operation data in the ICT system.

[0086] The technical scheme provided in the embodiments of the present application comprises a fault complexity determination module, which is configured to, in the case that it is predicted that there is fault risk operation data in an ICT system, identify business information, device information and fault type parameters associated with the fault risk operation data, and determine a fault complexity corresponding to the fault risk operation data according to the business information, the device information and the fault type parameters; a candidate strategy determination module, which is configured to, in the case that the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, determine a fault cause corresponding to the fault risk operation data in a preset ICT system knowledge base, and a plurality of candidate maintenance strategies corresponding to the fault cause; and a maintenance strategy generation module, which is configured to determine a plurality of target maintenance strategies corresponding to the fault cause according to strategy conditions in each candidate maintenance strategy and business constraint conditions in the business information, and integrate the plurality of target maintenance strategies to generate an ICT system maintenance strategy. Through the above-mentioned ICT system maintenance strategy generation device, the problems of untimely system maintenance and low maintenance efficiency in the prior art are solved. By determining the fault complexity corresponding to the predicted fault risk operation data, in the case that the fault complexity is greater than or equal to the preset multi-strategy maintenance fault complexity threshold, the plurality of target maintenance strategies corresponding to the fault cause are determined according to the strategy conditions in each candidate maintenance strategy and the business constraint conditions in the business information, and the plurality of target maintenance strategies are integrated to generate the ICT system maintenance strategy, the purpose of generating the ICT system maintenance strategy in advance before the system fault occurs can be achieved, the timeliness and efficiency of system maintenance are improved, and the generation of the ICT system maintenance strategy can be combined with the business constraint conditions of the operation data, so that the success rate of the ICT system maintenance is ensured.

[0087] The ICT system maintenance strategy generation device in the embodiments of the present application can be configured in a device, or in a component, an integrated circuit or a chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a Network Attached Storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., and the embodiments of the present application are not limited in this regard.

[0088] The ICT system maintenance strategy generation apparatus in the embodiments of the present applicationapplicationbe an operating system. The operating systemapplicationbe an Android operating system, an ios operating system, or other possible operating systems, and the embodiments of the present application do not make specific limitations.

[0089] The ICT system maintenance strategy generation apparatus provided in the embodiments of the present applicationapplicationimplement the processes implemented by the method embodiments described above, and thus the details are not described herein again.

[0090] As shown in Figure 6 The embodiments of the present application further provide an electronic device 600, which includes a processor 601, a memory 602, and a program or instruction stored in the memory 602 and executable on the processor 601. When the program or instruction is executed by the processor 601, the processes of the above-described ICT system maintenance strategy generation method embodiments are implemented, and the same technical effects are achieved. Thus, the details are not described herein again.

[0091] It should be noted that the electronic device in the embodiments of the present applicationapplicationinclude the mobile electronic device and the non-mobile electronic device described above.

[0092] The embodiments of the present application further provide a readable storage medium, which stores a program or instruction. When the program or instruction is executed by a processor, the processes of the above-described ICT system maintenance strategy generation method embodiments are implemented, and the same technical effects are achieved. Thus, the details are not described herein again.

[0093] The processorapplicationbe the processor in the electronic device described in the above embodiments. The readable storage mediumapplicationinclude a computer readable storage medium, such as a computer readable only memory (ROM), a random access memory (RAM), a magnetic disc or an optical disc, etc.

[0094] The embodiments of the present application further provide a program product, which includes program code. When the program product is run on a computer device, the program codeapplicationbe used to make the computer device execute the steps in the methods according to the various exemplary embodiments of the present application described above in the specification, for example, the computer deviceapplicationexecute the ICT system maintenance strategy generation method described in the embodiments of the present application. The program productapplicationbe implemented by using any combination of one or more readable media.

[0095] It should be noted that, in the present document, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a", "comprising", or "includes a", does not, without more constraints, preclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element. Additionally, it should be noted that the methods and apparatus of the present embodiments are not limited by the order of the steps or the sequence for performing the steps, as some steps can occur in different orders and / or concurrently with each other. Furthermore, the features of certain examples can be combined with features of other examples.

[0096] From the above description of the embodiments, it is apparent that the above-mentioned method of the embodiments can be realized by means of software and necessary universal hardware platforms, of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a plurality of instructions for making a terminal (which can be a mobile phone, computer, server, or network equipment, etc.) execute the method described in each embodiment of the present application.

[0097] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, not restrictive, and those skilled in the art can make many forms under the inspiration of the present application without departing from the scope of the present application and the protection scope of the claims.

[0098] The above are only the preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, re-adjustments and replacements made by those skilled in the art do not deviate from the protection scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without deviating from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A method for generating ICT system maintenance strategies, characterized in that, The method includes: When it is predicted that there is fault risk in the ICT system, the business information, equipment information and fault type parameters associated with the fault risk are identified, and the fault complexity corresponding to the fault risk is determined based on the business information, the equipment information and the fault type parameters. When the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, determine the fault cause corresponding to the fault risk operation data in the preset ICT system knowledge base, and multiple candidate maintenance strategies corresponding to the fault cause. Based on the policy conditions in each of the candidate maintenance strategies and the business constraints in the business information, multiple target maintenance strategies corresponding to the cause of the fault are determined, and the multiple target maintenance strategies are integrated to generate an ICT system maintenance strategy.

2. The ICT system maintenance strategy generation method according to claim 1, characterized in that, The step of determining multiple target maintenance strategies corresponding to the cause of the fault based on the strategy conditions in each of the candidate maintenance strategies and the business constraints in the business information includes: Perform a consistency check between the strategy conditions in each of the candidate maintenance strategies and the business constraints in the business information; When the policy conditions are consistent with the business constraints, the candidate maintenance policies corresponding to the policy conditions are filtered out to obtain multiple target maintenance policies corresponding to the cause of the failure.

3. The ICT system maintenance strategy generation method according to claim 2, characterized in that, The business constraints include maintenance time constraints, available resource constraints, and maintenance risk dimension constraints. The consistency verification of the policy conditions in each of the candidate maintenance strategies and the business constraints in the business information includes: Based on the policy conditions in each of the candidate maintenance policies, determine the execution time constraints, policy resource dependency constraints, and policy risk dimensions of each candidate maintenance policy, and generate a set of policy requirements corresponding to each candidate maintenance policy based on the execution time constraints, policy resource dependency constraints, and policy risk dimensions. Based on the maintenance time constraints, available resource constraints, and maintenance risk dimension constraints, a set of maintenance constraints corresponding to the fault risk operation data is generated, and the intersection of the strategy requirement set and the maintenance constraint set is calculated. Determine whether the intersection is empty. If the intersection is not empty, determine that the strategy condition is consistent with the business constraint condition.

4. The ICT system maintenance strategy generation method according to claim 1, characterized in that, The business information includes basic business information, business constraints, and business operating conditions; the equipment information includes faulty equipment information and dependent equipment information. The step of determining the fault complexity corresponding to the fault risk operation data based on the business information, the equipment information, and the fault type parameter includes: The proportion of the business constraints in the business operation conditions is determined based on the first number of the business constraints and the second number of the business operation conditions, thereby obtaining the business constraint strength corresponding to the fault risk operation data. The business operation logic is determined based on the business basic information. The equipment dependency is determined based on the business operation logic, the first equipment function in the fault equipment information, and the second equipment function in the dependent equipment information, so as to obtain the equipment correlation degree corresponding to the fault risk operation data. Multiple basic fault type parameters corresponding to the business operation logic are determined, the overlap between each basic fault type parameter and the fault type parameter is calculated, the fault impact range corresponding to the fault risk operation data is obtained, and the business constraint strength, the equipment correlation degree and the fault impact range are integrated to obtain the fault complexity corresponding to the fault risk operation data.

5. The ICT system maintenance strategy generation method according to claim 4, characterized in that, The number of dependent device information items is multiple; The step of determining device dependency based on the business operation logic, the first device function in the fault device information, and the second device function in the dependent device information includes: The first device function in the fault device information and the second device function in each of the dependent device information are matched with the business stage function corresponding to the business operation logic to determine the first business stage corresponding to the first device function and the second business stage corresponding to each of the second device functions. Determine the business processing sequence relationship between the first business stage and each of the second business stages, determine the number of functions affected by the first device function based on the sequence relationship, and determine the number of associated functions of the first device function based on the number of the plurality of second device functions; Device dependency is determined based on preset influence weights, preset association weights, the number of influencing functions, and the number of associated functions.

6. The ICT system maintenance strategy generation method according to claim 1, characterized in that, The construction process of the preset ICT system knowledge base includes: Acquire multi-source ICT system knowledge data, perform semantic recognition on the entity names of multiple entities in the multi-source ICT system knowledge data, perform unified naming processing on entity names with the same semantics according to the semantic recognition results, and construct a basic ICT system knowledge base based on preset knowledge base construction rules and unified naming processing results. The multiple entities include multiple fault entities and multiple device entities. Obtain multiple historical operation data of the ICT system, determine multiple implicitly related fault entities under the same fault occurrence time based on the fault occurrence time corresponding to each fault entity in the multiple historical operation data, and determine the implicitly related device entities corresponding to each of the implicitly related fault entities. The basic relationships in the basic ICT system knowledge base are updated based on the implicitly associated fault entities and the implicitly associated device entities to obtain the final ICT system knowledge base.

7. The ICT system maintenance strategy generation method according to claim 1, characterized in that, The process of predicting whether there are operational data risks in ICT systems includes: Obtain current operating data and multiple historical operating data of the ICT system, and determine the changing patterns of the operating data of the ICT system based on the multiple historical operating data; Determine the running timestamp of the current running data, and predict the theoretical running data corresponding to the running timestamp based on the changing pattern of the running data; The data deviation between the theoretical operating data and the current operating data is calculated. If the data deviation is greater than a preset data deviation threshold, it is determined that there is fault-risk operating data in the ICT system.

8. An ICT system maintenance strategy generation device, characterized in that, The device includes: The fault complexity determination module is used to identify the business information, equipment information and fault type parameters associated with the fault risk operation data when it is predicted that there is fault risk operation data in the ICT system, and to determine the fault complexity corresponding to the fault risk operation data based on the business information, the equipment information and the fault type parameters. The candidate strategy determination module is used to determine, when the fault complexity is greater than or equal to a preset multi-strategy maintenance fault complexity threshold, the fault cause corresponding to the fault risk operation data in the preset ICT system knowledge base, and multiple candidate maintenance strategies corresponding to the fault cause. The maintenance strategy generation module is used to determine multiple target maintenance strategies corresponding to the cause of the fault based on the strategy conditions in each of the candidate maintenance strategies and the business constraints in the business information, and to integrate the multiple target maintenance strategies to generate an ICT system maintenance strategy.

9. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and running on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the ICT system maintenance strategy generation method as described in any one of claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the ICT system maintenance strategy generation method as described in any one of claims 1-7.

Images