A fault feature-based inspection method and device

Abstract

Embodiments of the present application disclose a fault feature-based inspection method and device, which are used to improve fault inspection accuracy and efficiency. The method of the embodiments of the present application comprises: a computing device performing a first inspection task, collecting fault feature information corresponding to an occurred fault, the fault feature information being used to indicate a fault context corresponding to a fault signal. The computing device performs inspection strategy retrieval in a historical fault mode library based on the fault feature information, and determines a first target inspection strategy, the first target inspection strategy being an inspection strategy corresponding to a target fault feature, the target fault feature including a historical fault feature in the historical fault mode, a similarity of which to the fault feature information being higher than a similarity threshold. The first target inspection strategy is used to adjust the first inspection strategy, and an adjusted first inspection strategy is used to perform a second inspection task, the second inspection task being separated from the first inspection task by one or more inspection periods.

Description

Technical Field

[0001] This application relates to the field of computers, and more particularly to a method and apparatus for inspection based on fault characteristics. Background Technology

[0002] As more and more enterprises move to the cloud and customer businesses continue to expand, the complexity of business operations and the volume of data carried by cloud platforms are also growing explosively. At the same time, cloud platform customers are also raising high expectations for the stability of cloud platforms. For example, cloud platform customers hope to discover potential risks to applications and cloud platforms in advance and accurately, so as to avoid failures.

[0003] In current cloud platform service assurance solutions, inspection is used by various cloud platforms as a means to discover potential faults in the cloud environment in advance. In the current inspection solution, operation and maintenance personnel need to formulate an inspection plan for the operation and maintenance object in advance. The inspection plan can define a set of inspection contents that can support the needs of daily inspections, and set different inspection cycles according to the importance of the inspection contents. During the inspection process, the inspection tasks only need to be executed repeatedly according to the established inspection plan cycle.

[0004] However, because the current inspection system operates according to a fixed inspection plan, it can only identify the fault modes included in the plan and cannot dynamically supplement or improve its inspection capabilities based on business operation status and fault characteristics. This results in poor inspection accuracy. Furthermore, when the inspection plan is complex, executing the complete plan requires collecting and processing a large amount of data, leading to a waste of inspection resources. Summary of the Invention

[0005] This application provides a fault feature-based inspection method to improve the accuracy and efficiency of fault inspection. This application also provides a fault feature-based inspection device, computing device, computing device cluster, computer-readable storage medium, and computer program product corresponding to the fault feature-based inspection method.

[0006] In a first aspect, embodiments of this application provide a fault feature-based inspection method. This method can be executed by a computing device, or by a component of the computing device, such as a processor, chip, or chip system, or by a logic module or software capable of implementing all or part of the functions of the computing device. The method provided in the first aspect includes: the computing device executing a first inspection task to collect fault feature information corresponding to an occurring fault. The first inspection task can be generated based on a first inspection strategy, the occurring fault can be a fault detected during the execution of the first inspection task, and the fault feature information is used to indicate the fault context corresponding to a fault signal, which includes alarm inputs received by the inspection system. The computing device performs an inspection strategy retrieval in a historical fault pattern library based on the fault feature information to determine a first target inspection strategy. The historical fault pattern library includes historical fault features and corresponding inspection strategies. The first target inspection strategy is the inspection strategy corresponding to the target fault features, and the target fault features include historical fault features in historical fault patterns with a similarity higher than a similarity threshold to the fault feature information. The computing device adjusts the first inspection strategy based on the first target inspection strategy. The adjusted inspection strategy is used to execute the second inspection task. The second inspection task and the first inspection task are separated by one or more inspection cycles.

[0007] In this embodiment, the computing device executes the first inspection task by retrieving a first target inspection strategy from a historical fault mode library based on the fault feature information corresponding to the fault signal that caused the fault, and adjusting the first inspection strategy based on the first target inspection strategy. Then, the second inspection task is executed according to the adjusted inspection strategy. Compared to the prior art where the computing device performs inspections based on a fixed inspection strategy package, the inspection method in this embodiment can dynamically adjust the inspection strategy based on fault features, thereby improving the accuracy of fault inspection. Furthermore, since the computing device can retrieve the first target inspection strategy based on fault feature information, it can selectively select inspection content from the inspection strategy, avoiding some invalid inspection content, further improving the efficiency of fault inspection.

[0008] In one possible implementation, during the process of the computing device retrieving inspection strategies from a historical fault pattern database based on fault feature information, the computing device performs similarity matching between the fault feature information and one or more historical fault features in the historical fault pattern database to determine the target fault feature. Specifically, the computing device identifies fault features in the historical fault pattern database with similarity scores higher than a similarity threshold as target fault features. The computing device then determines the inspection strategy corresponding to the target fault feature as a first target inspection strategy, which includes inspection content and inspection cycle.

[0009] In this embodiment, the computing device performs similarity matching based on fault feature information and one or more historical fault features in the historical fault mode library, thereby determining the inspection strategy corresponding to the historical fault feature information with a similarity higher than the similarity threshold as the target inspection strategy, thereby improving the efficiency of generating the target inspection strategy.

[0010] In one possible implementation, when there are multiple historical fault feature information in the historical fault mode library that have a similarity higher than the similarity threshold with the fault feature information, the inspection subsystem can determine the target fault feature from the historical fault feature information with the highest similarity, and the inspection strategy corresponding to the historical fault feature information with the highest similarity is determined as the first target inspection strategy.

[0011] In this embodiment of the application, the computing device can determine the target fault feature by the historical fault feature information with the highest similarity, thereby improving the feasibility of the computing device in determining the target fault feature and the first target inspection strategy.

[0012] In one possible implementation, the computing device generates a first fault feature vector based on fault feature information. The first fault feature vector includes multiple dimensions, with different dimensions corresponding to different fault features. The computing device calculates the similarity between the first fault feature vector and a second fault feature vector. The second fault feature vector includes one or more fault feature vectors from a historical fault pattern library. The similarity is used by the inspection system for similarity matching.

[0013] In this embodiment of the application, the computing device can represent fault feature information based on fault feature vectors, thereby improving the similarity between the fault feature information corresponding to the fault signal of the computing device and the fault feature information in the historical fault mode library.

[0014] In one possible implementation, when the number of fault signals received by the inspection system within a first time threshold exceeds the number threshold, the first target inspection strategy is adjusted. Adjusting the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy, or increasing the inspection content in the first target inspection strategy.

[0015] In this embodiment, the computing device can adjust the inspection strategy in a timely manner based on the number of fault signals received within a first time threshold. That is, the computing device can supplement the inspection capability based on the business status, thereby improving the accuracy of the inspection strategy.

[0016] In one possible implementation, during the process of adding inspection content to the first target inspection strategy, the computing device performs a union operation on the set of inspection items corresponding to fault feature information and the set of inspection items corresponding to the target fault features. The set of inspection items corresponding to fault feature information includes newly added inspection items obtained based on the analysis of fault feature information. That is, the computing device can analyze and add new inspection items based on fault feature information and merge the new inspection items with the inspection items in the current inspection content.

[0017] In this embodiment, the computing device can perform a union operation on the set of inspection items corresponding to the fault feature information and the set of inspection items corresponding to the target fault feature, thereby obtaining a new inspection strategy and improving the inspection accuracy of the inspection strategy.

[0018] In one possible implementation, when no fault feature information is matched in the historical fault mode library (i.e., the similarity calculated by the computing device based on the fault feature information and one or more historical fault feature information in the historical fault mode library is less than a similarity threshold), the computing device generates a second target inspection strategy based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information. The computing device can adjust the first inspection strategy based on the second target inspection strategy.

[0019] In this embodiment of the application, when no fault feature information is matched in the historical fault mode library, the computing device can generate a second target inspection strategy based on the fault feature information, and adjust the first inspection strategy based on the second target inspection strategy, thereby improving the feasibility of dynamically adjusting the first inspection strategy.

[0020] In one possible implementation, during the process of generating a second target inspection strategy based on fault feature information, the computing device can analyze the inspection items corresponding to the fault feature information based on expert knowledge, or it can analyze the inspection items corresponding to the fault feature information based on an artificial intelligence model. The artificial intelligence model is an artificial intelligence model trained based on the training data of the fault feature information and the inspection strategy, and there is no specific limitation.

[0021] In the embodiments of this application, the computing device can generate a second target inspection strategy in a variety of ways, thereby improving the feasibility of the computing device to generate a second target inspection strategy.

[0022] In one possible implementation, after the computing device generates a second target inspection strategy based on fault feature information, it adds the second target inspection strategy and the corresponding fault features to a historical fault mode library. The newly added fault feature information and the corresponding second target inspection strategy can be used as historical fault features for similarity matching.

[0023] In this embodiment of the application, after the computing device generates a target inspection strategy based on fault feature information, it can add the target inspection strategy and the fault features corresponding to the target inspection strategy to the historical fault mode library, thereby expanding the inspection strategy of the historical fault mode library and improving the similarity matching success rate of fault feature information in the historical fault mode library.

[0024] In one possible implementation, if the inspection system does not receive a fault signal within a second time threshold, it adjusts the first target inspection strategy. Adjusting the first target inspection strategy includes increasing the inspection cycle within the first target inspection strategy, wherein the increased inspection cycle cannot exceed the inspection cycle threshold. The percentage increase in the inspection cycle is positively correlated with the duration of the unreceived fault signal.

[0025] In this embodiment of the application, if the computing device does not receive a fault signal within a certain period of time, the inspection cycle in the first target inspection strategy can be increased, that is, the inspection frequency can be reduced, thereby saving fault inspection resources.

[0026] In one possible implementation, the fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. The monitoring information is used to indicate the operating status of the inspected object, the log information is used to indicate the operating history of the inspected object, and the topology information is used to indicate the topological relationship of the inspected object, including network topology or system topology.

[0027] In this embodiment of the application, the computing device can dynamically adjust the first inspection strategy based on various types of fault characteristic information, thereby improving the accuracy of the inspection strategy.

[0028] Secondly, embodiments of this application provide a fault feature-based inspection device, including a data acquisition unit and a processing unit. The processing unit executes a first inspection task, while the data acquisition unit acquires fault feature information corresponding to an occurring fault. This fault feature information indicates the fault context corresponding to a fault signal, which includes alarm inputs received by the inspection system. The processing unit performs an inspection strategy retrieval in a historical fault pattern database based on the fault feature information to determine a first target inspection strategy. The historical fault pattern database includes historical fault features and corresponding inspection strategies. The first target inspection strategy is the inspection strategy corresponding to the target fault features, which include historical fault features in historical fault patterns with a similarity higher than a similarity threshold to the fault feature information. The processing unit further adjusts the first inspection strategy based on the first target inspection strategy. The adjusted first inspection strategy is used to execute a second inspection task, which is spaced one or more inspection cycles apart from the first task.

[0029] In one possible implementation, the processing unit is further configured to perform similarity matching based on fault feature information and one or more historical fault features in a historical fault pattern library to determine a target fault feature. The inspection strategy corresponding to the target fault feature is determined as a first target inspection strategy, which includes inspection content and inspection cycle.

[0030] In one possible implementation, the processing unit is further configured to generate a first fault feature vector based on fault feature information. The first fault feature vector includes multiple dimensions, with different dimensions corresponding to different fault features. A similarity is calculated between the first fault feature vector and a second fault feature vector. The second fault feature vector includes one or more fault feature vectors from a historical fault pattern library. The similarity is used by the inspection system for similarity matching.

[0031] In one possible implementation, the processing unit is further configured to adjust the first target inspection strategy when the number of fault signals received by the inspection system within a first time threshold exceeds the number threshold. The adjustment of the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy and increasing the inspection content in the first target inspection strategy.

[0032] In one possible implementation, the processing unit is specifically used to perform a union process on the set of inspection items corresponding to the fault feature information and the set of inspection items corresponding to the target fault feature. The set of inspection items corresponding to the fault feature information includes newly added inspection items obtained based on the analysis of the fault feature information.

[0033] In one possible implementation, when no fault feature information is matched in the historical fault mode library, the processing unit is further configured to generate a second target inspection strategy based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information.

[0034] In one possible implementation, the processing unit is further configured to add the second target inspection strategy and the fault features corresponding to the second target inspection strategy to the historical fault mode library.

[0035] In one possible implementation, if the inspection system does not receive a fault signal within a second time threshold, the processing unit is specifically used to adjust the first target inspection strategy, which includes increasing the inspection cycle in the first target inspection strategy.

[0036] In one possible implementation, the fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. The monitoring information is used to indicate the operating status of the inspected object, the log information is used to indicate the operating history of the inspected object, and the topology information is used to indicate the topological relationship of the inspected object.

[0037] Thirdly, embodiments of this application provide a computing device including a processor coupled to a memory. The processor stores instructions, which, when executed by the processor, cause the computing device to perform the method described in the first aspect or any possible implementation thereof.

[0038] Fourthly, embodiments of this application provide a computing device cluster, which includes one or more computing devices. Each computing device includes a processor coupled to a memory. The processor is used to store instructions, which, when executed by the processor, cause the computing device cluster to perform the method described in the first aspect or any possible implementation thereof.

[0039] Fifthly, embodiments of this application provide a computer-readable storage medium having instructions stored thereon, which, when executed, cause a computer to perform the method described in the first aspect or any possible implementation thereof.

[0040] Sixthly, embodiments of this application provide a computer program product including instructions that, when executed, cause a computer to implement the method described in the first aspect or any possible implementation thereof.

[0041] It is understood that the beneficial effects achieved by any of the fault-characteristic-based inspection devices, computing devices, computing device clusters, computer-readable media, or computer program products provided above can be referred to the beneficial effects in the corresponding methods, and will not be repeated here. Attached Figure Description

[0042] Figure 1 A schematic diagram of the system architecture of a fault feature-based inspection system provided in this application embodiment;

[0043] Figure 2 A flowchart illustrating a fault feature-based inspection method provided in this application embodiment;

[0044] Figure 3 A flowchart illustrating another inspection method based on fault characteristics provided in this application embodiment;

[0045] Figure 4 This is a schematic diagram illustrating similarity matching based on fault feature information, provided as an embodiment of this application.

[0046] Figure 5 A schematic diagram illustrating the calculation of similarity based on fault feature vectors, provided as an embodiment of this application;

[0047] Figure 6 A schematic diagram illustrating the addition of inspection content as provided in an embodiment of this application;

[0048] Figure 7 A schematic diagram of a fault feature-based inspection device provided in an embodiment of this application;

[0049] Figure 8 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application;

[0050] Figure 9 This is a schematic diagram of the structure of a computing device cluster provided in an embodiment of this application;

[0051] Figure 10 This is a schematic diagram of another computing device cluster provided in an embodiment of this application. Detailed Implementation

[0052] This application provides a fault feature-based inspection method and apparatus to improve the accuracy and efficiency of fault inspection.

[0053] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0054] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0055] First, some of the terms used in the embodiments of this application are introduced to facilitate understanding of the technical solutions by those skilled in the art.

[0056] Fault inspection refers to a systematic inspection of equipment or systems. Its purpose is to promptly identify and locate fault points, analyze the causes of faults, and take rapid measures to repair them, thereby ensuring the normal operation of the equipment or system.

[0057] Inspection content refers to the specific items that need to be checked during the inspection process, including indicator values ​​that characterize whether a specific observation point of the inspected object is normal. The accuracy of the inspection content directly determines the completeness and accuracy of detecting sub-health in the system. For example, if threshold alarm rules are configured for the indicators in the inspection content, an alarm can be generated to notify the operations and maintenance personnel when the indicator value of the inspection content is abnormal.

[0058] The inspection cycle is the time interval for inspecting equipment or systems, usually measured in minutes, hours, days, etc. Adjusting the inspection cycle directly determines the timeliness of detecting sub-health in the system.

[0059] To make the technical solution of this application clearer and easier to understand, the system architecture of this application will be described below with reference to the accompanying drawings.

[0060] Please see Figure 1 , Figure 1 This application provides a schematic diagram of the system architecture for an inspection system. Figure 1 In the example shown, the inspection system 10 includes an inspection subsystem 101, an operation and maintenance subsystem 102, and an operation and maintenance object 103. The inspection subsystem 101 includes an inspection task management module 1011, an inspection strategy storage module 1012, a fault mode learner 1013, and a fault feature collector 1014. The functions of each part of the system are described in detail below.

[0061] The inspection subsystem 101 is used to generate inspection tasks according to the inspection strategy and send the inspection tasks to the operation and maintenance subsystem 101 or the operation and maintenance object 103. The inspection subsystem 101 is also used to generate and update the inspection strategy. Specifically, the inspection subsystem 101 can receive fault characteristic information reported by the operation and maintenance subsystem 102, and generate a new inspection strategy based on the fault characteristic information. The inspection subsystem 101 can update the current inspection strategy based on the newly generated inspection strategy, thereby realizing the dynamic adjustment of the inspection strategy.

[0062] The inspection subsystem 101 includes an inspection task management module 1011, an inspection strategy storage module 1012, a fault mode learner 1013, and a fault feature collector 1014. The inspection task management module 1011 manages inspection strategies and issues inspection tasks. Specifically, the inspection task management module 1011 can obtain inspection strategies from the fault mode learner 1013, generate inspection tasks according to the inspection strategies, and send the generated inspection tasks to the maintenance object 103.

[0063] The inspection strategy storage module 1012 is used to acquire and store the inspection strategy generated by the fault mode learner 1013, wherein the inspection strategy includes inspection content and inspection cycle. Since the inspection strategy in this embodiment is a dynamically updated inspection strategy based on fault feature information, the inspection strategy storage module 1012 can also dynamically update the stored inspection strategy and interact with the inspection task management module 1011 to update the inspection strategy.

[0064] The fault mode learner 1013 receives fault feature information sent by the fault feature collector 1014 and generates an inspection strategy based on the fault feature information. Specifically, the fault mode learner 1013 can perform similarity matching between the fault feature information and historical fault feature information in the historical fault mode library, thereby selecting the inspection strategy corresponding to the historical fault feature information with a similarity higher than the similarity threshold as the inspection strategy for the input fault feature information. When no historical fault feature information with a similarity higher than the similarity threshold is matched in the historical fault mode library, the fault mode learner 1013 can also generate an inspection strategy based on the input fault feature information and add the fault feature information and the corresponding inspection strategy to the historical fault mode library.

[0065] The fault mode learner 1013 is also used to dynamically adjust the inspection strategy. For example, if the inspection subsystem 101 does not receive a fault signal for a long time, the fault mode learner 1013 can increase the inspection cycle in the inspection strategy while keeping the inspection content unchanged. As another example, if the inspection subsystem 101 receives more than a certain number of fault signals within a certain period, the fault mode learner 1013 can decrease the inspection cycle in the inspection strategy and increase the inspection content.

[0066] The fault feature collector 1014 is used to collect fault feature information from the operation and maintenance subsystem 102. Specifically, the fault feature collector 1014 obtains fault feature information based on the fault context corresponding to the fault signal of the operation and maintenance subsystem 102. The fault feature information includes fault features in multiple dimensions, such as monitoring information, log information, and topology information.

[0067] The fault feature collector 1014 is also used to convert the collected fault feature information into a multi-dimensional fault feature vector and send the fault feature vector to the fault pattern learner 1013. Each dimension of the fault feature vector corresponds to a fault feature. When the fault pattern learner 1013 performs similarity matching, it also calculates the similarity based on the fault feature vector converted by the fault feature collector 1014.

[0068] The operation and maintenance subsystem 102 is used to monitor and manage the operating status and performance indicators of the operation and maintenance object 103. Specifically, the operation and maintenance subsystem 102 can receive the operation and maintenance data reported by the operation and maintenance object 103, and perform fault judgment based on the operation and maintenance data. When the operation and maintenance object 103 has a fault signal, the operation and maintenance subsystem 102 can also report the fault characteristic information corresponding to the fault signal to the inspection subsystem.

[0069] The operation and maintenance subsystem 102 includes multiple functional modules, such as a data access and cleaning module 1021, a data storage module 1022, an alarm module 1023, a monitoring module 1024, a log module 1025, and a topology module 1026.

[0070] The data access and cleaning module 1021 is used to receive the operation and maintenance data reported by the operation and maintenance object 103 and perform deduplication and completion processing on the operation and maintenance data. The data storage module 1022 is used to store the operation and maintenance data processed by the data access and cleaning module 1021.

[0071] The alarm module 1023 is used to generate fault signals based on operation and maintenance data. The fault signals can be divided into different fault levels, such as attention, serious and fatal. In this embodiment, the fault signals are also called alarm signals. The fault signals can be signals generated during the inspection process. There is a mapping relationship between the fault signals and the fault feature information.

[0072] The monitoring module 1024 is used to monitor the operating status of the maintenance object 103 based on maintenance data and generate monitoring indicators, which can serve as fault characteristic information. The logging module 1025 is used to generate various types of log information based on maintenance data, which can also serve as fault characteristic information. The topology module 1026 is used to generate network topology relationships between different maintenance objects 103, including horizontal and vertical topologies between maintenance objects 103.

[0073] Operation and maintenance object 103 is the specific object monitored and managed by operation and maintenance subsystem 102. Operation and maintenance object 103 can also be called inspection object. It can receive and execute inspection tasks issued by inspection subsystem 101 or operation and maintenance subsystem 102. Operation and maintenance object 103 includes various types of hardware and software resources; for example, it can be a server, switch, host machine, virtual machine, cloud disk, and database, etc., without specific limitations.

[0074] It should be noted that the inspection subsystem 101, operation and maintenance subsystem 102 and operation and maintenance object 103 in the above-mentioned inspection system 10 can all be deployed on computing devices or computing device clusters. Therefore, in this embodiment, computing devices or computing device clusters can also be used to refer to the various modules in the inspection system 10.

[0075] based on Figure 1 The inspection system 10 shown in this application also provides an inspection method based on fault characteristics. The following describes the inspection method based on fault characteristics provided by this application in conjunction with embodiments.

[0076] Please see Figure 2 , Figure 2 This is a flowchart illustrating a fault characteristic-based inspection method provided in an embodiment of this application. Figure 2 In the example shown, the method includes the following steps:

[0077] 201. The inspection system performs the first inspection task, collecting fault characteristic information corresponding to the fault that occurred. The fault characteristic information is used to indicate the fault context corresponding to the fault signal. The fault signal includes the alarm input received by the inspection system.

[0078] In this embodiment, after the inspection subsystem 101 sends the first inspection task according to the first inspection strategy, the operation and maintenance system 103 begins to execute the first inspection task. Since the operation and maintenance system 103 is executing the first inspection task, the first inspection strategy corresponding to the first inspection task can also be called the current inspection strategy. At the same time, the operation and maintenance object 103 can report operation and maintenance data to the operation and maintenance subsystem 102, and the operation and maintenance subsystem 102 monitors the operating status of the operation and maintenance object 103 based on the reported operation and maintenance data.

[0079] When the operation and maintenance subsystem 102 determines that the operation and maintenance object 103 has failed, the operation and maintenance subsystem 102 generates a fault signal and the corresponding fault feature information. The fault may be a fault detected during the execution of the first inspection task. At this time, the inspection subsystem 101 can collect the fault feature information. That is, the operation and maintenance subsystem 102 reports the fault signal and the corresponding fault feature information to the subsystem 101. The fault feature information is used to indicate the fault context corresponding to the fault signal. The fault context includes, for example, the fault background corresponding to the fault signal, the fault description, and the fault association record.

[0080] In this embodiment, the fault feature information includes one or more dimensions of fault features, and different indicators can be used for different dimensions of fault features. For example, when the inspection content is to detect whether the application interface is capable of providing interface call services to the outside world, the fault feature information includes multiple dimensions of fault features such as interface call information and topology information. The indicators corresponding to the interface call information dimension include the number of interface calls and the call success rate.

[0081] In one possible implementation, the fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. Monitoring information indicates the operating status of the inspected object; log information indicates the operating history of the inspected object; and topology information indicates the topological relationships of the inspected object, including network topology or system topology. For example, when the maintenance object 103 is a server, monitoring information may include the server's CPU utilization and memory utilization, log information may include the server's system logs, and topology information may include the connection relationships or logical relationships between the server and other objects such as databases and application systems.

[0082] The fault signals in this embodiment can be classified into different fault levels, such as attention, serious, and fatal. Since the maintenance subsystem 102 will send an alarm to the inspection subsystem 101 after determining that the maintenance object 103 has failed, the fault signal can also be called an alarm signal.

[0083] It should be noted that the fault signal in this embodiment can be a fault determined by the maintenance object 103 performing the first inspection task, or a fault determined by the maintenance subsystem 102 based on maintenance data; there is no specific limitation. Correspondingly, the fault feature information in this embodiment can be the fault feature information collected when the inspection result of the first inspection task fails, or the fault feature information collected when a fault report is input; there is no specific limitation.

[0084] Please see Figure 3 , Figure 3 This is a schematic diagram of another inspection method based on fault characteristics provided in an embodiment of this application. Figure 3 In step 1 of the example shown, the fault feature collector of the inspection subsystem 101 collects fault feature information from the operation and maintenance subsystem 102. The fault feature information includes the context information corresponding to the fault signal.

[0085] For example, in Figure 3 In the example shown, the inspection task sent by the inspection subsystem 101 to the maintenance object 103 is to detect whether the interface of the application APP1 in the inspection object 103 is capable of providing interface call services to the outside world. When the inspection result is that the interface of the application APP1 is faulty, the maintenance subsystem 102 sends the fault signal and the corresponding fault feature information to the fault feature collector 1014. The fault feature information includes multiple fault features. For example, the fault feature information sent by the maintenance subsystem 10 includes 4 fault features, of which fault feature 1 is the interface call information of APP1, fault feature 2 is the alarm information of APP1, fault feature 3 is the horizontal topology information and vertical topology information of APP1, and fault feature 4 is the abnormal log fragment of APP1.

[0086] 202. The inspection system retrieves inspection strategies from the historical fault mode database based on fault feature information to determine the first target inspection strategy. The historical fault mode database includes historical fault features and the corresponding inspection strategies.

[0087] After collecting fault feature information, the inspection subsystem 103 determines a first target inspection strategy based on the fault feature information. Specifically, the inspection subsystem 103 searches for inspection strategies in a historical fault pattern database based on the fault feature information to determine the first target inspection strategy. The historical fault pattern database includes historical fault features and corresponding inspection strategies. The first target inspection strategy is the inspection strategy corresponding to the target fault features. The target fault features include historical fault features in the historical fault patterns whose similarity to the fault feature information is higher than a similarity threshold. Historical fault feature information in the historical fault pattern database with a similarity higher than the similarity threshold can be referred to as "hit fault feature information" in the historical fault pattern database. The similarity threshold can be, for example, 0.8, 0.85, 0.9, or 0.95.

[0088] In one possible implementation, during the process of the inspection subsystem 101 retrieving inspection strategies from the historical fault mode database based on fault feature information, the inspection subsystem 101 performs similarity matching between the fault feature information and one or more historical fault features in the historical fault mode database to determine the target fault feature. Specifically, the inspection subsystem 101 identifies fault features in the historical fault mode database with a similarity higher than a similarity threshold as target fault features. The inspection subsystem 101 then determines the inspection strategy corresponding to the target fault feature as a first target inspection strategy, which includes inspection content and inspection cycle.

[0089] Understandably, when there are multiple historical fault feature information in the historical fault mode library that have a similarity to the fault feature information that is higher than the similarity threshold, the inspection subsystem 101 can determine the target fault feature from the historical fault feature information with the highest similarity.

[0090] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating similarity matching based on fault feature information, provided as an embodiment of this application. Figure 4 In the example shown, after the inspection subsystem 101 collects fault feature information, it performs similarity matching based on the fault feature information and historical fault features in the historical fault mode library. The historical fault features in the historical fault mode library have a mapping relationship with the inspection strategy. Through similarity matching, the inspection subsystem 101 can identify historical fault features with similarity higher than the similarity threshold as target fault features, and determine the inspection strategy corresponding to the target fault features as the first target inspection strategy based on the mapping relationship.

[0091] For example, in Figure 4 In the example shown, the inspection subsystem 101 collects fault feature information including four fault features, namely fault feature 1 to fault feature 4. The historical fault pattern library contains multiple historical fault feature information and corresponding inspection strategies. There is a mapping relationship between the historical fault feature information and the inspection strategies. For example, historical fault features 1 to historical fault features 4 correspond to inspection strategy 1, and historical fault features 1, 3 and 5 correspond to inspection strategy 2. The inspection subsystem 101 performs similarity matching based on the fault feature information. That is, the similarity between the fault feature information and historical fault features 1 to historical fault features 4 in the historical fault pattern library is 100%. Therefore, historical fault features 1 to historical fault features 4 are determined as target fault features, and the inspection strategy 1 corresponding to historical fault features 1 to historical fault features 4 is the first target inspection strategy.

[0092] In one possible implementation, during the process of similarity matching based on fault feature information and historical fault feature information in the historical fault pattern library, the fault feature collector 1014 of the inspection subsystem 101 generates a first fault feature vector based on the fault feature information. The first fault feature vector includes multiple dimensions, and different dimensions correspond to different fault features.

[0093] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating a method for calculating similarity based on fault feature vectors, provided as an embodiment of this application. Figure 5 In the example shown, after the inspection subsystem 101 collects fault feature information, it converts the fault feature information into a first fault feature vector. The first fault feature vector is a multi-dimensional vector, with each dimension corresponding to a fault feature. For example, the first fault feature vector generated by the inspection subsystem 101 based on the fault feature information contains fault features in four dimensions: fault feature 1 is the interface call information of APP1, fault feature 2 is the alarm information of APP1, fault feature 3 is the horizontal and vertical topology information of APP1, and fault feature 4 is the abnormal log fragment of APP1.

[0094] exist Figure 5 In the example shown, the inspection subsystem 101 generates a second fault feature vector based on historical fault feature information from the historical fault mode library. For example, the second fault feature vector may contain fault features in four dimensions, namely historical fault feature 1 to historical fault feature 4, or the second fault feature vector may contain fault features in three dimensions, namely historical fault feature 1, historical fault feature 3 and historical fault feature 5.

[0095] In one possible implementation, after the fault feature collector 1014 of the inspection subsystem 101 generates a first fault feature vector based on fault feature information, the fault mode learner 1013 of the inspection subsystem 101 calculates the similarity between the first fault feature vector and the second fault feature vector. The second fault feature vector includes one or more fault feature vectors from a historical fault mode library, and the similarity is used by the inspection system for similarity matching. After calculating the similarity between one or more fault feature vectors in the historical fault mode library, the inspection subsystem 101 determines the fault feature vectors with similarity higher than a similarity threshold as target fault features, and determines the inspection strategy corresponding to the target fault features as the first target inspection strategy. For example, the inspection subsystem 101 determines the fault feature vector with the highest similarity among the fault feature vectors with similarity higher than the similarity threshold as the target fault feature.

[0096] In this embodiment, the fault mode learner 1013 calculates the cosine similarity between the first fault feature vector and the second fault feature vector, and the calculation result satisfies the following formula:

[0097]

[0098] Where cos(θ) is the cosine similarity, x 1k Let x be the first fault feature vector. 2k For the second fault feature vector, the closer cos(θ) is to 1, the higher the similarity; the closer cos(θ) is to -1, the lower the similarity.

[0099] Please continue reading. Figure 3 ,exist Figure 3 In steps 2 to 4 of the example shown, after the fault feature acquisition module 1014 acquires fault feature information, it generates a first fault feature vector based on the fault feature information and sends the first fault feature vector to the fault mode learner 1013. The fault mode learner 1013 receives the first fault feature vector and performs similarity matching based on the first fault feature vector and one or more fault feature information from the historical fault mode library, wherein the fault feature information from the historical fault mode library can be represented as a second fault feature vector. The fault mode learner 1013 performs similarity matching based on the fault feature vector.

[0100] exist Figure 3 In steps 2 to 4 of the example shown, the fault mode learner 1013 calculates cosine similarity based on the first fault feature vector and one or more second fault feature vectors. If there is a second fault feature vector with a cosine similarity greater than the similarity threshold, the fault mode learner 1013 determines the fault feature corresponding to the second fault feature vector as the target fault feature and determines the inspection strategy in the historical fault mode library corresponding to the target fault feature as the first target inspection strategy.

[0101] Please continue reading. Figure 5 ,exist Figure 5 In the example shown, the fault mode learner 1013 of the inspection subsystem 101 calculates the cosine similarity between the first fault feature vector and the second fault feature vector, wherein the first fault feature vector... = (APP1's interface call information, APP1's alarm information, APP1's horizontal and vertical topology information, APP1's abnormal log fragments), the second fault feature vector includes one or more historical fault feature vectors from the historical fault mode library. The inspection subsystem 101 is based on the first fault feature vector. The cosine similarity is calculated between the second fault feature vector and the historical fault feature vector with the largest cosine similarity is determined as the target fault feature, and the inspection strategy corresponding to the historical fault feature vector with the largest cosine similarity is determined as the first target inspection strategy.

[0102] In one possible implementation, when the number of fault signals received by the inspection subsystem 101 within a first time threshold exceeds a certain threshold, the first target inspection strategy is adjusted. Adjusting the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy, or increasing the inspection content in the first target inspection strategy. For example, if the inspection subsystem 101 receives more than 3 fault signals within 3 days, the inspection subsystem 101 reduces the inspection cycle of the first target inspection strategy or increases the inspection content of the first target inspection strategy. Reducing the inspection cycle in the first target inspection strategy, for example, involves the inspection subsystem 101 adjusting the inspection cycle from once a day to once every 8 hours.

[0103] It should be noted that during the process of reducing the inspection cycle in the first target inspection strategy, the proportion of the reduction in the inspection cycle is positively correlated with the number of fault signals received within the first time threshold. That is, the more fault signals the inspection subsystem 101 receives within the first time threshold, the shorter the inspection cycle will be.

[0104] In this embodiment of the application, when the inspection subsystem 101 is in a faulty state or when the business is in a sub-healthy state for a long time, it can more quickly discover potential risks in the system by reasonably adjusting the inspection cycle and inspection content. At the same time, based on the system's own resource capacity, it can stagger inspection tasks to improve inspection accuracy and make reasonable use of system resources.

[0105] In one possible implementation, adding inspection content to the first target inspection strategy includes unifying the set of inspection items corresponding to fault feature information with the set of inspection items corresponding to the target fault features. The set of inspection items corresponding to fault feature information includes newly added inspection items obtained based on the analysis of fault feature information. That is, the inspection subsystem 101 can analyze and add new inspection items based on fault feature information and merge the new inspection items with the inspection items in the current inspection strategy.

[0106] Please see Figure 6 , Figure 6 This is an example diagram illustrating an embodiment of the present application that adds inspection content to a first target inspection strategy. Figure 6 In the example shown, if the inspection subsystem 101 determines that the fault feature information corresponding to the fault signal within the first time threshold hits the fault mode library multiple times, then the inspection subsystem 101 adjusts the inspection strategy, that is, adds inspection items to the inspection content in the inspection strategy. The inspection subsystem 101 determines the set of inspection items corresponding to the fault feature information based on the fault feature information, and performs a union operation on the set of inspection items with the set of inspection items in the current inspection strategy, thereby obtaining the inspection strategy after adding inspection content.

[0107] For example, in Figure 6 In the example shown, the inspection subsystem 101 determines the set of inspection items corresponding to the target fault feature based on the historical fault mode library. This set includes inspection items 1 to 3, which are respectively APP interface inspection, interface call indicator inspection, and alarm inspection. This set of inspection items corresponding to the target fault feature is the inspection item in the current inspection strategy. The inspection subsystem 101 determines the set of inspection items corresponding to the fault feature information, including inspection items 1, 4, and 5. These items are respectively APP interface inspection, topology information inspection, and abnormal log fragment inspection. The combined set of inspection items by the inspection subsystem 101 includes inspection items 1 to 5, which are respectively APP interface inspection, interface call indicator inspection, alarm inspection, topology information inspection, and abnormal log fragment inspection. This combined set of inspection items is the inspection content of the adjusted target inspection strategy.

[0108] In one possible implementation, when no fault feature information is matched in the historical fault mode library, that is, the similarity between the fault feature information and the fault feature information in the historical fault mode library is less than the similarity threshold, the inspection subsystem 101 generates a second target inspection strategy based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information. The inspection cycle corresponding to the fault feature information in the second target inspection strategy can be a default inspection cycle, for example, the default inspection cycle is 1 day.

[0109] It is understandable that, in the process of generating the second target inspection strategy based on fault feature information, the inspection subsystem 101 can analyze the inspection items corresponding to the fault feature information based on expert knowledge, or it can analyze the inspection items corresponding to the fault feature information based on an artificial intelligence model. The artificial intelligence model is an artificial intelligence model trained based on the training data of the fault feature information and the inspection strategy, and there is no specific limitation.

[0110] In one possible implementation, after the inspection subsystem 101 generates a second target inspection strategy based on the fault feature information, it adds the second target inspection strategy and the fault features corresponding to the second target inspection strategy to the historical fault mode library. The newly added fault feature information and the second target inspection strategy corresponding to the fault feature information can be used as historical fault features for similarity matching.

[0111] In one possible implementation, if the inspection subsystem 101 does not receive a fault signal within a second time threshold, the inspection subsystem 101 adjusts the first target inspection strategy. Adjusting the first target inspection strategy includes increasing the inspection cycle within the first target inspection strategy. For example, if the inspection subsystem 101 does not receive a fault signal for 3 days, the inspection subsystem 101 increases the inspection cycle from 1 day to 3 days, while the inspection content remains unchanged.

[0112] It should be noted that during the process of increasing the inspection cycle in the first target inspection strategy, the increase in the inspection cycle of the inspection subsystem 101 is positively correlated with the time during which no fault signal was received. That is, the longer the inspection subsystem 101 does not receive a fault signal, the greater the increase in the inspection cycle. In addition, the increased inspection cycle cannot exceed the inspection cycle threshold, for example, the inspection cycle threshold is 7 days.

[0113] In this embodiment of the application, the inspection subsystem 101 can reduce the consumption of computing, storage, and network resources by the inspection tasks in the inspection system by reasonably increasing the inspection cycle (reducing the inspection frequency) when the business is in a healthy state under non-fault conditions. This can reasonably reduce the pressure on the entire inspection system.

[0114] 203. The inspection system adjusts the first inspection strategy based on the first target inspection strategy. The adjusted first inspection strategy is used to execute the second inspection task. The second inspection task and the first inspection task are separated by one or more inspection cycles.

[0115] After the inspection subsystem 101 generates the first target inspection strategy, it adjusts the first inspection strategy based on the first target inspection strategy. The adjusted first inspection strategy is used to execute the second inspection task. At this time, the adjusted first inspection strategy is the first target inspection strategy. The second inspection task and the first inspection task are separated by one or more inspection cycles. Specifically, the fault mode learning 1013 sends the first target inspection strategy to the inspection task management module 1011. The inspection task management module 1011 generates the second inspection task based on the first target inspection strategy and sends the second inspection task to the maintenance object 103.

[0116] In one possible implementation, after the inspection task management module 1011 generates the second inspection task, the operation and maintenance subsystem 102 can also send the second inspection task to the operation and maintenance object 103. The specific implementation is not limited.

[0117] Please continue reading. Figure 3 ,exist Figure 3 In steps 5 to 6 of the example shown, after the fault mode learning 1013 generates the first target inspection strategy, it sends the first target inspection strategy to the inspection task management module 1011. The inspection task management module 1011 generates inspection tasks based on the first target inspection strategy and sends the inspection tasks to the operation and maintenance subsystem 102. The operation and maintenance subsystem 102 then sends the inspection tasks to the operation and maintenance object 103.

[0118] It is understandable that when the fault feature information does not find the first target inspection strategy in the historical fault mode database, the inspection subsystem 101 can also adjust the first inspection strategy based on the second target inspection strategy after generating the second target inspection strategy based on the fault feature information, without any specific limitations.

[0119] As can be seen from the above embodiments, in the embodiments of this application, the computing device can search for the target inspection strategy from the historical fault mode library based on the fault feature information corresponding to the fault signal, and adjust the current inspection strategy based on the target inspection strategy. That is, the computing device can dynamically adjust the inspection strategy based on the fault features, and can selectively select the inspection content in the target inspection strategy based on the fault feature information, thereby improving the inspection accuracy and efficiency of fault inspection.

[0120] Based on the above method embodiments, this application also provides a fault feature-based inspection device. The fault feature-based inspection device provided by this application embodiment is described in detail below.

[0121] Please see Figure 7 , Figure 7 This is a schematic diagram of a fault-characteristic-based inspection device provided in an embodiment of this application. Figure 7In the example shown, the fault feature-based inspection device 700 is used to implement the various steps performed by the fault feature-based inspection system in the above embodiments. The fault feature-based inspection device 700 includes a data acquisition unit 701 and a processing unit 702.

[0122] The processing unit 702 executes a first inspection task, and the acquisition unit 701 acquires fault feature information corresponding to the occurring fault. The fault feature information indicates the fault context corresponding to the fault signal, which includes alarm inputs received by the inspection system. The processing unit 702 performs an inspection strategy retrieval in a historical fault pattern library based on the fault feature information to determine a first target inspection strategy. The historical fault pattern library includes historical fault features and corresponding inspection strategies. The first target inspection strategy is the inspection strategy corresponding to the target fault features, which include historical fault features in the historical fault patterns with a similarity higher than a similarity threshold to the fault feature information. The processing unit 702 also adjusts the first inspection strategy based on the first target inspection strategy. The adjusted first inspection strategy is used to execute a second inspection task, which is spaced one or more inspection cycles apart from the first task.

[0123] In one possible implementation, the processing unit 702 is further configured to perform similarity matching based on fault feature information and one or more historical fault features in a historical fault pattern library to determine a target fault feature. The inspection strategy corresponding to the target fault feature is determined as a first target inspection strategy, which includes inspection content and inspection cycle.

[0124] In one possible implementation, the processing unit 702 is further configured to generate a first fault feature vector based on fault feature information. The first fault feature vector includes multiple dimensions, with different dimensions corresponding to different fault features. A similarity is calculated between the first fault feature vector and a second fault feature vector. The second fault feature vector includes one or more fault feature vectors from a historical fault mode library. The similarity is used by the inspection system for similarity matching.

[0125] In one possible implementation, the processing unit 702 is further configured to adjust the first target inspection strategy when the number of fault signals received by the inspection system within a first time threshold exceeds the number threshold. The adjustment of the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy and increasing the inspection content in the first target inspection strategy.

[0126] In one possible implementation, the processing unit 702 is specifically used to perform a union processing on the set of inspection items corresponding to the fault feature information and the set of inspection items corresponding to the target fault feature. The set of inspection items corresponding to the fault feature information includes newly added inspection items obtained based on the analysis of the fault feature information.

[0127] In one possible implementation, when no fault feature information is matched in the historical fault mode library, the processing unit 702 is further configured to generate a second target inspection strategy based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information.

[0128] In one possible implementation, the processing unit 702 is further configured to add the second target inspection strategy and the fault features corresponding to the second target inspection strategy to the historical fault mode library.

[0129] In one possible implementation, when the inspection system does not receive a fault signal within a second time threshold, the processing unit 702 is specifically used to adjust the first target inspection strategy, which includes increasing the inspection cycle in the first target inspection strategy.

[0130] In one possible implementation, the fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. The monitoring information is used to indicate the operating status of the inspected object, the log information is used to indicate the operating history of the inspected object, and the topology information is used to indicate the topological relationship of the inspected object.

[0131] It is understandable that the acquisition unit 701 and processing unit 702 in the fault characteristic-based inspection device 700 can serve as functional modules. Figure 1 The various modules in the inspection system 10 are mapped to each other, thereby realizing the functions of each module in the inspection system 10.

[0132] It should be understood that the division of units in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, all units in the device can be implemented entirely through software calls from processing elements; all units can be implemented entirely in hardware; or some units can be implemented through software calls from processing elements, and others in hardware. For example, each unit can be a separate processing element, or it can be integrated into a chip within the device. Alternatively, it can be stored as a program in memory, called and executed by a processing element of the device. Moreover, these units can be fully or partially integrated together, or implemented independently. The processing element mentioned here can also be called a processor, which can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above units can be implemented through integrated logic circuits in the processor element or through software calls from processing elements.

[0133] It is worth noting that, for the sake of simplicity, the above method embodiments are described as a series of actions. However, those skilled in the art should know that this application is not limited to the order of the described actions. Furthermore, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by this application.

[0134] Other reasonable combinations of steps that can be conceived by those skilled in the art based on the above description also fall within the scope of protection of this application. Furthermore, those skilled in the art should also be aware that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to this application.

[0135] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Figure 8 As shown, the computing device 800 includes a processor 801, a memory 802, a communication interface 803, and a bus 804. The processor 801, memory 802, and communication interface 803 are coupled via the bus (not shown in the figure). The memory 802 stores instructions. When the instructions in the memory 802 are executed, the computing device 800 executes the method performed by the inspection system in the above method embodiment.

[0136] The computing device 800 may be one or more integrated circuits configured to implement the methods described above, such as: one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these forms of integrated circuits. Furthermore, when the units in the device can be implemented in the form of a processing element scheduler, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these units may be integrated together to implement a system-on-a-chip (SOC).

[0137] The processor 801 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0138] The memory 802 can be volatile memory or non-volatile memory, or it can include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0139] The memory 802 stores executable program code, and the processor 801 executes the executable program code to implement the functions of the aforementioned units or modules, thereby realizing the above-mentioned fault characteristic-based inspection method. That is, the memory 802 stores instructions for executing the above-mentioned fault characteristic-based inspection method.

[0140] The communication interface 803 uses transceiver modules, such as, but not limited to, network interface cards and transceivers, to enable communication between the computing device 800 and other devices or communication networks.

[0141] In addition to the data bus, the 804 bus can also include a power bus, a control bus, and a status signal bus. The bus can be a Peripheral Component Interconnect Express (PCIe) bus, an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. The bus can be divided into address bus, data bus, and control bus.

[0142] Please see Figure 9 , Figure 9 This is a schematic diagram of a computing device cluster provided in an embodiment of this application. Figure 9 As shown, the computing device cluster 900 includes at least one computing device 800.

[0143] like Figure 9 As shown, the computing device cluster 900 includes at least one computing device 800. The memory 802 of one or more computing devices 800 in the computing device cluster 900 may store the same instructions for executing the aforementioned fault-characteristic-based inspection method.

[0144] In some possible implementations, the memory 802 of one or more computing devices 800 in the computing device cluster 900 may also store partial instructions for executing the aforementioned fault characteristic-based inspection method. In other words, a combination of one or more computing devices 800 can jointly execute the instructions for executing the aforementioned fault characteristic-based inspection method.

[0145] It should be noted that the memories 802 in the different computing devices 800 within the computing device cluster 900 can store different instructions, which are used to execute certain functions of the aforementioned node load control device. That is, the instructions stored in the memories 802 of the different computing devices 800 can implement the functions of one or more modules in the acquisition unit and processing unit.

[0146] In some possible implementations, one or more computing devices 800 in the computing device cluster 900 can be connected via a network. This network can be a wide area network (WAN) or a local area network (LAN), etc.

[0147] Please see Figure 10 , Figure 10 This is a schematic diagram illustrating the network connection of computer devices in a computer cluster, as provided in an embodiment of this application. Figure 10 As shown, the two computing devices 800A and 800B are connected via a network. Specifically, they are connected to the network through the communication interfaces in each computing device.

[0148] In one possible implementation, the memory in computing device 800A stores instructions for executing the acquisition unit function. Meanwhile, the memory in computing device 800B stores instructions for executing the processing unit function.

[0149] It should be understood that Figure 10 The functions of computing device 800A shown can also be performed by multiple computing devices. Similarly, the functions of computing device 800B can also be performed by multiple computing devices.

[0150] In another embodiment of this application, a computer-readable storage medium is also provided, which stores computer-executable instructions. When the processor of the device executes the computer-executable instructions, the device executes the method performed by the inspection system in the above method embodiment.

[0151] In another embodiment of this application, a computer program product is also provided, which includes computer-executable instructions stored in a computer-readable storage medium. When the processor of the device executes the computer-executable instructions, the device performs the method executed by the inspection system in the above-described method embodiment.

[0152] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A fault characteristic-based inspection method, characterized in that, include: The first inspection task is performed to collect fault feature information corresponding to the fault that occurred. The fault feature information is used to indicate the fault context corresponding to the fault signal. The fault signal includes the alarm input received by the inspection system. Based on the fault feature information, an inspection strategy is retrieved in the historical fault pattern library to determine a first target inspection strategy. The historical fault pattern library includes historical fault features and inspection strategies corresponding to the historical fault features. The first target inspection strategy is the inspection strategy corresponding to the target fault features. The target fault features include historical fault features in the historical fault patterns whose similarity to the fault feature information is higher than a similarity threshold. The first inspection strategy is adjusted based on the first target inspection strategy. The adjusted first inspection strategy is used to execute a second inspection task, which is spaced one or more inspection cycles apart from the first inspection task.

2. The method according to claim 1, characterized in that, The step of retrieving inspection strategies from the historical fault mode database based on the fault feature information includes: The target fault feature is determined by performing similarity matching between the fault feature information and one or more historical fault features in the historical fault pattern library. The inspection strategy corresponding to the target fault characteristics is determined as the first target inspection strategy, which includes inspection content and inspection cycle.

3. The method according to claim 1 or 2, characterized in that, The method further includes: A first fault feature vector is generated based on the fault feature information. The first fault feature vector includes multiple dimensions, and different dimensions correspond to different fault features. The similarity between the first fault feature vector and the second fault feature vector is calculated. The second fault feature vector includes one or more fault feature vectors in the historical fault mode library. The similarity is used by the inspection system for similarity matching.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: If the number of fault signals received by the inspection system within the first time threshold exceeds the number threshold, the first target inspection strategy is adjusted. The adjustment of the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy, and increasing the inspection content in the first target inspection strategy.

5. The method according to claim 4, characterized in that, The addition of inspection content to the first target inspection strategy includes: The set of inspection items corresponding to the fault feature information is combined with the set of inspection items corresponding to the target fault feature. The set of inspection items corresponding to the fault feature information includes newly added inspection items obtained based on the analysis of the fault feature information.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: If the fault feature information is not matched in the historical fault mode library, a second target inspection strategy is generated based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information.

7. The method according to claim 6, characterized in that, After generating the second target inspection strategy based on the fault feature information, the method further includes: Add the second target inspection strategy and the corresponding fault features to the historical fault mode library.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes: If the inspection system does not receive the fault signal within the second time threshold, the first target inspection strategy is adjusted, and the adjustment of the first target inspection strategy includes increasing the inspection cycle in the first target inspection strategy.

9. The method according to any one of claims 1 to 6, characterized in that, The fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. The monitoring information is used to indicate the operating status of the inspected object, the log information is used to indicate the operating history of the inspected object, and the topology information is used to indicate the topological relationship of the inspected object.

10. A fault characteristic-based inspection device, characterized in that, include: The processing unit is used to execute the first inspection task; The acquisition unit is used to acquire fault feature information corresponding to the fault that occurred. The fault feature information is used to indicate the fault context corresponding to the fault signal. The fault signal includes the alarm input received by the inspection system. The processing unit is used to perform inspection strategy retrieval in the historical fault pattern library based on the fault feature information, and determine a first target inspection strategy. The historical fault pattern library includes historical fault features and inspection strategies corresponding to the historical fault features. The first target inspection strategy is the inspection strategy corresponding to the target fault features. The target fault features include historical fault features in the historical fault patterns whose similarity to the fault feature information is higher than a similarity threshold. The processing unit is further configured to adjust the first inspection strategy based on the first target inspection strategy, and the adjusted first inspection strategy is used to execute a second inspection task, wherein the second inspection task is spaced one or more inspection cycles apart from the first inspection task.

11. The apparatus according to claim 10, characterized in that, The processing unit is also used for: The target fault feature is determined by performing similarity matching between the fault feature information and one or more historical fault features in the historical fault pattern library. The inspection strategy corresponding to the target fault characteristics is determined as the first target inspection strategy, which includes inspection content and inspection cycle.

12. The apparatus according to claim 10 or 11, characterized in that, The processing unit is also used for: A first fault feature vector is generated based on the fault feature information. The first fault feature vector includes multiple dimensions, and different dimensions correspond to different fault features. The similarity between the first fault feature vector and the second fault feature vector is calculated. The second fault feature vector includes one or more fault feature vectors in the historical fault mode library. The similarity is used by the inspection system for similarity matching.

13. The apparatus according to any one of claims 10 to 12, characterized in that, The processing unit is also used for: If the number of fault signals received by the inspection system within the first time threshold exceeds the number threshold, the first target inspection strategy is adjusted. The adjustment of the first target inspection strategy includes one or more of the following: reducing the inspection cycle in the first target inspection strategy, and increasing the inspection content in the first target inspection strategy.

14. The apparatus according to claim 13, characterized in that, The processing unit is specifically used for: The set of inspection items corresponding to the fault feature information is combined with the set of inspection items corresponding to the target fault feature. The set of inspection items corresponding to the fault feature information includes newly added inspection items obtained based on the analysis of the fault feature information.

15. The apparatus according to any one of claims 10 to 14, characterized in that, The processing unit is also used for: If the fault feature information is not matched in the historical fault mode library, a second target inspection strategy is generated based on the fault feature information. The inspection content in the second target inspection strategy includes the inspection items corresponding to the fault feature information.

16. The apparatus according to claim 15, characterized in that, The processing unit is also used for: Add the second target inspection strategy and the corresponding fault features to the historical fault mode library.

17. The apparatus according to any one of claims 10 to 16, characterized in that, The processing unit is specifically used for: If the inspection system does not receive the fault signal within the second time threshold, the first target inspection strategy is adjusted, and the adjustment of the first target inspection strategy includes increasing the inspection cycle in the first target inspection strategy.

18. The apparatus according to any one of claims 10 to 17, characterized in that, The fault characteristic information includes one or more of the following: monitoring information, log information, and topology information. The monitoring information is used to indicate the operating status of the inspected object, the log information is used to indicate the operating history of the inspected object, and the topology information is used to indicate the topological relationship of the inspected object.

19. A computing device cluster, characterized in that, The device includes at least one computing device, the computing device including a processor coupled to a memory, the processor being used to store instructions that, when executed by the processor, cause the cluster of computing devices to perform the method of any one of claims 1 to 9.

20. A computer-readable storage medium having instructions stored thereon, characterized in that, When the instructions are executed, they cause the computer to perform the method of any one of claims 1 to 9.

21. A computer program product, the computer program product comprising instructions, characterized in that, When the instructions are executed, they cause the computer to perform the method of any one of claims 1 to 9.

Images