Hardware equipment maintenance method, device and electronic equipment

Abstract

The embodiment of the present application provides a method, device and electronic device for hardware equipment maintenance, the method comprising obtaining status information, wherein the status information is information corresponding to each hardware device in the server to be maintained, inputting the status information into a trained multi-level SVM model, so that the multi-level SVM model performs multiple identification and classification processing on the status information, obtains a fault information set containing at least one fault type, and performs fault maintenance on each hardware device in the server to be maintained according to the fault information set containing at least one fault type. The present application improves the accuracy and efficiency of fault type identification, thereby ensuring the normal operation of the server.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of equipment maintenance, and in particular, to a hardware equipment maintenance method, device and electronic equipment. Background Art

[0002] With the development of computer technology, more and more technologies are applied in the financial field. The traditional financial industry is gradually transforming to financial technology (Fintech), and equipment maintenance technology is no exception. However, due to the security and real-time requirements of the financial industry, higher requirements are also placed on equipment maintenance technology.

[0003] In the prior art, in order to meet the growing needs of various financial services, the number of server applications is increasing, and the types of hardware devices used in the server are also increasing. During the operation of each hardware device in the server, failures may occur due to reasons such as long operation time or model mismatch. In order to troubleshoot as quickly as possible, the acquired hardware device failure information can be identified through the SVM (Support Vector Machine) model to obtain the failure types corresponding to different hardware devices, and then maintenance is performed according to the failure types corresponding to different hardware devices to ensure the normal operation of the server.

[0004] However, the existing SVM model is limited to binary classification scenarios, that is, it can only identify situations where there are two or one fault type. Due to the wide variety of hardware devices, the corresponding number of fault types is also relatively large, resulting in poor versatility of the SVM model, which reduces the accuracy and efficiency of fault type identification, and thus affects the normal operation of the server. Summary of the invention

[0005] The embodiments of the present application provide a hardware equipment maintenance method, apparatus and electronic equipment to improve the accuracy and efficiency of fault type identification.

[0006] In a first aspect, an embodiment of the present application provides a hardware device maintenance method, including:

[0007] Acquire status information, wherein the status information is information corresponding to each hardware device in the server to be maintained;

[0008] Inputting the state information into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type;

[0009] Fault maintenance is performed on each hardware device in the server to be maintained according to the fault information set including at least one fault type.

[0010] Optionally, the state information is input into a trained multi-level SVM model so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type, including:

[0011] Inputting the state information into a trained multi-level SVM model, so that a first-level SVM module in the multi-level SVM model identifies and classifies the state information to obtain first fault type information and other fault type information;

[0012] If the other fault type information contains more than one type of fault type, the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification processing to obtain the second fault type information and new other fault type information;

[0013] If the new other fault type information still contains more than one type of fault type, then re-execute the step of inputting the other fault type information into the second-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps until the new other fault type information contains only one type of fault type, and obtain a fault information set containing at least one fault type.

[0014] Optionally, after obtaining the fault information set including at least one fault type, the method further includes:

[0015] Obtaining a non-fault information set according to the state information and the fault information set including at least one fault type;

[0016] The non-fault information set is input into the multi-level SVM model, so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set including at least one risk level.

[0017] Optionally, the non-fault information set is input into the multi-level SVM model so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set containing at least one risk level, including:

[0018] Inputting the non-fault information set into the multi-level SVM model, so that the third-level SVM module in the multi-level SVM model identifies and classifies each non-fault information in the non-fault information set to obtain first-level information and other-level information;

[0019] If the number of risk levels contained in the other level information is greater than one, the other level information is input into the fourth level SVM module in the multi-level SVM model for identification and classification processing to obtain the second level information and new other level information;

[0020] If the number of risk levels contained in the new other level information is still greater than one, the step of inputting the other level information into the fourth-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps is re-executed until the number of risk levels contained in the new other level information is one, thereby obtaining a level information set containing at least one risk level.

[0021] Optionally, before obtaining the status information, the method further includes:

[0022] Acquire a state training information set, wherein the state training information set includes multiple groups of state training information, each group of the state training information includes state information of a hardware device, and a target fault type corresponding to the state information of the hardware device;

[0023] The initial multi-level SVM model is trained according to the status information of the hardware devices contained in each group of the status training information and the target fault type corresponding to the status information of the hardware devices to obtain a trained multi-level SVM model.

[0024] Optionally, after obtaining the trained multi-level SVM model, the method further includes:

[0025] Get status test information set;

[0026] Inputting the state test information set into the trained multi-level SVM model for multiple identification and classification processes to obtain a fault test information set;

[0027] The accuracy of the multi-level SVM model is evaluated according to the fault test information set, and when the result of the accuracy evaluation meets the preset evaluation condition, a multi-level SVM model test pass prompt is generated.

[0028] Optionally, the performing accuracy evaluation on the multi-level SVM model according to the fault test information set includes:

[0029] According to the expression: Accuracy = Avg (Acc i ) determining the classification accuracy of the multi-level SVM model;

[0030] Among them, Acc i represents the classification accuracy of each level of SVM module in the multi-level SVM model,

[0031]

[0032] T pl represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, F p represents the number of samples that do not belong to the target fault type but are mistakenly classified as the target fault type, Fnn represents the number of samples that belong to the target fault type but are mistakenly classified as other fault types, T n1 represents the number of samples that do not belong to the target fault type and are correctly classified, α i Represents the accurate factor coefficient of each fault type, α i =1-X*X T , X represents the matrix corresponding to the state test information in the state test information set, X T represents the transpose of X, i represents the number of fault types;

[0033] and / or,

[0034] According to the expression: Determining the recall of the multi-level SVM model;

[0035] Wherein, F represents the recall rate of the multi-level SVM model, β represents the ratio of the target samples corresponding to the target fault type in the samples classified into the target fault type, β = max{α1, α2, ...α i},

[0036] Optionally, when the result of the accuracy evaluation meets the preset evaluation condition, generating a prompt indicating that the multi-level SVM model test has passed includes:

[0037] When the classification accuracy of the multi-level SVM model exceeds a preset accuracy threshold, and / or the recall rate of the multi-level SVM model exceeds a preset recall rate threshold, a multi-level SVM model test pass prompt is generated.

[0038] In a second aspect, an embodiment of the present application provides a hardware device maintenance device, including:

[0039] An acquisition module, used to acquire status information, wherein the status information is information corresponding to each hardware device in the server to be maintained;

[0040] A processing module, used for inputting the state information into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type;

[0041] The processing module is further configured to perform fault maintenance on each hardware device in the server to be maintained according to the fault information set including at least one fault type.

[0042] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0043] The memory stores computer-executable instructions;

[0044] The processor executes the computer-executable instructions stored in the memory to implement the hardware device maintenance method as described in the first aspect and various possible designs of the first aspect.

[0045] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the hardware device maintenance method described in the first aspect and various possible designs of the first aspect is implemented.

[0046] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the hardware device maintenance method described in the first aspect and various possible designs of the first aspect.

[0047] The embodiments of the present application provide a hardware equipment maintenance method, apparatus and electronic device. After adopting the above scheme, the status information corresponding to each hardware device in the server to be maintained can be obtained first, and then the status information can be input into the trained multi-level SVM model for identification and classification processing to obtain a fault information set containing at least one fault type. Then, fault maintenance is performed on each hardware device in the server to be maintained based on the fault information set containing at least one fault type. The pre-trained multi-level SVM model is used to realize the identification of multiple fault types, thereby improving the versatility of the SVM model, and improving the accuracy and efficiency of fault type identification, thereby ensuring the normal operation of the server. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0049] Figure 1 A schematic diagram of the architecture of an application system for the hardware device maintenance method provided in an embodiment of the present application;

[0050] Figure 2 A flowchart of a hardware device maintenance method provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of the application of the multi-level SVM model recognition process provided in the embodiment of the present application;

[0052] Figure 4 A schematic diagram of the structure of a hardware device maintenance device provided in an embodiment of the present application;

[0053] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can also include other sequential instances in addition to those illustrated or described. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0056] The number of server applications is increasing. In order to better meet the needs of financial services, the types of hardware devices used in servers are also increasing. Exemplarily, the types of hardware devices used in servers may include memory, hard disk, processor, fan, etc. And each hardware device in the server may fail during operation due to reasons such as too long running time or model mismatch. In order to troubleshoot as soon as possible, the acquired hardware device fault information can be identified through the SVM (Support Vector Machine) model to obtain the fault type corresponding to different hardware devices, and then perform maintenance according to the fault type corresponding to different hardware devices to ensure the normal operation of the server. However, the existing SVM model is limited to the binary classification scenario, that is, it can only identify the situation where the fault type is two or one. Due to the wide variety of hardware devices, the number of corresponding fault types is also relatively large, resulting in poor versatility of the SVM model, reducing the accuracy and efficiency of fault type identification, and thus affecting the normal operation of the server.

[0057] Based on the above technical problems, the present application realizes the identification of various fault types through a pre-trained multi-level SVM model, which not only improves the versatility of the SVM model, but also improves the accuracy and efficiency of fault type identification, thereby ensuring the normal operation of the server.

[0058] Figure 1 A schematic diagram of the architecture of the application system of the hardware device maintenance method provided in the embodiment of the present application, such as Figure 1 As shown, in the application system, it can include: server 101, server to be maintained 102. Server to be maintained 102 can be a running device or a non-running device, and different hardware devices are deployed in server to be maintained 102, which can be hard disk, fan, processor, etc., for example, and different hardware devices can correspond to different status information. Server 101 can obtain status information corresponding to each hardware device from server to be maintained 102, and input the status information corresponding to different hardware devices into pre-trained multi-level SVM model 103 for identification and classification processing, so as to obtain a fault information set containing at least one fault type, and then generate a fault maintenance request from the fault information set and send it to the corresponding server to be maintained 102.

[0059] In addition, the server 101 can also serve as a server to be maintained to send status information corresponding to each hardware device to other servers. The server 101 can also directly obtain status information corresponding to different hardware devices in each server to be maintained from the database.

[0060] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0061] Figure 2 1 is a flow chart of a hardware device maintenance method provided in an embodiment of the present application. The method in this embodiment can be executed by the server 101. Figure 2 As shown, the method of this embodiment may include:

[0062] S201: Acquire status information, where the status information is information corresponding to each hardware device in the server to be maintained.

[0063] In this embodiment, in order to ensure the normal operation of each financial service, the status of each hardware device in the server running the financial service can be monitored to determine whether each hardware device is operating normally. In addition, when it is determined that a hardware device fails, in order to improve the efficiency of fault maintenance, the faults can be classified first, and then further fault maintenance can be performed according to the determined categories.

[0064] The status information may be information corresponding to each hardware device in a running server to be maintained, or may be information corresponding to each hardware device in a non-running server to be maintained.

[0065] Furthermore, the status information may include hardware brand, model, running time, abnormal alarm log, etc.

[0066] S202: Inputting the state information into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple recognition and classification processes on the state information to obtain a fault information set containing at least one fault type.

[0067] In this embodiment, after obtaining the status information, the status information can be input into a pre-trained multi-level SVM model that can identify multiple fault types. The multi-level SVM model can perform multiple identification and classification processes on the status information, and finally obtain a fault information set containing at least one fault type.

[0068] Furthermore, the state information is input into the trained multi-level SVM model, so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type, which may specifically include:

[0069] The state information is input into the trained multi-level SVM model, so that the first-level SVM module in the multi-level SVM model identifies and classifies the state information to obtain first fault type information and other fault type information.

[0070] If the other fault type information contains more than one type of fault type, the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification to obtain second fault type information and new other fault type information.

[0071] If the new other fault type information still contains more than one type of fault type, then re-execute the step of inputting the other fault type information into the second-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps until the new other fault type information contains only one type of fault type, and obtain a fault information set containing at least one fault type.

[0072] Specifically, the multi-level SVM model includes different SVM modules, and the more SVM modules there are, the more fault types the multi-level SVM model can identify. Exemplarily, if the multi-level SVM model includes two SVM modules, the multi-level SVM model can identify three fault types.

[0073] Correspondingly, after the state information is input into the trained multi-level SVM model, the first-level SVM module in the multi-level SVM model can first identify and classify the state information to obtain the first fault type information and other fault type information. Then, it is determined whether the fault type types contained in the other fault type information are greater than one. If the fault type types contained in the other fault type information are less than or equal to one (i.e., one or zero), the identification is completed and the subsequent steps are no longer executed. If the fault type types contained in the other fault type information are greater than one, the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification, and the second fault type information and new other fault type information are obtained. Then, it is determined whether the fault type types contained in the new other fault type information are greater than one. If the fault type types contained in the new other fault type information are equal to one, the identification is completed and the subsequent steps are no longer executed. If the new other fault type information contains more than one type of fault type, the new other fault type information is used as other fault type information, and the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps until the new other fault type information contains only one type of fault type, thereby obtaining a fault information set containing at least one fault type.

[0074] Exemplarily, Table 1 is a fault information table. As shown in Table 1, Table 1 includes four different types of fault information tables.

[0075] Table 1 Fault information table

[0076]

[0077] The advantage of the SVM model for binary classification is used to classify the status information containing fault information. Each time one of the fault types is distinguished, it can stop. The remaining unclassified fault types are then classified using the SVM model, and the cycle is repeated until all fault types are distinguished. This can detect fault types faster and increase the number of proactive server maintenance times, thereby reducing maintenance costs.

[0078] S203: Perform fault maintenance on each hardware device in the server to be maintained according to a fault information set containing at least one fault type.

[0079] In this embodiment, after obtaining a fault information set including at least one fault type, in order to ensure the normal operation of each financial service, maintenance can be performed as soon as possible for each fault type in the fault information set.

[0080] Furthermore, a corresponding fault maintenance request may be generated according to each fault type in the fault information set, and the fault maintenance request may be sent to the corresponding server to be maintained, so that the server to be maintained may automatically complete the repair of the fault according to the fault maintenance request. For a server to be maintained that cannot automatically complete the repair of the fault, the server to be maintained may send a maintenance request to the terminal device of the corresponding operation and maintenance personnel. The server may generate a corresponding fault maintenance request according to each fault type in the fault information set, and send the fault maintenance request to the terminal device of the corresponding operation and maintenance personnel.

[0081] After adopting the above scheme, the status information corresponding to each hardware device in the server to be maintained can be obtained first, and then the status information can be input into the trained multi-level SVM model for identification and classification processing to obtain a fault information set containing at least one fault type, and then fault maintenance can be performed on each hardware device in the server to be maintained based on the fault information set containing at least one fault type. The pre-trained multi-level SVM model can be used to recognize multiple fault types, thereby improving the versatility of the SVM model, and improving the accuracy and efficiency of fault type recognition, thereby ensuring the normal operation of the server.

[0082] based on Figure 2 The method, the examples of this specification also provide some specific implementation plans of the method, which are described below.

[0083] In addition, in another embodiment, after obtaining a fault information set including at least one fault type, the method may further include:

[0084] A non-fault information set is obtained according to the state information and the fault information set including at least one fault type.

[0085] The non-fault information set is input into the multi-level SVM model, so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set including at least one risk level.

[0086] In this embodiment, the status information corresponding to each hardware device in the server to be maintained may include not only the status information corresponding to the faulty hardware device, but also the status information corresponding to the normally operating hardware device. By classifying the status information corresponding to the normally operating hardware device according to the risk level, the hardware devices that may fail can be predicted in advance, thereby further reducing the frequency of possible server failures.

[0087] Specifically, information related to the fault information set in the status information can be eliminated to obtain a non-fault information set, and then the non-fault information set is input into a multi-level SVM model for identification and classification to obtain a level information set containing at least one risk level.

[0088] In addition, the multi-level SVM model includes different SVM modules, and the number of levels that can be divided is positively correlated with the number of SVM modules, that is, the more SVM modules there are, the more levels that can be divided, and the more risk levels included in the level information set.

[0089] Further, the non-fault information set is input into the multi-level SVM model, so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set containing at least one risk level, which may specifically include:

[0090] The non-fault information set is input into the multi-level SVM model, so that the third-level SVM module in the multi-level SVM model identifies and classifies each non-fault information in the non-fault information set to obtain first-level information and other-level information.

[0091] If the number of risk levels contained in the other level information is greater than one, the other level information is input into the fourth level SVM module in the multi-level SVM model for identification and classification processing to obtain the second level information and new other level information;

[0092] If the number of risk levels contained in the new other level information is still greater than one, the step of inputting the other level information into the fourth-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps is re-executed until the number of risk levels contained in the new other level information is one, thereby obtaining a level information set containing at least one risk level.

[0093] Specifically, after the non-fault information set is input into the trained multi-level SVM model, the third-level SVM module in the multi-level SVM model can first identify and classify the non-fault information set to obtain the first-level information and other-level information. Then, it is determined whether the number of risk levels contained in the other-level information is greater than one. If the number of risk levels contained in the other-level information is less than or equal to one (i.e., one or zero), the identification is completed and the subsequent steps are not executed. If the number of risk levels contained in the other-level information is greater than one, the other-level information is input into the fourth-level SVM module in the multi-level SVM model for identification and classification, and the second-level information and new other-level information are obtained. Then, it is determined whether the number of risk levels contained in the new other-level information is greater than one. If the number of risk levels contained in the new other-level information is equal to one, the identification is completed and the subsequent steps are not executed. If the number of risk levels contained in the new other level information is greater than one, the new other level information will be used as other level information, and the other level information will be input into the fourth-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps will be re-executed until the number of risk levels contained in the new other level information is one, thereby obtaining a level information set containing at least one risk level.

[0094] For example, Figure 3 The following is a schematic diagram of the application of the multi-level SVM model recognition process provided in the embodiment of the present application. Figure 3As shown, in this embodiment, there are three types of fault types and three risk levels. Then, the state information can be obtained first, and then the state information can be input into the first-level SVM module for identification and classification to obtain the first fault type information and other fault type information. Then, it is determined whether the fault type contained in the other fault type information is greater than one. If it is greater than one, the other fault type information is input into the second-level SVM module for identification and classification to obtain the second fault type information and new other fault type information, and the new other fault type information only contains one fault type. Then, a fault information set can be obtained based on the first fault type information, the second fault type information and the new other fault type information. Then, a non-fault information set can be obtained based on the state information and the fault information set. Then, the non-fault information set can be input into the third-level SVM module for identification and classification to obtain the first level information and other level information. Then, it is determined whether the number of risk levels contained in the other level information is greater than one. If it is greater than one, the other level information is input into the fourth-level SVM module for identification and classification to obtain the second level information and new other level information, and the number of risk levels contained in the new other level information is one. Then, a level information set is obtained according to the first level information, the second level information and the new other level information.

[0095] In addition, in another embodiment, before acquiring the status information, the method may further include:

[0096] A state training information set is obtained, wherein the state training information set includes multiple groups of state training information, each group of the state training information includes state information of a hardware device, and a target fault type corresponding to the state information of the hardware device.

[0097] The initial multi-level SVM model is trained according to the status information of the hardware devices contained in each group of the status training information and the target fault type corresponding to the status information of the hardware devices to obtain a trained multi-level SVM model.

[0098] In this embodiment, before applying the multi-level SVM model to identify the state information and determine the fault information set, the initial multi-level SVM model can be trained with the state training information set to obtain a trained multi-level SVM model, and then the trained multi-level SVM model is applied to identify and classify the newly acquired state information, and finally a fault information set containing at least one fault type is obtained.

[0099] The state training information set may include multiple sets of state training information, each set of state training information may include the state information of the hardware device and the target fault type corresponding to the state information of the hardware device. Correspondingly, the process of determining the state training information set may specifically include: collecting historical fault data D from server fault work orders within a preset time period d Then, the data set D can be cleaned according to the preset data cleaning rules. d Perform data cleaning and integration processing to remove incomplete and invalid data. It should be noted that data cleaning here can retain duplicate data. Then, according to the goal to be achieved, reduce the data set D d The unnecessary feature dimensions in the , get the aggregated rule set D d Then, the aggregated rule set D can be mined according to the preset fault feature d Perform feature mining and classify the state training information set M k ={M1, M2, M3…M n}, where n represents the type of fault. For example, M1, M2, and M3 can represent three types of faults. The mined features can be fault time, faulty components, abnormal alarm information, etc.

[0100] In addition, in order to further reduce the frequency of server failures, each set of state training information in the state training information set may include, in addition to the state information of the failed hardware device and the target failure type corresponding to the state information of the failed hardware device, the state information of the non-failed hardware device and the risk level corresponding to the state information of the non-failed hardware device. The initial multi-level SVM model can be trained with the state training information set to obtain a trained multi-level SVM model, and then the trained multi-level SVM model is applied to identify and classify the newly acquired state information, and finally a failure information set containing at least one failure type is obtained, and then the state information is further identified and classified to obtain a level information set containing at least one risk level.

[0101] In addition, after obtaining the trained multi-level SVM model, the method may further include:

[0102] Get the status test information set.

[0103] The state test information set is input into the trained multi-level SVM model for multiple identification and classification processes to obtain a fault test information set.

[0104] The accuracy of the multi-level SVM model is evaluated according to the fault test information set, and when the result of the accuracy evaluation meets the preset evaluation condition, a multi-level SVM model test pass prompt is generated.

[0105] In this embodiment, after obtaining the trained multi-level SVM model, in order to improve the accuracy of subsequent applications of the multi-level SVM model, the trained multi-level SVM model can be tested using a state test information set, and after the test passes, the multi-level SVM model can be applied to identify and classify the acquired state information.

[0106] Further, performing accuracy evaluation on the multi-level SVM model according to the fault test information set may specifically include:

[0107] According to the expression:

[0108] Accuracy=Avg(Acc i ) determine the classification accuracy of the multi-level SVM model,

[0109] Among them, Acc i represents the classification accuracy of each level of SVM module in the multi-level SVM model,

[0110] T p1 represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, F pp represents the number of samples that do not belong to the target fault type but are mistakenly classified as the target fault type, Fnn represents the number of samples that belong to the target fault type but are mistakenly classified as other fault types, T n1 represents the number of samples that do not belong to the target fault type and are correctly classified, α represents the accurate factor coefficient of each fault type, and α i =1-X*X T , X represents the matrix corresponding to the state test information in the state test information set, X T represents the transpose of X, and i represents the number of fault types.

[0111] and / or according to the expression: The recall of the multi-class SVM model is determined.

[0112] Wherein, F represents the recall rate of the multi-level SVM model, β represents the ratio of the target samples corresponding to the target fault type in the samples classified into the target fault type, β = max{α1, α2, ...α i},

[0113] Specifically, α i It can represent the accurate factor coefficients of various fault types, and can dynamically change according to the corresponding features in the iteration process according to the frequent item sets of different features. For example, there are three fault types. At the beginning of model training, each fault type αi Set an initial value, which is 1 by default. During the network training process, as the amount of data continues to increase, the weight contribution of the data sample in each iteration is also changing (the value gradually decreases). When the network model area is stable, α i The value is basically stable, that is, the final coefficient α can be obtained i , the specific calculation formula can be: i =1-X*X T X represents the matrix corresponding to the status test information in the status test information set, but it needs to be converted into a matrix format for machine recognition. However, in actual consideration, it is not a single feature, and each feature will contain more. The article only classifies the set to facilitate understanding of the actual considerations of business classification, brand, type, faulty component, subsystem type, etc.

[0114] For example, taking Table 1 as an example, T p1 It represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, for example, the number of samples that belong to type A1 and are also classified into type A1. n1 Indicates the number of samples that do not belong to the target fault type and are correctly classified, for example: A4 is classified as A2 regardless of the brand. pp Indicates the number of samples that do not belong to the target fault type but are mistakenly classified as such, for example: A2 does not belong to A1 but is classified as A1. nn Indicates the number of samples that belong to the target fault type but are mistakenly classified as other classes, for example, samples that belong to A4 are classified as A1.

[0115] In addition, you can also use the expression: Determine the recall of the multi-class SVM model.

[0116] Wherein, F represents the recall rate of the multi-level SVM model, β represents the ratio of the target samples corresponding to the target fault type in the samples classified into the target fault type, β = max{α1, α2, ...α i};

[0117] Specifically, F is a model measurement indicator that takes into account both precision and recall. For example, the multi-level SVM model can identify three types of faults and three risk levels. By repeatedly training the multi-level SVM model, a trained multi-level SVM model can be obtained. Table 2 is a test result table of the multi-level SVM model provided in the embodiment of the present application. As shown in Table 2, after obtaining the trained multi-level SVM model, the trained multi-level SVM model can be tested using a test information set to output the corresponding six-category results, where M k(can be M1, M2, M3) indicates the type of fault, D r (It can be D1, D2, D3) indicates the risk level.

[0118] Table 2 Multi-level SVM model test results

[0119]

[0120] In addition, when the result of the accuracy evaluation meets the preset evaluation conditions, a prompt indicating that the multi-level SVM model test has passed is generated, which may specifically include:

[0121] When the classification accuracy of the multi-level SVM model exceeds a preset accuracy threshold, and / or the recall rate of the multi-level SVM model exceeds a preset recall rate threshold, a multi-level SVM model test pass prompt is generated.

[0122] Specifically, the multi-level SVM model test can be determined to have passed when either the classification accuracy of the multi-level SVM model or the recall rate of the multi-level SVM model meets the conditions, or the multi-level SVM model test can be determined to have passed when both the classification accuracy of the multi-level SVM model and the recall rate of the multi-level SVM model meet the conditions. Among them, the accuracy threshold and the recall threshold can be customized according to the actual application scenario, and no detailed restrictions are given here.

[0123] In addition, when determining F, in order to improve the accuracy of the determined F, M can be determined separately. k and D r Determine F, then M k and D r The average value of the corresponding F is determined as the final F.

[0124] Exemplarily, Table 3 is a comparison table of algorithm classification accuracy results provided in an embodiment of the present application. As shown in Table 3, the multi-level SVM model can be tested, and the classification performance of the multi-level SVM model can be compared with that of the existing algorithms at this stage. It can be clearly seen that the Accuracy value obtained according to the multi-level SVM model is improved to a certain extent compared with the traditional Decision tree, Randomforest, and SVM algorithms.

[0125] Table 3 Algorithm classification accuracy results comparison table

[0126]

[0127]

[0128] Table 4 is a comparison table of the recall rate results of the algorithms provided in this application. As shown in Table 4, the F determined by the multi-level SVM model is 0.83, which is comparable to the performance of the traditional algorithm. In addition, the multi-level SVM model has a better classification effect on various hardware devices of the server under certain conditions.

[0129] Table 4 Algorithm recall rate comparison table

[0130]

[0131] Through active management of server hardware devices, server hardware resource management can be more refined, facilitating early detection of problems so that corresponding decisions can be made, improving server reliability, and improving server performance at the bottom level to ensure server stability.

[0132] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method. Figure 4 A schematic diagram of the structure of the hardware device maintenance device provided in the embodiment of the present application is shown in FIG. Figure 4 As shown, the device provided in this embodiment may include:

[0133] The acquisition module 401 is used to acquire status information, wherein the status information is information corresponding to each hardware device in the server to be maintained.

[0134] The processing module 402 is used to input the state information into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type.

[0135] In this embodiment, the processing module 402 is further used for:

[0136] The state information is input into a trained multi-level SVM model, so that the first-level SVM module in the multi-level SVM model identifies and classifies the state information to obtain first fault type information and other fault type information.

[0137] If the other fault type information contains more than one type of fault type, the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification to obtain second fault type information and new other fault type information.

[0138] If the new other fault type information still contains more than one type of fault type, then re-execute the step of inputting the other fault type information into the second-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps until the new other fault type information contains only one type of fault type, and obtain a fault information set containing at least one fault type.

[0139] Furthermore, the processing module 402 is further configured to:

[0140] A non-fault information set is obtained according to the state information and the fault information set including at least one fault type.

[0141] The non-fault information set is input into the multi-level SVM model, so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set including at least one risk level.

[0142] The processing module 402 is further configured to:

[0143] The non-fault information set is input into the multi-level SVM model, so that the third-level SVM module in the multi-level SVM model identifies and classifies each non-fault information in the non-fault information set to obtain first-level information and other-level information.

[0144] If the number of risk levels contained in the other level information is greater than one, the other level information is input into the fourth-level SVM module in the multi-level SVM model for identification and classification to obtain the second-level information and new other level information.

[0145] If the number of risk levels contained in the new other level information is still greater than one, the step of inputting the other level information into the fourth-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps is re-executed until the number of risk levels contained in the new other level information is one, thereby obtaining a level information set containing at least one risk level.

[0146] The processing module 402 is further configured to perform fault maintenance on each hardware device in the server to be maintained according to the fault information set including at least one fault type.

[0147] In addition, in another embodiment, the processing module 402 is further configured to:

[0148] A state training information set is obtained, wherein the state training information set includes multiple groups of state training information, each group of the state training information includes state information of a hardware device, and a target fault type corresponding to the state information of the hardware device.

[0149] The initial multi-level SVM model is trained according to the status information of the hardware devices contained in each group of the status training information and the target fault type corresponding to the status information of the hardware devices to obtain a trained multi-level SVM model.

[0150] In addition, the processing module 402 is further used for:

[0151] Get the status test information set.

[0152] The state test information set is input into the trained multi-level SVM model for multiple identification and classification processes to obtain a fault test information set.

[0153] The accuracy of the multi-level SVM model is evaluated according to the fault test information set, and when the result of the accuracy evaluation meets the preset evaluation condition, a multi-level SVM model test pass prompt is generated.

[0154] Specifically, the processing module 402 is further used for:

[0155] According to the expression: Accuracy = Avg (Acc i ) determine the classification accuracy of the multi-level SVM model,

[0156] Among them, Acc i represents the classification accuracy of each level of SVM module in the multi-level SVM model,

[0157] T p1 represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, F pp represents the number of samples that do not belong to the target fault type but are mistakenly classified as the target fault type, Fnn represents the number of samples that belong to the target fault type but are mistakenly classified as other fault types, T n1 represents the number of samples that do not belong to the target fault type and are correctly classified; α i Represents the accurate factor coefficient of each fault type, α i =1-X*X T , X represents the matrix corresponding to the state test information in the state test information set, XT represents the transpose of X, and i represents the number of fault types.

[0158] and / or,

[0159] According to the expression: The recall of the multi-class SVM model is determined.

[0160] Wherein, F represents the recall rate of the multi-level SVM model, β represents the ratio of the target samples corresponding to the target fault type in the samples classified into the target fault type, β = max{α1, α2, ...α n},

[0161] In addition, the processing module 402 is further used for:

[0162] When the classification accuracy of the multi-level SVM model exceeds a preset accuracy threshold, and / or the recall rate of the multi-level SVM model exceeds a preset recall rate threshold, a multi-level SVM model test pass prompt is generated.

[0163] The device provided in the embodiment of the present application can realize the above-mentioned Figure 2 The implementation principle and technical effect of the method in the embodiment shown are similar and will not be described in detail here.

[0164] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, the device 500 provided in this embodiment includes: a processor 501 and a memory connected to the processor in communication. The processor 501 and the memory 502 are connected via a bus 503 .

[0165] In the specific implementation process, the processor 501 executes the computer execution instructions stored in the memory 502, so that the processor 501 executes the hardware device maintenance method in the above method embodiment.

[0166] The specific implementation process of the processor 501 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0167] In the above Figure 5 In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0168] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0169] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0170] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When a processor executes the computer-executable instructions, the hardware device maintenance method of the above method embodiment is implemented.

[0171] An embodiment of the present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the hardware device maintenance method as described above is implemented.

[0172] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0173] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0174] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for maintaining hardware equipment, characterized in that: include: Acquire status information, wherein the status information is information corresponding to each hardware device in the server to be maintained; The state information is input into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple identification and classification processing on the state information to obtain a fault information set containing at least one fault type; the multi-level SVM model performs an accuracy evaluation on the trained multi-level SVM model through the state test information set, and the result of the accuracy evaluation is obtained when it meets the preset evaluation conditions, and the accuracy evaluation includes determining the classification accuracy Accuracy of the trained multi-level SVM model, wherein Accuracy = Avg(Acc i ), Acc i represents the classification accuracy of each level of SVM module in the multi-level SVM model, T p1 represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, F pp represents the number of samples that do not belong to the target fault type but are mistakenly classified as the target fault type, Fnn represents the number of samples that belong to the target fault type but are mistakenly classified as other fault types, T n1 represents the number of samples that do not belong to the target fault type and are correctly classified, α i Represents the accurate factor coefficient of each fault type, α i =1-X*X T , X represents the matrix corresponding to the state test information in the state test information set, X T represents the transpose of X, i represents the number of fault types; Fault maintenance is performed on each hardware device in the server to be maintained according to the fault information set including at least one fault type.

2. The method according to claim 1, characterized in that: The state information is input into the trained multi-level SVM model, so that the multi-level SVM model performs multiple identification and classification processes on the state information to obtain a fault information set containing at least one fault type, including: Inputting the state information into a trained multi-level SVM model, so that a first-level SVM module in the multi-level SVM model identifies and classifies the state information to obtain first fault type information and other fault type information; If the other fault type information contains more than one type of fault type, the other fault type information is input into the second-level SVM module in the multi-level SVM model for identification and classification processing to obtain the second fault type information and new other fault type information; If the new other fault type information still contains more than one type of fault type, then re-execute the step of inputting the other fault type information into the second-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps until the new other fault type information contains only one type of fault type, and obtain a fault information set containing at least one fault type.

3. The method according to claim 2, characterized in that After obtaining the fault information set including at least one fault type, the method further includes: Obtaining a non-fault information set according to the state information and the fault information set including at least one fault type; The non-fault information set is input into the multi-level SVM model, so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set including at least one risk level.

4. The method according to claim 3, characterized in that The non-fault information set is input into the multi-level SVM model so that the multi-level SVM model identifies and classifies the non-fault information set to obtain a level information set containing at least one risk level, including: Inputting the non-fault information set into the multi-level SVM model, so that the third-level SVM module in the multi-level SVM model identifies and classifies each non-fault information in the non-fault information set to obtain first-level information and other-level information; If the number of risk levels contained in the other level information is greater than one, the other level information is input into the fourth level SVM module in the multi-level SVM model for identification and classification processing to obtain the second level information and new other level information; If the number of risk levels contained in the new other level information is still greater than one, the step of inputting the other level information into the fourth-level SVM module in the multi-level SVM model for identification and classification processing and subsequent steps is re-executed until the number of risk levels contained in the new other level information is one, thereby obtaining a level information set containing at least one risk level.

5. The method according to any one of claims 1 to 4, characterized in that: Before obtaining the status information, the method further includes: Acquire a state training information set, wherein the state training information set includes multiple groups of state training information, each group of the state training information includes state information of a hardware device, and a target fault type corresponding to the state information of the hardware device; The initial multi-level SVM model is trained according to the status information of the hardware devices contained in each group of the status training information and the target fault type corresponding to the status information of the hardware devices to obtain a trained multi-level SVM model.

6. The method according to claim 5, characterized in that After the trained multi-level SVM model is obtained, the method further includes: Get status test information set; Inputting the state test information set into the trained multi-level SVM model for multiple identification and classification processes to obtain a fault test information set; The accuracy of the multi-level SVM model is evaluated according to the fault test information set, and when the result of the accuracy evaluation meets the preset evaluation condition, a multi-level SVM model test pass prompt is generated.

7. The method according to claim 6, characterized in that The performing accuracy evaluation on the multi-level SVM model according to the fault test information set further comprises: According to the expression: Determining the recall rate of the multi-level SVM model; Wherein, F represents the recall rate of the multi-level SVM model, β represents the ratio of the target samples corresponding to the target fault type in the samples classified into the target fault type, β = max{α1, α2, ...α i }, 8. The method according to claim 7, characterized in that When the result of the accuracy evaluation meets the preset evaluation conditions, a multi-level SVM model test passing prompt is generated, including: When the classification accuracy of the multi-level SVM model exceeds a preset accuracy threshold, and / or the recall rate of the multi-level SVM model exceeds a preset recall rate threshold, a multi-level SVM model test pass prompt is generated.

9. A hardware equipment maintenance device, characterized in that: include: An acquisition module, used to acquire status information, wherein the status information is information corresponding to each hardware device in the server to be maintained; A processing module is used to input the state information into a trained multi-level support vector machine (SVM) model, so that the multi-level SVM model performs multiple identification and classification processing on the state information to obtain a fault information set containing at least one fault type; the multi-level SVM model performs an accuracy evaluation on the trained multi-level SVM model through the state test information set, and the result of the accuracy evaluation is obtained when it meets the preset evaluation conditions, and the accuracy evaluation includes determining the classification accuracy Accuracy of the trained multi-level SVM model, wherein Accuracy=Avg(Acc i ), Acc i represents the classification accuracy of each level of SVM module in the multi-level SVM model, T p1 represents the number of samples that belong to the target fault type and are correctly classified into the target fault type, F pp represents the number of samples that do not belong to the target fault type but are mistakenly classified as the target fault type, Fnn represents the number of samples that belong to the target fault type but are mistakenly classified as other fault types, T n1 represents the number of samples that do not belong to the target fault type and are correctly classified, α i Represents the accurate factor coefficient of each fault type, α i =1-X*X T , X represents the matrix corresponding to the state test information in the state test information set, X T represents the transpose of X, i represents the number of fault types; The processing module is further configured to perform fault maintenance on each hardware device in the server to be maintained according to the fault information set including at least one fault type.

10. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the hardware device maintenance method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the processor executes the computer-executable instructions, the hardware device maintenance method according to any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the hardware device maintenance method according to any one of claims 1 to 8 is implemented.

Images