WO2022134353A1 - 硬件状态检测方法、装置、计算机设备及存储介质 - Google Patents

硬件状态检测方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2022134353A1
WO2022134353A1 PCT/CN2021/083760 CN2021083760W WO2022134353A1 WO 2022134353 A1 WO2022134353 A1 WO 2022134353A1 CN 2021083760 W CN2021083760 W CN 2021083760W WO 2022134353 A1 WO2022134353 A1 WO 2022134353A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
error reporting
hardware
historical
state detection
Prior art date
Application number
PCT/CN2021/083760
Other languages
English (en)
French (fr)
Inventor
高南海
邵敏
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2022134353A1 publication Critical patent/WO2022134353A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C29/44—Indication or identification of errors, e.g. for repair
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2455—Query execution
    • G06F16/24564—Applying rules; Deductive queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
    • G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
    • G11C2029/4402—Internal storage of test result, quality data, chip identification, repair information

Definitions

  • the present application relates to the technical field of hardware management, and belongs to the application scenario of intelligently detecting the status of hardware equipment in a smart city, and in particular, to a hardware status detection method, device, computer equipment and storage medium.
  • Embodiments of the present application provide a hardware state detection method, device, computer device, and storage medium, which aim to solve the problem that the state of a magnetic tape and other hardware devices cannot be efficiently detected in the prior art methods.
  • an embodiment of the present application provides a hardware state detection method, which includes:
  • the historical error reporting information contained in the pre-stored historical error reporting information table is matched to obtain the error reporting level of each of the historical error reporting information;
  • the abnormal hardware information is obtained from the hardware detection information according to the state level of the hardware device and fed back to the administrator of the management server.
  • an embodiment of the present application provides a hardware state detection device, which includes:
  • An error reporting level obtaining unit configured to match the historical error reporting information contained in the pre-stored historical error reporting information table according to a preset error reporting level matching rule to obtain the error reporting level of each of the historical error reporting information;
  • an error reporting quantification information acquisition unit configured to quantify each of the historical error reporting information in the historical error reporting information table according to a preset information quantification rule to obtain error reporting quantification information
  • a state detection model generation unit configured to generate a corresponding state detection model according to the error reporting level matching rule and the information quantification rule
  • a model training unit configured to iteratively train the state detection model according to preset model training rules and the error reporting quantification information and error reporting level of each of the historical error reporting information, to obtain a trained state detection model
  • a hardware detection information acquisition unit configured to perform periodic detection on hardware devices included in each of the backup terminals according to a preset detection period, to acquire hardware detection information of each of the hardware devices;
  • a state level acquisition unit configured to input the hardware detection information acquired at the current detection time point into the state detection model to acquire the state level corresponding to each of the hardware devices;
  • the abnormal hardware information feedback unit is configured to obtain abnormal hardware information from the hardware detection information according to the state level of the hardware device and feed it back to the administrator of the management server.
  • an embodiment of the present application further provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer
  • the hardware state detection method described in the first aspect above is implemented in a program.
  • an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when executed by a processor, the computer program causes the processor to execute the above-mentioned first step.
  • Embodiments of the present application provide a hardware state detection method, apparatus, and computer-readable storage medium.
  • Obtain the error reporting level of each historical error reporting information in the historical error reporting information table quantify each historical error reporting information to obtain quantified error reporting information, build a status detection model, and iteratively train the status detection model according to the quantified error reporting information and the corresponding error reporting level.
  • Obtain the hardware detection information obtained by periodic detection of the hardware devices included in the backup terminal into the state detection model obtain the state level corresponding to each hardware device, and obtain abnormal hardware information for feedback according to the state level.
  • the state level of each hardware device is obtained based on the hardware detection information, and abnormal hardware information is obtained to give an early warning to the abnormal hardware device, and the state of the magnetic tape and other hardware devices can be detected accurately and efficiently.
  • FIG. 1 is a schematic flowchart of a hardware state detection method provided by an embodiment of the present application.
  • FIG. 2 is a schematic diagram of an application scenario of a hardware state detection method provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of a sub-flow of a hardware state detection method provided by an embodiment of the present application.
  • FIG. 4 is a schematic diagram of another sub-flow of the hardware state detection method provided by the embodiment of the present application.
  • FIG. 5 is a schematic diagram of another sub-flow of the hardware state detection method provided by the embodiment of the present application.
  • FIG. 6 is a schematic diagram of another sub-flow of the hardware state detection method provided by the embodiment of the present application.
  • FIG. 7 is a schematic diagram of another sub-flow of the hardware state detection method provided by the embodiment of the present application.
  • FIG. 8 is another schematic flowchart of a hardware state detection method provided by an embodiment of the present application.
  • FIG. 9 is a schematic block diagram of a hardware state detection apparatus provided by an embodiment of the present application.
  • FIG. 10 is a schematic block diagram of a computer device provided by an embodiment of the present application.
  • FIG. 1 is a schematic flowchart of a hardware state detection method provided by an embodiment of the present application
  • FIG. 2 is a schematic diagram of an application scenario of the hardware state detection method provided by an embodiment of the present application
  • the hardware state detection method is applied to In the management server 10
  • the hardware state detection method is executed by the application software installed in the management server 10
  • the management server 10 is connected to a plurality of backup terminals 20 through a network to realize the transmission of data information
  • the management server 10 is the backup terminal.
  • the server side for intelligently detecting the state of the hardware equipment in the 20, the backup terminal 20 can be the terminal equipment configured by enterprises or government agencies in different areas for backing up and storing business data, and each backup terminal 20 is configured with a A plurality of hardware devices, wherein the hardware device may be a storage tape, for example, the backup terminal 20 may be a distributed storage terminal, a cluster storage terminal, and the like. As shown in FIG. 1, the method includes steps S110-S170.
  • the historical error reporting information contained in the pre-stored historical error reporting information table is matched according to the preset error reporting level matching rule to obtain the error reporting level of each historical error reporting information.
  • the historical error reporting information table contains the historical error reporting information corresponding to the storage tape, and errors will occur in the process of reading the data information stored in the storage tape, which can be obtained through the management server.
  • the historical error information is obtained by recording the specific information of the error in the tape. If an error occurs, the corresponding historical error information will be recorded.
  • the error degree of the data information corresponding to each historical error information is different.
  • Each historical error reporting information is matched to obtain a corresponding error reporting level, and the error reporting level can identify the error degree of the data information corresponding to each historical error reporting information.
  • the error reporting level matching rule includes a weighted value calculation formula and multiple error reporting level matching intervals.
  • step S110 includes sub-steps S111 , S112 and S113 .
  • the historically maintained information includes the proportion of important information and the error interval.
  • the proportion of important information is the proportion of the important information contained in the errored data information corresponding to the historical error reporting information.
  • the important information can be the customer ID number, transaction amount
  • the ratio of important information is the ratio between the number of data pieces of important information and the total number of pieces of important information in the data information with errors
  • the error interval time is the interval between the storage time of the data information with errors and the current time Time
  • step S111 further includes S111a before step S111.
  • S111a Perform data cleaning on the invalid information contained in the historical error reporting information table according to a preset data cleaning rule to obtain a historical error reporting information table after removing the invalid information.
  • Part of the historical error information also includes the tape information of the storage tape, and some of the historical error information does not contain the tape information of the storage tape.
  • the data cleaning rule can be to remove the historical error information that does not contain the tape information in the historical error information table, then Part of the historical error reporting information that does not contain tape information can be removed from the historical error reporting information table as invalid information to obtain the historical error reporting information table after removing the invalid information.
  • the weighted data volume of each historical error reporting information can be obtained by multiplying the data volume of each historical error reporting information by the weighted value corresponding to the historical error reporting information, wherein the data volume is the data of the data information in which the error occurred.
  • the amount of data can be expressed in kb.
  • the corresponding weighted data volume is 70.1875kb.
  • the error reporting level matching rule includes multiple error reporting level matching intervals. Each error reporting level matching interval corresponds to an error reporting level.
  • the error reporting level corresponding to the error reporting level matching interval to which each weighted data volume belongs can be obtained as the error reporting of the corresponding weighted data volume. Level, that is, to obtain the error level corresponding to each historical error message.
  • the error level corresponding to the error level matching interval [0, 15] is minor
  • the error level corresponding to the error level matching interval (15, 40) is generally serious
  • the error level corresponding to the error level matching interval (40, 100) is serious, and an error is reported.
  • the error level corresponding to the level matching interval (100,+ ⁇ ) is very serious, then the weighted data volume of 70.1875kb belongs to the error reporting level matching interval (40,100], and the error level corresponding to the weighted data volume is serious.
  • Each piece of the historical error reporting information in the historical error reporting information table is quantified according to a preset information quantification rule to obtain quantified error reporting information.
  • the information quantification rules are specific rules for quantifying each historical error reporting information in the historical error reporting information table. In the actual application process, the historical error reporting information contained in the historical error reporting information table after removing invalid information can be quantified.
  • the information quantification rule includes multiple quantification items, and each historical error reporting information can be converted into error reporting quantification information represented by normalized eigenvalues through the multiple quantification items included in the information quantification rule.
  • the various characteristics of the historical error reporting information can be quantified and represented by the error reporting quantification information, so as to facilitate the quantitative calculation based on the obtained error reporting quantification information, and the error reporting quantification information can be expressed as a Multidimensional vector, the number of dimensions of the multidimensional vector in the error quantization information is equal to the number of conversion items included in the information quantization rule.
  • step S120 includes sub-steps S121 and S122.
  • the quantitative items in the information quantification rule include information such as tape type, tape manufacturer, loading times, drive number, usage time, etc.
  • the tape type is the information that identifies the specific type of storage tape.
  • the tape manufacturer is the information that identifies the manufacturer of the storage tape, the loading times is the number of times the corresponding storage tape is loaded when the historical error information is obtained, and the drive number is the drive number information that drives the corresponding storage tape.
  • quantification processing can be carried out according to the item rules of each quantification item, and the quantification items can be divided into non-numerical quantification items. and numerical quantification projects.
  • the non-numerical quantization item of tape type in the information quantification rule corresponds to the keyword of each tape type, and the data corresponding to the keyword of "A type” is "0.1", and the keyword of "B type” The corresponding data is "0.3”, and the tape type in the attribute information of a certain item is type B, and the corresponding characteristic value is "0.3".
  • the corresponding quantification rule in the information quantification rule is an activation function and an intermediate value. Get the corresponding quantized value.
  • the activation function can be expressed as Wherein, x is an item of information corresponding to the numerical quantization item, and v is an intermediate value corresponding to the numerical quantization item.
  • the intermediate value of the numerical quantification item can be preset by the administrator, or can be obtained from the historical error reporting information table. If the intermediate value of the corresponding numerical quantification item is obtained from the historical error reporting information table, the corresponding steps include the following steps: obtaining the historical error reporting information table. Multiple attribute values corresponding to each numerical quantization item in the error reporting information table; calculating the average value of the multiple attribute values of each of the numerical quantization items as a median value corresponding to each of the numerical quantization items.
  • a plurality of attribute values corresponding to the historical error reporting information table may be obtained according to the numerical quantization item, and the average value of the plurality of attribute values may be calculated as an intermediate value of the corresponding numerical quantization item.
  • a corresponding state detection model is generated according to the error reporting level matching rule and the information quantification rule.
  • the generated state detection model is composed of multiple input nodes, multiple output nodes, and fully connected layers, and the corresponding multiple input nodes can be configured according to the number of quantification items included in the information quantification rules, and can be configured according to the error report.
  • the number of error reporting level matching intervals included in the level matching rule configures multiple output nodes, then each input node corresponds to a feature value in the error reporting quantization information, and each output node corresponds to an error reporting level matching interval, also That is, each output node corresponds to an error level.
  • a fully-connected layer is included between the input node and the output node.
  • the number of feature units included in the fully-connected layer can be preset by the administrator.
  • the first formula group is generated according to the input node and the feature unit, and the first formula group includes all input nodes.
  • the formulas in the first formula group take the input node value as the input value and the feature unit value as the output value; generate the second formula group according to the feature unit and the output node, then the second formula group contains all the features
  • the formulas from the unit to all output nodes, the formulas in the second formula group all take the characteristic unit value as the input value and the output node value as the output value.
  • the error quantization information corresponding to a historical error report information can be input into the state detection model, and the output node value corresponding to the historical error report information and each output node can be obtained after analysis.
  • S140 Perform iterative training on the state detection model according to a preset model training rule and the error reporting quantification information and error reporting level of each historical error reporting information, to obtain a trained state detection model.
  • the state detection model is iteratively trained according to the preset model training rules, the error reporting quantification information and the error reporting level of each historical error reporting information, and a trained state detection model is obtained. Before using the state detection model, it is necessary to use the above error quantification information and iteratively train the model according to the model training rules. Training the state detection model is to adjust the parameter values of the formulas in the model, and the model training rules are to The specific rules for training the state detection model, wherein the model training rules include loss value calculation formula, gradient calculation formula and loss threshold.
  • step S140 includes sub-steps S141 , S142 , S143 , S144 and S145 .
  • the model output information After inputting multiple eigenvalues contained in an error quantization information into the state detection model for calculation, the model output information is obtained.
  • the model output information includes the output node values of multiple output nodes, and each output node value is the error quantization information and the corresponding output.
  • the matching probability between the error level matching intervals of the nodes, and the value range of each output node value is [0, 1].
  • the loss value calculation formula can be expressed as Among them, f p is the matching probability of an output node in the model output information that matches the error level of the training data, and f n is the matching probability of the nth output node in the model output information, where n is the same as that in the risk rating model.
  • the number of included output nodes is equal, and the value ranges of f p and f n are both [0, 1].
  • S143 determine whether the loss value is less than the loss threshold; S144, if the loss value is not less than the loss threshold, calculate each parameter in the state detection model according to the gradient calculation formula and the loss value and update the parameter value of each of the parameters, and return to the step of inputting a piece of the error reporting quantification information into the state detection model to obtain the corresponding model output information; S145, if the loss value If it is less than the loss threshold, the state detection model is determined as the trained state detection model.
  • the parameter values of the parameters in the state detection model need to be updated.
  • the loss value and the calculated value of the state detection model can be calculated to obtain each state detection model.
  • the update value of the parameter after the parameter value in the state detection model is updated once, the next piece of error reporting quantization information can be obtained and the process returns to step S141.
  • the calculated value obtained by calculating a parameter in the state detection model for a piece of error reporting quantification information is input into the gradient calculation formula, and combined with the above loss value, the update value corresponding to the parameter can be calculated.
  • This calculation process also This is the gradient descent calculation.
  • the parameter value of each parameter in the state detection model is updated correspondingly, that is, a training process of the state detection model is completed.
  • the gradient calculation formula can be expressed as:
  • ⁇ s is the original parameter value of the parameter s
  • ⁇ is the preset learning rate in the gradient calculation formula
  • the calculated loss value is less than the loss threshold, it means that the state detection model at this time can meet the usage requirements, and the state detection model obtained at this time is used as the state detection model after training.
  • the hardware device can be a storage tape configured in the backup terminal, and a detection period can be set to periodically detect the storage tape. Specifically, it can be determined whether the interval between the current time and the last detection time point is not less than the detection period. If the interval between the current time and the last detection time point is not less than this time period, it indicates that the current time point is the detection time point of this detection, and the storage tapes contained in each backup terminal are stored at the current time. The state detection is performed to obtain the tape detection information of each storage tape.
  • the tape detection information at least includes information such as the tape type, the tape manufacturer, the number of times of loading, the drive number, and the usage time.
  • the state level of each hardware device can be obtained by sequentially inputting each piece of hardware detection information obtained at the current detection time point into the state detection model for analysis.
  • each tape detection information can be quantified according to the information quantification rules to obtain the corresponding tape detection quantification information, and the tape detection quantification information can be input into the state detection model to obtain the corresponding state.
  • the status level is the level information obtained by analyzing the status of the storage tape. By obtaining the status level of each tape detection information, an early warning can be given to the abnormal storage tape.
  • step S160 includes sub-steps S161 , S162 and S163 .
  • Each tape detection information can be quantized according to the above-mentioned information quantification rules to obtain the tape detection quantization information, the tape detection quantization information contains a plurality of characteristic values, the specific process of quantizing the tape detection information and the specific process of quantifying the historical error information. The same is not repeated here.
  • S162 input the tape detection quantization information into the state detection model, and obtain detection output information corresponding to each tape detection quantization information; S163, determine the error level corresponding to an output node with the largest output node value in the detection output information for the corresponding state level.
  • the tape detection quantification information is input into the state detection model for analysis and calculation, and the detection output information corresponding to each tape detection quantification information can be obtained, and the detection output information includes the output node value corresponding to each output node.
  • the error level corresponding to an output node is used as the status level of the corresponding tape detection quantization information.
  • the abnormal hardware information is obtained from the hardware detection information according to the state level of the hardware device and fed back to the administrator of the management server.
  • the abnormal hardware information is the abnormal tape information.
  • the storage tapes with serious and very serious status levels can be determined as the abnormal storage tapes, and the specific information of the abnormal storage tapes can be obtained as the abnormal tapes.
  • the information is fed back to the administrator.
  • the administrator is the user of the management server.
  • the abnormal tape information includes the tape type of the abnormal storage tape, the tape manufacturer, the usage time, the number of loading times, the tape identification code, the backup terminal to which it belongs, the region it belongs to, and the physical location.
  • the tape identification code is the identification code information uniquely corresponding to each storage tape
  • the backup terminal to which it belongs is the backup terminal to which the abnormal storage tape belongs
  • the area to which it belongs is the backup terminal corresponding to the abnormal storage tape.
  • the physical address is the specific location information of the storage tape in the backup terminal.
  • the physical address of a storage tape can be: P01 computer room-H rack-Q column-F row.
  • the abnormal tape information can also be analyzed to obtain analysis results and fed back to the administrator.
  • the status rate of the storage tape contained in each backup terminal is counted, and the status rate of the storage tape contained in each backup terminal is counted according to the abnormal storage tape.
  • Area count the status rate of the storage tapes contained in each area, and feed back the status rate of each backup terminal and the status rate of each area as analysis results to the administrator.
  • step S180 is further included after step S170 .
  • the corresponding summary information is obtained based on the abnormal tape information.
  • the summary information is obtained by hashing the abnormal tape information. , for example, processed by sha256 algorithm.
  • Uploading summary information to the blockchain ensures its security and fairness and transparency to users.
  • the user equipment can download the summary information from the blockchain in order to verify whether the abnormal tape information has been tampered with.
  • the blockchain referred to in this example is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
  • the technical method in this application can be applied to smart government affairs/smart city management/smart community/smart security/smart logistics/smart medical care/smart education/smart environmental protection/smart transportation and other application scenarios including intelligent detection of the status of storage tapes , so as to promote the construction of smart cities.
  • the error reporting level of each historical error reporting information in the historical error reporting information table is obtained, each historical error reporting information is quantized to obtain the error reporting quantization information, a state detection model is constructed, and a state detection model is constructed according to the error reporting method.
  • the quantitative information and the corresponding error reporting level are used to iteratively train the state detection model, and the hardware detection information obtained by periodic detection of the hardware devices included in the backup terminal is input into the state detection model, and the state level corresponding to each hardware device is obtained.
  • the status level obtains abnormal hardware information for feedback.
  • the embodiment of the present application further provides a hardware state detection apparatus, which can be configured in the management server 10, and is used to execute any embodiment of the aforementioned hardware state detection method.
  • a hardware state detection apparatus which can be configured in the management server 10, and is used to execute any embodiment of the aforementioned hardware state detection method.
  • FIG. 9 is a schematic block diagram of a hardware state detection apparatus provided by an embodiment of the present application.
  • the hardware state detection device 100 includes an error reporting level acquiring unit 110, an error reporting quantization information acquiring unit 120, a state detection model generating unit 130, a model training unit 140, a hardware detection information acquiring unit 150, a state level acquiring unit 160 and Abnormal hardware information feedback unit 170 .
  • the error reporting level obtaining unit 110 is configured to match the historical error reporting information contained in the pre-stored historical error reporting information table according to a preset error reporting level matching rule to obtain the error reporting level of each historical error reporting information.
  • the error reporting level obtaining unit 110 includes a subunit: a weighted value calculation unit, configured to calculate the important information ratio and error interval time of each of the historical error reporting information according to the weighted value calculation formula to Obtain the weighted value of each of the historical error reporting information; the weighted data volume acquisition unit is used to perform a weighted calculation on the data volume of each of the historical error reporting information according to the weighted value to obtain a corresponding data volume of each historical error reporting information. Weighted data volume; an error reporting level determination unit, configured to determine an error reporting level corresponding to each historical error reporting information according to the error reporting level matching interval to which the weighted data volume of each historical error reporting information belongs.
  • the error reporting level obtaining unit 110 further includes a subunit: a data cleaning unit, configured to perform data cleaning on the invalid information contained in the historical error reporting information table according to a preset data cleaning rule, and obtain the removal of invalid data.
  • the history error message table after the message after the message.
  • the error reporting quantification information obtaining unit 120 is configured to quantify each historical error reporting information in the historical error reporting information table according to a preset information quantification rule to obtain error reporting quantification information.
  • the error reporting quantification information obtaining unit 120 includes a subunit: an item attribute information obtaining unit, configured to obtain the corresponding item attribute from each of the historical error reporting information according to the quantification item in the information quantification rule. information; an information quantification processing unit, configured to quantify the item attribute information of each of the historical error reporting information according to the item rule of each of the quantified items to obtain corresponding quantified error reporting information.
  • the state detection model generating unit 130 is configured to generate a corresponding state detection model according to the error reporting level matching rule and the information quantification rule.
  • the model training unit 140 is configured to iteratively train the state detection model according to the preset model training rules and the error reporting quantification information and error reporting level of each of the historical error reporting information to obtain a trained state detection model.
  • the model training unit 140 includes subunits: a model output information acquisition unit, configured to input a piece of the error reporting quantization information into the state detection model to acquire corresponding model output information; a loss value calculation unit, is used to calculate the loss value between the error reporting level of the error reporting quantization information and the model output information according to the loss value calculation formula; the loss value judgment unit is used to judge whether the loss value is less than the loss threshold; A parameter value update unit, configured to obtain an update value of each parameter in the state detection model according to the gradient calculation formula and the loss value if the loss value is not less than the loss threshold The parameter value of the parameter is updated, and the step of inputting a piece of the error reporting quantification information into the state detection model to obtain the corresponding model output information is returned to execute; the state detection model determination unit is used for if the loss value is less than the loss threshold, and the state detection model is determined as the state detection model after training.
  • a model output information acquisition unit configured to input a piece of the error reporting quantization information into the state detection model to acquire
  • the hardware detection information acquiring unit 150 is configured to periodically detect the hardware devices included in each of the backup terminals according to a preset detection period, so as to acquire hardware detection information of each of the hardware devices.
  • the state level acquiring unit 160 is configured to input the hardware detection information acquired at the current detection time point into the state detection model to acquire the state level corresponding to each of the hardware devices.
  • the state level acquisition unit 160 includes a subunit: a tape detection quantization information acquisition unit, configured to quantify each of the tape detection information according to the information quantization rule to obtain corresponding tape detection quantization information;
  • the detection output information acquisition unit is used for inputting the tape detection quantization information into the state detection model to obtain detection output information corresponding to each tape detection quantization information;
  • the state level determination unit is used for outputting the detection output information in the output node The error level corresponding to an output node with the largest value is determined as the corresponding status level.
  • the abnormal hardware information feedback unit 170 is configured to obtain abnormal hardware information from the hardware detection information according to the state level of the hardware device and feed it back to the administrator of the management server.
  • the hardware state detection apparatus 100 further includes a subunit: a synchronization storage unit, configured to synchronously upload the abnormal hardware information to the blockchain for storage.
  • the hardware state detection device provided by the embodiment of the present application applies the above hardware state detection method, obtains the error level of each historical error message in the historical error message table, performs quantization processing on each historical error message to obtain the error message quantization information, and constructs a state
  • the detection model is iteratively trained on the status detection model according to the error quantification information and the corresponding error level, and the hardware detection information obtained by periodic detection of the hardware devices included in the backup terminal is input into the status detection model to obtain the corresponding information of each hardware device.
  • Status level and obtain abnormal hardware information for feedback according to the status level.
  • the state level of each hardware device is obtained based on the hardware detection information, and the abnormal hardware information is obtained to give an early warning to the abnormal hardware device, and the state of the magnetic tape and other hardware devices can be detected accurately and efficiently.
  • the above hardware state detection apparatus can be implemented in the form of a computer program, and the computer program can be executed on a computer device as shown in FIG. 10 .
  • FIG. 10 is a schematic block diagram of a computer device provided by an embodiment of the present application.
  • the computer device may be a management server 10 for executing a hardware state detection method to intelligently detect the state of the hardware device.
  • the computer device 500 includes a processor 502 , a memory and a network interface 505 connected through a system bus 501 , wherein the memory may include a storage medium 503 and an internal memory 504 .
  • the storage medium 503 can store an operating system 5031 and a computer program 5032 .
  • the computer program 5032 When executed, it can cause the processor 502 to execute the hardware state detection method, wherein the storage medium 503 can be a volatile storage medium or a non-volatile storage medium.
  • the processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .
  • the internal memory 504 provides an environment for running the computer program 5032 in the storage medium 503.
  • the processor 502 can execute the hardware state detection method.
  • the network interface 505 is used for network communication, such as providing transmission of data information.
  • the network interface 505 is used for network communication, such as providing transmission of data information.
  • FIG. 10 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied.
  • the specific computer device 500 may include more or fewer components than shown, or combine certain components, or have a different arrangement of components.
  • the processor 502 is configured to run the computer program 5032 stored in the memory, so as to realize the corresponding functions in the above-mentioned hardware state detection method.
  • the embodiment of the computer device shown in FIG. 10 does not constitute a limitation on the specific structure of the computer device. Either some components are combined, or different component arrangements.
  • the computer device may only include a memory and a processor.
  • the structures and functions of the memory and the processor are the same as those of the embodiment shown in FIG. 10 , which will not be repeated here.
  • the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor can be a microprocessor or the processor can also be any conventional processor or the like.
  • a computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.
  • the computer-readable storage medium stores a computer program, wherein when the computer program is executed by the processor, the steps included in the above-mentioned hardware state detection method are implemented.
  • the disclosed apparatus, apparatus and method may be implemented in other manners.
  • the device embodiments described above are only illustrative.
  • the division of the units is only logical function division.
  • there may be other division methods, or units with the same function may be grouped into one Units, such as multiple units or components, may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solutions of the embodiments of the present application.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium.
  • the read storage medium includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned computer-readable storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), magnetic disk or optical disk and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Fuzzy Systems (AREA)
  • Software Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mathematical Physics (AREA)
  • Debugging And Monitoring (AREA)

Abstract

一种硬件状态检测方法、装置、计算机设备及存储介质，方法包括：获取历史报错信息表中每一历史报错信息的报错等级，对每一历史报错信息进行量化处理得到报错量化信息，构建状态检测模型并根据报错量化信息及对应的报错等级对状态检测模型进行迭代训练，将对备份终端所包含的硬件设备进行周期性检测得到的硬件检测信息输入状态检测模型，获取对应的状态等级并进一步获取异常硬件信息进行反馈。该方案基于主机设备检测技术，属于硬件管理技术领域，涉及区块链技术，基于硬件检测信息得到每一硬件设备的状态等级，并获取异常硬件信息以对存在异常的硬件设备进行预警，可准确、高效地对硬件设备的状态进行检测。

Description

硬件状态检测方法、装置、计算机设备及存储介质
本申请要求于2020年12月25日提交中国专利局、申请号为202011561806.X，发明名称为“硬件状态检测方法、装置、计算机设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及硬件管理技术领域，属于智慧城市中对硬件设备的状态进行智能化检测的应用场景，尤其涉及一种硬件状态检测方法、装置、计算机设备及存储介质。
背景技术
随着大数据时代的蓬勃发展，企业对业务连续性的重视程度也越来越高。为了避免数据丢失造成的企业出现不可挽回的业务损失，对业务数据进行备份存储显得尤为重要。目前从性价比上来看，长期和大量数据还是备份到磁带设备上，磁带介质成了主流备份数据存储介质，为了验证备份数据的有效性，我们需要定期对磁带进行状态监测，对有问题的磁带进行及时处理。
申请人发现，现有技术中均是采用专用设备对磁带的磁带头芯片的数据进行读取，进而获取磁带在进行数据读取过程中所记录的运行信息来判断磁带状态情况。这一检测方法，需要将正在运行的带库停机并取出磁带，检测完成之后再将磁带放回，所对应的流程较为复杂、繁琐，检测过程中需要耗费大量时间，这一检测方法存在效率不高的问题，且答复增加了企业的运营成本。因此，现有技术方法中存在无法高效地对磁带等硬件设备的状态进行检测的问题。
发明内容
本申请实施例提供了一种硬件状态检测方法、装置、计算机设备及存储介质，旨在解决现有技术方法中所存在的无法高效地对磁带等硬件设备的状态进行检测的问题。
第一方面，本申请实施例提供了一种硬件状态检测方法，其包括：
根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理 服务器的管理员。
第二方面，本申请实施例提供了一种硬件状态检测装置，其包括：
报错等级获取单元，用于根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
报错量化信息获取单元，用于根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
状态检测模型生成单元，用于根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
模型训练单元，用于根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
硬件检测信息获取单元，用于根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
状态等级获取单元，用于将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
异常硬件信息反馈单元，用于根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。
第三方面，本申请实施例又提供了一种计算机设备，其包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述第一方面所述的硬件状态检测方法。
第四方面，本申请实施例还提供了一种计算机可读存储介质，其中所述计算机可读存储介质存储有计算机程序，所述计算机程序当被处理器执行时使所述处理器执行上述第一方面所述的硬件状态检测方法。
本申请实施例提供了一种硬件状态检测方法、装置、计算机可读存储介质。获取历史报错信息表中每一历史报错信息的报错等级，对每一历史报错信息进行量化处理得到报错量化信息，构建状态检测模型并根据报错量化信息及对应的报错等级对状态检测模型进行迭代训练，将对备份终端所包含的硬件设备进行周期性检测得到的硬件检测信息输入状态检测模型，获取每一硬件设备对应的状态等级，并根据状态等级获取异常硬件信息进行反馈。通过上述方法，基于硬件检测信息得到每一硬件设备的状态等级，并获取异常硬件信息以对存在异常的硬件设备进行预警，可准确、高效地对磁带等硬件设备的状态进行检测。
附图说明
为了更清楚地说明本申请实施例技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的硬件状态检测方法的流程示意图；
图2为本申请实施例提供的硬件状态检测方法的应用场景示意图；
图3为本申请实施例提供的硬件状态检测方法的子流程示意图；
图4为本申请实施例提供的硬件状态检测方法的另一子流程示意图；
图5为本申请实施例提供的硬件状态检测方法的另一子流程示意图；
图6为本申请实施例提供的硬件状态检测方法的另一子流程示意图；
图7为本申请实施例提供的硬件状态检测方法的另一子流程示意图；
图8为本申请实施例提供的硬件状态检测方法的另一流程示意图；
图9为本申请实施例提供的硬件状态检测装置的示意性框图；
图10为本申请实施例提供的计算机设备的示意性框图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
应当理解，当在本说明书和所附权利要求书中使用时，术语“包括”和“包含”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。
还应当理解，在此本申请说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本申请。如在本申请说明书和所附权利要求书中所使用的那样，除非上下文清楚地指明其它情况，否则单数形式的“一”、“一个”及“该”意在包括复数形式。
还应当进一步理解，在本申请说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。
请参阅图1及图2，图1是本申请实施例提供的硬件状态检测方法的流程示意图，图2为本申请实施例提供的硬件状态检测方法的应用场景示意图；该硬件状态检测方法应用于管理服务器10中，该硬件状态检测方法通过安装于管理服务器10中的应用软件进行执行，管理服务器10与多台备份终端20进行网络连接以实现数据信息的传输，管理服务器10即是对备份终端20中硬件设备的状态进行智能化检测的服务器端，备份终端20可以为企业或政府机构在不同区域所配置的用于对业务数据进行备份存储的终端设备，每一备份终端20中均配置有多个硬件设备，其中硬件设备可以是存储磁带，例如，备份终端20可以是分布式存储终端、集群存储终端等。如图1所示，该方法包括步骤S110～S170。
S110、根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级。
根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级。以硬件设备为存储磁带为例，则历史报错信息表中包含与存储磁带对应的历史报错信息，对存储磁带中已存储的数据信息进行读取的过程会出现错误，可通过管理服务器获取对存储磁带出现错误的具体信息进行记录所得到历史报错信息，则出现一次错误均会记录得到对应的历史报错信息，每一历史报错信息对应的数据信息的错误程度不同，则可根据报错等级匹配规则对每一历史报错信息进行匹配以得到相应的 报错等级，报错等级即可对每一历史报错信息所对应的数据信息的错误程度进行标识。所述报错等级匹配规则中包括加权值计算公式及多个报错等级匹配区间。
在一实施例中，如图3所示，步骤S110包括子步骤S111、S112和S113。
S111、根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值。
历史保持信息中包括重要信息比例及差错间隔时间，重要信息比例即为历史报错信息所对应的出现错误的数据信息中所包含的重要信息的比例值，重要信息可以是客户身份证号、交易金额等信息，重要信息比例即为出现错误的数据信息中重要信息的数据条数与总数量条数之间的比值，差错间隔时间即为出现错误的数据信息的存储时间与当前时间之间的间隔时间，可将每一历史报错信息的重要信息比例及差错间隔时间输入加权值计算公式并计算得到对应的加权值。
例如，加权值计算公式可以是J＝ln(1.5+12/t)×(e r/0.3-1)/2.5，其中，r为重要信息比例，t为差错间隔时间，若r为0.6，t为8个月，则对应得到的加权值J为2.8075。
在一实施例中，如图4所示，步骤S111之前还包括S111a。
S111a、根据预置的数据清洗规则对所述历史报错信息表所包含的无效信息进行数据清洗，得到去除无效信息后的历史报错信息表。
部分历史报错信息中还包含存储磁带的磁带信息，部分历史报错信息中未包含存储磁带的磁带信息，具体的，数据清洗规则可以是去除历史报错信息表中未包含磁带信息的历史报错信息，则可将部分未包含磁带信息的历史报错信息作为无效信息从历史报错信息表中进行去除，得到去除无效信息后的历史报错信息表。
S112、根据所述加权值对每一所述历史报错信息的数据量进行加权计算得到与每一所述历史报错信息对应的加权数据量。
具体的，将每一历史报错信息的数据量与该历史报错信息对应的加权值相乘，即可得到每一历史报错信息的加权数据量，其中，数据量即为出现错误的数据信息的数据量大小，数据量可采用kb进行表示。
例如，历史报错信息的数据量为25kb，加权值J为2.8075，则对应得到的加权数据量为70.1875kb。
S113、根据每一所述历史报错信息的加权数据量所属的报错等级匹配区间确定与每一所述历史报错信息对应的报错等级。
报错等级匹配规则中包括多个报错等级匹配区间，每一报错等级匹配区间均对应一个报错等级，可获取每一加权数据量所属的报错等级匹配区间对应的报错等级，作为相应加权数据量的报错等级，也即是获取得到每一历史报错信息对应的报错等级。
例如，报错等级匹配区间[0,15]对应的报错等级为轻微，报错等级匹配区间(15,40]对应的报错等级为一般严重，报错等级匹配区间(40,100]对应的报错等级为严重，报错等级匹配区间(100,+∞)对应的报错等级为十分严重，则加权数据量70.1875kb属于报错等级匹配区间(40,100]，与该加权数据量对应的报错等级为严重。
S120、根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息。
根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息。信息量化规则即为对历史报错信息表中每一历史报错信息进行量化的具体规则，在实际应用过程中，可对去除无效信息后的历史报错信息表中包含的历史报错信息进行量化。信息量化规则中包含多个量化项目，可通过信息量化规则包含的多个量化项目将与每一历史报错信息转换为以归一化的特征值进行表示的报错量化信息。将与历史报错信息转换为报错量化信息，即可通过报错量化信息对与该历史报错信息的各项特征进行量化表示，以方便基于得到的报错量化信息进行量化计算，报错量化信息可表示为一个多维向量，报错量化信息中多维向量的维度数与信息量化规则中所包含的转换项目的数量相等。
在一实施例中，如图5所示，步骤S120包括子步骤S121和S122。
S121、根据所述信息量化规则中的量化项目从每一所述历史报错信息中获取对应的项目属性信息。
以硬件设备为存储磁带为例，则信息量化规则中的量化项目包括磁带类型、磁带厂商、加载次数、驱动器编号、使用时长等信息，磁带类型即为对存储磁带的具体类型进行标识的信息，磁带厂商即为对存储磁带的生产厂商进行标识的信息，加载次数即为获取历史报错信息时对应存储磁带被加载的次数，驱动器编号即为对相应存储磁带进行驱动的驱动器编号信息，使用时长即为获取历史报错信息时对应存储磁带被使用的时长，获取到上述量化项目对应的项目属性信息后，即可根据每一量化项目的项目规则对应进行量化处理，量化项目可分为非数值量化项目及数值量化项目。
S122、根据每一所述量化项目的项目规则对每一所述历史报错信息的项目属性信息分别进行量化得到对应的报错量化信息。
则对一个历史报错信息的项目属性信息进行量化，即可得到对应的一条报错量化信息，对项目属性信息中的每一项信息进行量化所得特征值的范围均为[0,1]。具体的，对于与量化项目对应的一项信息以非数值方式进行表示的情况，则直接获取非数值量化项目中与该非数值相匹配的关键字所对应的数据，作为与该非数值对应的量化值。
例如，信息量化规则中磁带类型这一非数值量化项目对应包含每种磁带类型的关键字，与“A类型”这一关键字对应的数据为“0.1”、与“B类型”这一关键字对应的数据为“0.3”，某一项目属性信息中的磁带类型为B类型，则对应的特征值为“0.3”。
对于与量化项目对应的信息以数值方式表示的情况，信息量化规则中对应的量化规则为一个激活函数及一个中间值，根据激活函数对中间值及数值量化项目的一项信息进行计算，即可得到对应的量化值。
例如，激活函数可表示为
Figure PCTCN2021083760-appb-000001
其中，x为与数值量化项目对应的一项信息，v为与该数值量化项目对应的中间值。与加载次数这一数值量化项目对应的中间值为v＝160，某一项目属性信息中的加载次数为x＝247，则根据上述激活函数计算得到对应的特征值为0.3673。
其中，数值量化项目的中间值可由管理员预设，也可从历史报错信息表中获取，若从历史报错信息表中获取相应数值量化项目的中间值，则对应包含以下步骤：获取所述历史报错信息表中与每一数值量化项目对应的多个属性值；计算每一所述数值量化项目的多个属性值的平均值，作为与每一所述数值量化项目对应的中间值。
具体的，可根据数值量化项目获取历史报错信息表中对应的多个属性值，计算多个属性值的平均值作为相应数值量化项目的中间值。
S130、根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型。
根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型。具体的，所生成的状态检测模型中由多个输入节点、多个输出节点及全连接层组成，可根据信息量化规则中所包含的量化项目的数量配置对应的多个输入节点，可根据报错等级匹配规则中所包含的报错等级匹配区间的数量配置对应的多个输出节点，则每一输入节点均对应报错量化信息中的一个特征值，每一输出节点均对应一个报错等级匹配区间，也即是每一输出节点与一个报错等级相对应。输入节点与输出节点之间包含全连接层，全连接层中包含的特征单元的数量可由管理员预先设定，根据输入节点及特征单元生成第一公式组，则第一公式组包含所有输入节点至所有特征单元的公式，第一公式组中的公式均以输入节点值作为输入值、特征单元值作为输出值；根据特征单元及输出节点生成第二公式组，则第二公式组包含所有特征单元至所有输出节点的公式，第二公式组中的公式均以特征单元值作为输入值、输出节点值作为输出值。可将一个历史报错信息对应的报错量化信息输入状态检测模型，进行分析即可得到该历史报错信息与每一输出节点对应的输出节点值。
S140、根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型。
根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型。在使用状态检测模型之前，需要使用上述报错量化信息并根据模型训练规则对模型进行迭代训练，对状态检测模型进行训练也即是对模型中包含公式的参数值进行调整，模型训练规则即为对状态检测模型进行训练的具体规则，其中模型训练规则包括损失值计算公式、梯度计算公式及损失阈值。
在一实施例中，如图6所示，步骤S140包括子步骤S141、S142、S143、S144和S145。
S141、将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息。
将一条报错量化信息包含的多个特征值输入状态检测模型进行计算后得到模型输出信息，模型输出信息即包含多个输出节点的输出节点值，每一输出节点值即为报错量化信息与相应输出节点的报错等级匹配区间之间的匹配概率，每一输出节点值的取值范围均为[0，1]。
S142、根据所述损失值计算公式计算得到所述报错量化信息的报错等级与所述模型输出信息之间的损失值。
例如，损失值计算公式可表示为
Figure PCTCN2021083760-appb-000002
其中，f p为模型输出信息中与该训练数据的报错等级相匹配的一个输出节点的匹配概率，f n为模型输出信息中第n个输 出节点的匹配概率，其中，n与风险评级模型中包含的输出节点的数量相等，f p及f n的取值范围均为[0，1]。
S143、判断所述损失值是否小于所述损失阈值；S144、若所述损失值不小于所述损失阈值，根据所述梯度计算公式及所述损失值计算得到所述状态检测模型中每一参数的更新值并对每一所述参数的参数值进行更新，返回执行所述将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息的步骤；S145、若所述损失值小于所述损失阈值，将所述状态检测模型确定为训练后的状态检测模型。
若损失值不小于损失阈值，即需要对状态检测模型中参数的参数值进行更新，具体的，根据梯度计算公式对损失值及状态检测模型的计算值进行计算即可得到状态检测模型中每一参数的更新值，对状态检测模型中参数值进行一次更新后，即可获取下一条报错量化信息并返回执行步骤S141。
具体的，将状态检测模型中一个参数对一条报错量化信息进行计算所得到的计算值输入梯度计算公式，并结合上述损失值，即可计算得到与该参数对应的更新值，这一计算过程也即为梯度下降计算。基于所计算得到更新值对状态检测模型中每一参数的参数值对应更新，即完成对状态检测模型的一次训练过程。
具体的，梯度计算公式可表示为：
Figure PCTCN2021083760-appb-000003
其中，
Figure PCTCN2021083760-appb-000004
为计算得到的参数s的更新值，ω s为参数s的原始参数值，η为梯度计算公式中预置的学习率，
Figure PCTCN2021083760-appb-000005
为基于损失值及参数s对应的计算值对该参数s的偏导值(这一计算过程中需使用参数对应的计算值)。
若计算得到的损失值小于损失阈值，即表明此时的状态检测模型可满足使用需求，将此时所得到的状态检测模型作为训练后的状态检测模型。
S150、根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息。
根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息。硬件设备可以是备份终端内所配置的存储磁带，可设置检测周期以对存储磁带进行周期性检测，具体的，可判断当前时间与上一次检测时间点之间的间隔时间是否不小于检测周期中的时间周期，若当前时间与上一次检测时间点之间的间隔时间不小于该时间周期，则表明当前时间点为本次检测的检测时间点，在当前时间对每一备份终端包含的存储磁带进行状态检测，得到每一存储磁带的磁带检测信息，磁带检测信息中至少包括磁带类型、磁带厂商、加载次数、驱动器编号、使用时长等信息。
S160、将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级。
将当前检测时间点获取到的每一条硬件检测信息依次输入状态检测模型进行分析，即可获取每一硬件设备的状态等级。以硬件检测信息为磁带检测信息为例，具体的，可根据信息量化规则对每一条磁带检测信息进行量化得到对应的磁带检测量化信息，将磁带检测量化信 息输入状态检测模型即可获取对应的状态等级，状态等级即为对存储磁带的状态进行分析所得到的等级信息，通过获取每一磁带检测信息的状态等级即可对存在异常的存储磁带进行预警。
在一实施例中，如图7所示，步骤S160包括子步骤S161、S162和S163。
S161、根据所述信息量化规则对每一所述磁带检测信息进行量化得到对应的磁带检测量化信息。
可根据上述信息量化规则对每一磁带检测信息进行量化得到磁带检测量化信息，磁带检测量化信息中包含多个特征值，对磁带检测信息进行量化的具体过程与对历史报错信息进行量化的具体过程相同，在此不作赘述。
S162、将所述磁带检测量化信息输入所述状态检测模型，得到与每一磁带检测量化信息对应的检测输出信息；S163、将检测输出信息中输出节点值最大的一个输出节点对应的报错等级确定为对应的状态等级。
将磁带检测量化信息输入状态检测模型进行分析计算，即可得到每一磁带检测量化信息对应的检测输出信息，检测输出信息中包括与每一输出节点对应的输出节点值，获取输出节点值最大的一个输出节点对应的报错等级作为相应磁带检测量化信息的状态等级。
S170、根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。
根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。具体的，以硬件设备为存储磁带为例，则异常硬件信息即为异常磁带信息，可将状态等级为严重及十分严重的存储磁带确定为异常存储磁带，获取异常存储磁带的具体信息作为异常磁带信息并反馈至管理员，管理员即为管理服务器的使用者，异常磁带信息中包含异常存储磁带的磁带类型、磁带厂商、使用时长、加载次数、磁带识别码、所属备份终端、所属区域、物理地址等具体信息，其中，磁带识别码即为与每一存储磁带唯一对应的识别码信息，所属备份终端即为异常存储磁带对应所属的备份终端，所属区域即为异常存储磁带对应所属备份终端的配置区域信息，物理地址即为存储磁带在备份终端内所对应的具体位置信息，例如，某一存储磁带的物理地址可以是：P01机房-H机架-Q列-F行。管理员获取异常磁带信息后即可将相应异常存储磁带内存储的数据信息转移至其他存储磁带内进行存储，或对异常存储磁带进行更新替换。
具体的，还可对异常磁带信息进行分析得到分析结果并反馈至管理员，例如，根据异常存储磁带的所属备份终端，统计每一备份终端所包含存储磁带的状态率，根据异常存储磁带的所属区域，统计每一区域包含的存储磁带的状态率，将每一备份终端的状态率及每一区域的状态率作为分析结果反馈至管理员。
在一实施例中，如图8所示，步骤S170之后还包括步骤S180。
S180、将所述异常硬件信息同步上传至区块链进行存储。
将所述异常硬件信息上传至区块链进行存储，以异常硬件信息为异常磁带信息为例，基于异常磁带信息得到对应的摘要信息，具体来说，摘要信息由异常磁带信息进行散列处理得 到，比如利用sha256算法处理得到。将摘要信息上传至区块链可保证其安全性和对用户的公正透明性。用户设备可以从区块链中下载得该摘要信息，以便查证异常磁带信息是否被篡改。本示例所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain)，本质上是一个去中心化的数据库，是一串使用密码学方法相关联产生的数据块，每一个数据块中包含了一批次网络交易的信息，用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
本申请中的技术方法可应用于智慧政务/智慧城管/智慧社区/智慧安防/智慧物流/智慧医疗/智慧教育/智慧环保/智慧交通等包含对存储磁带的状态进行智能化检测的应用场景中，从而推动智慧城市的建设。
在本申请实施例所提供的硬件状态检测方法中，获取历史报错信息表中每一历史报错信息的报错等级，对每一历史报错信息进行量化处理得到报错量化信息，构建状态检测模型并根据报错量化信息及对应的报错等级对状态检测模型进行迭代训练，将对备份终端所包含的硬件设备进行周期性检测得到的硬件检测信息输入状态检测模型，获取每一硬件设备对应的状态等级，并根据状态等级获取异常硬件信息进行反馈。通过上述方法，基于硬件检测信息得到每一硬件设备的状态等级，并获取异常硬件信息以对存在异常的硬件设备进行预警，可准确、高效地对磁带等硬件设备的状态进行检测。
本申请实施例还提供一种硬件状态检测装置，该硬件状态检测装置可配置于管理服务器10中，该硬件状态检测装置用于执行前述的硬件状态检测方法的任一实施例。具体地，请参阅图9，图9为本申请实施例提供的硬件状态检测装置的示意性框图。
如图9所示，硬件状态检测装置100包括报错等级获取单元110、报错量化信息获取单元120、状态检测模型生成单元130、模型训练单元140、硬件检测信息获取单元150、状态等级获取单元160和异常硬件信息反馈单元170。
报错等级获取单元110，用于根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级。
在一实施例中，所述报错等级获取单元110包括子单元：加权值计算单元，用于根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值；加权数据量获取单元，用于根据所述加权值对每一所述历史报错信息的数据量进行加权计算得到与每一所述历史报错信息对应的加权数据量；报错等级确定单元，用于根据每一所述历史报错信息的加权数据量所属的报错等级匹配区间确定与每一所述历史报错信息对应的报错等级。
在一实施例中，所述报错等级获取单元110还包括子单元：数据清洗单元，用于根据预置的数据清洗规则对所述历史报错信息表所包含的无效信息进行数据清洗，得到去除无效信息后的历史报错信息表。
报错量化信息获取单元120，用于根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息。
在一实施例中，所述报错量化信息获取单元120包括子单元：项目属性信息获取单元，用于根据所述信息量化规则中的量化项目从每一所述历史报错信息中获取对应的项目属性信息；信息量化处理单元，用于根据每一所述量化项目的项目规则对每一所述历史报错信息的项目属性信息分别进行量化得到对应的报错量化信息。
状态检测模型生成单元130，用于根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型。
模型训练单元140，用于根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型。
在一实施例中，所述模型训练单元140包括子单元：模型输出信息获取单元，用于将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息；损失值计算单元，用于根据所述损失值计算公式计算得到所述报错量化信息的报错等级与所述模型输出信息之间的损失值；损失值判断单元，用于判断所述损失值是否小于所述损失阈值；参数值更新单元，用于若所述损失值不小于所述损失阈值，根据所述梯度计算公式及所述损失值计算得到所述状态检测模型中每一参数的更新值并对每一所述参数的参数值进行更新，返回执行所述将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息的步骤；状态检测模型确定单元，用于若所述损失值小于所述损失阈值，将所述状态检测模型确定为训练后的状态检测模型。
硬件检测信息获取单元150，用于根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息。
状态等级获取单元160，用于将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级。
在一实施例中，所述状态等级获取单元160包括子单元：磁带检测量化信息获取单元，用于根据所述信息量化规则对每一所述磁带检测信息进行量化得到对应的磁带检测量化信息；检测输出信息获取单元，用于将所述磁带检测量化信息输入所述状态检测模型，得到与每一磁带检测量化信息对应的检测输出信息；状态等级确定单元，用于将检测输出信息中输出节点值最大的一个输出节点对应的报错等级确定为对应的状态等级。
异常硬件信息反馈单元170，用于根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。
在一实施例中，所述硬件状态检测装置100还包括子单元：同步存储单元，用于将所述异常硬件信息同步上传至区块链进行存储。
在本申请实施例所提供的硬件状态检测装置应用上述硬件状态检测方法，获取历史报错信息表中每一历史报错信息的报错等级，对每一历史报错信息进行量化处理得到报错量化信息，构建状态检测模型并根据报错量化信息及对应的报错等级对状态检测模型进行迭代训练，将对备份终端所包含的硬件设备进行周期性检测得到的硬件检测信息输入状态检测模型，获取每一硬件设备对应的状态等级，并根据状态等级获取异常硬件信息进行反馈。通过上述方法，基于硬件检测信息得到每一硬件设备的状态等级，并获取异常硬件信息以对存在异常的 硬件设备进行预警，可准确、高效地对磁带等硬件设备的状态进行检测。
上述硬件状态检测装置可以实现为计算机程序的形式，该计算机程序可以在如图10所示的计算机设备上运行。
请参阅图10，图10是本申请实施例提供的计算机设备的示意性框图。该计算机设备可以是用于执行硬件状态检测方法以对硬件设备的状态进行智能化检测的管理服务器10。
参阅图10，该计算机设备500包括通过系统总线501连接的处理器502、存储器和网络接口505，其中，存储器可以包括存储介质503和内存储器504。
该存储介质503可存储操作系统5031和计算机程序5032。该计算机程序5032被执行时，可使得处理器502执行硬件状态检测方法，其中，存储介质503可以为易失性的存储介质或非易失性的存储介质。
该处理器502用于提供计算和控制能力，支撑整个计算机设备500的运行。
该内存储器504为存储介质503中的计算机程序5032的运行提供环境，该计算机程序5032被处理器502执行时，可使得处理器502执行硬件状态检测方法。
该网络接口505用于进行网络通信，如提供数据信息的传输等。本领域技术人员可以理解，图10中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备500的限定，具体的计算机设备500可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
其中，所述处理器502用于运行存储在存储器中的计算机程序5032，以实现上述的硬件状态检测方法中对应的功能。
本领域技术人员可以理解，图10中示出的计算机设备的实施例并不构成对计算机设备具体构成的限定，在其他实施例中，计算机设备可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，在一些实施例中，计算机设备可以仅包括存储器及处理器，在这样的实施例中，存储器及处理器的结构及功能与图10所示实施例一致，在此不再赘述。
应当理解，在本申请实施例中，处理器502可以是中央处理单元(Central Processing Unit，CPU)，该处理器502还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。其中，通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
在本申请的另一实施例中提供计算机可读存储介质。该计算机可读存储介质可以为易失性或非易失性的计算机可读存储介质。该计算机可读存储介质存储有计算机程序，其中计算机程序被处理器执行时实现上述的硬件状态检测方法中所包含的步骤。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，上述描述的设备、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能 够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中，应该理解到，所揭露的设备、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为逻辑功能划分，实际实现时可以有另外的划分方式，也可以将具有相同功能的单元集合成一个单元，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口、装置或单元的间接耦合或通信连接，也可以是电的，机械的或其它的形式连接。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本申请实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以是两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分，或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个计算机可读存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的计算机可读存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以权利要求的保护范围为准。

Claims (20)

  1. 一种硬件状态检测方法，应用于管理服务器中，所述管理服务器同时与多台备份终端通过网络连接进行数据信息的传输，其中，所述方法包括：
    根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
    根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
    根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
    根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
    根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
    将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
    根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。
  2. 根据权利要求1所述的硬件状态检测方法，其中，所述报错等级匹配规则包括加权值计算公式及多个报错等级匹配区间，所述根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级，包括：
    根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值；
    根据所述加权值对每一所述历史报错信息的数据量进行加权计算得到与每一所述历史报错信息对应的加权数据量；
    根据每一所述历史报错信息的加权数据量所属的报错等级匹配区间确定与每一所述历史报错信息对应的报错等级。
  3. 根据权利要求2所述的硬件状态检测方法，其中，所述根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值之前，还包括：
    根据预置的数据清洗规则对所述历史报错信息表所包含的无效信息进行数据清洗，得到去除无效信息后的历史报错信息表。
  4. 根据权利要求1所述的硬件状态检测方法，其中，所述根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息，包括：
    根据所述信息量化规则中的量化项目从每一所述历史报错信息中获取对应的项目属性信息；
    根据每一所述量化项目的项目规则对每一所述历史报错信息的项目属性信息分别进行量化得到对应的报错量化信息。
  5. 根据权利要求1所述的硬件状态检测方法，其中，所述模型训练规则包括损失值计算 公式、梯度计算公式及损失阈值，所述根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型，包括：
    将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息；
    根据所述损失值计算公式计算得到所述报错量化信息的报错等级与所述模型输出信息之间的损失值；
    判断所述损失值是否小于所述损失阈值；
    若所述损失值不小于所述损失阈值，根据所述梯度计算公式及所述损失值计算得到所述状态检测模型中每一参数的更新值并对每一所述参数的参数值进行更新，返回执行所述将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息的步骤；
    若所述损失值小于所述损失阈值，将所述状态检测模型确定为训练后的状态检测模型。
  6. 根据权利要求1所述的硬件状态检测方法，其中，所述硬件检测信息为磁带检测信息，所述硬件设备为存储磁带，所述将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级，包括：
    根据所述信息量化规则对每一所述磁带检测信息进行量化得到对应的磁带检测量化信息；
    将所述磁带检测量化信息输入所述状态检测模型，得到与每一磁带检测量化信息对应的检测输出信息；
    将检测输出信息中输出节点值最大的一个输出节点对应的报错等级确定为对应的状态等级。
  7. 根据权利要求1所述的硬件状态检测方法，其中，所述根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员之后，还包括：
    将所述异常硬件信息同步上传至区块链进行存储。
  8. 一种硬件状态检测装置，其中，所述装置包括：
    报错等级获取单元，用于根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
    报错量化信息获取单元，用于根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
    状态检测模型生成单元，用于根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
    模型训练单元，用于根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
    硬件检测信息获取单元，用于根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
    状态等级获取单元，用于将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
    异常硬件信息反馈单元，用于根据所述硬件设备的状态等级从所述硬件检测信息中获取 异常硬件信息反馈至所述管理服务器的管理员。
  9. 一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中，所述处理器执行所述计算机程序时实现以下步骤：
    根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
    根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
    根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
    根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
    根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
    将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
    根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员。
  10. 根据权利要求9所述的计算机设备，其中，所述报错等级匹配规则包括加权值计算公式及多个报错等级匹配区间，所述根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级，包括：
    根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值；
    根据所述加权值对每一所述历史报错信息的数据量进行加权计算得到与每一所述历史报错信息对应的加权数据量；
    根据每一所述历史报错信息的加权数据量所属的报错等级匹配区间确定与每一所述历史报错信息对应的报错等级。
  11. 根据权利要求10所述的计算机设备，其中，所述根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值之前，还包括：
    根据预置的数据清洗规则对所述历史报错信息表所包含的无效信息进行数据清洗，得到去除无效信息后的历史报错信息表。
  12. 根据权利要求9所述的计算机设备，其中，所述根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息，包括：
    根据所述信息量化规则中的量化项目从每一所述历史报错信息中获取对应的项目属性信息；
    根据每一所述量化项目的项目规则对每一所述历史报错信息的项目属性信息分别进行量化得到对应的报错量化信息。
  13. 根据权利要求9所述的计算机设备，其中，所述模型训练规则包括损失值计算公式、梯度计算公式及损失阈值，所述根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型，包括：
    将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息；
    根据所述损失值计算公式计算得到所述报错量化信息的报错等级与所述模型输出信息之间的损失值；
    判断所述损失值是否小于所述损失阈值；
    若所述损失值不小于所述损失阈值，根据所述梯度计算公式及所述损失值计算得到所述状态检测模型中每一参数的更新值并对每一所述参数的参数值进行更新，返回执行所述将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息的步骤；
    若所述损失值小于所述损失阈值，将所述状态检测模型确定为训练后的状态检测模型。
  14. 根据权利要求9所述的计算机设备，其中，所述硬件检测信息为磁带检测信息，所述硬件设备为存储磁带，所述将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级，包括：
    根据所述信息量化规则对每一所述磁带检测信息进行量化得到对应的磁带检测量化信息；
    将所述磁带检测量化信息输入所述状态检测模型，得到与每一磁带检测量化信息对应的检测输出信息；
    将检测输出信息中输出节点值最大的一个输出节点对应的报错等级确定为对应的状态等级。
  15. 根据权利要求9所述的计算机设备，其中，所述根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理服务器的管理员之后，还包括：
    将所述异常硬件信息同步上传至区块链进行存储。
  16. 一种计算机可读存储介质，其中，所述计算机可读存储介质存储有计算机程序，当所述计算机程序被处理器执行时实现以下操作：
    根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级；
    根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息；
    根据所述报错等级匹配规则及所述信息量化规则生成对应的状态检测模型；
    根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型；
    根据预置的检测周期对每一所述备份终端所包含的硬件设备进行周期性检测，以获取每一所述硬件设备的硬件检测信息；
    将当前检测时间点获取到的所述硬件检测信息输入所述状态检测模型以获取每一所述硬件设备对应的状态等级；
    根据所述硬件设备的状态等级从所述硬件检测信息中获取异常硬件信息反馈至所述管理 服务器的管理员。
  17. 根据权利要求16所述的计算机可读存储介质，其中，所述报错等级匹配规则包括加权值计算公式及多个报错等级匹配区间，所述根据预置的报错等级匹配规则对预存的历史报错信息表中包含的历史报错信息进行匹配以获取每一所述历史报错信息的报错等级，包括：
    根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值；
    根据所述加权值对每一所述历史报错信息的数据量进行加权计算得到与每一所述历史报错信息对应的加权数据量；
    根据每一所述历史报错信息的加权数据量所属的报错等级匹配区间确定与每一所述历史报错信息对应的报错等级。
  18. 根据权利要求17所述的计算机可读存储介质，其中，所述根据所述加权值计算公式对每一所述历史报错信息的重要信息比例及差错间隔时间进行计算以得到每一所述历史报错信息的加权值之前，还包括：
    根据预置的数据清洗规则对所述历史报错信息表所包含的无效信息进行数据清洗，得到去除无效信息后的历史报错信息表。
  19. 根据权利要求16所述的计算机可读存储介质，其中，所述根据预置的信息量化规则对所述历史报错信息表中的每一所述历史报错信息进行量化得到报错量化信息，包括：
    根据所述信息量化规则中的量化项目从每一所述历史报错信息中获取对应的项目属性信息；
    根据每一所述量化项目的项目规则对每一所述历史报错信息的项目属性信息分别进行量化得到对应的报错量化信息。
  20. 根据权利要求16所述的计算机可读存储介质，其中，所述模型训练规则包括损失值计算公式、梯度计算公式及损失阈值，所述根据预置的模型训练规则及每一所述历史报错信息的报错量化信息、报错等级对所述状态检测模型进行迭代训练，得到训练后的状态检测模型，包括：
    将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息；
    根据所述损失值计算公式计算得到所述报错量化信息的报错等级与所述模型输出信息之间的损失值；
    判断所述损失值是否小于所述损失阈值；
    若所述损失值不小于所述损失阈值，根据所述梯度计算公式及所述损失值计算得到所述状态检测模型中每一参数的更新值并对每一所述参数的参数值进行更新，返回执行所述将一条所述报错量化信息输入所述状态检测模型以获取对应的模型输出信息的步骤；
    若所述损失值小于所述损失阈值，将所述状态检测模型确定为训练后的状态检测模型。
PCT/CN2021/083760 2020-12-25 2021-03-30 硬件状态检测方法、装置、计算机设备及存储介质 WO2022134353A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011561806.XA CN112652351A (zh) 2020-12-25 2020-12-25 硬件状态检测方法、装置、计算机设备及存储介质
CN202011561806.X 2020-12-25

Publications (1)

Publication Number Publication Date
WO2022134353A1 true WO2022134353A1 (zh) 2022-06-30

Family

ID=75362985

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/083760 WO2022134353A1 (zh) 2020-12-25 2021-03-30 硬件状态检测方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN112652351A (zh)
WO (1) WO2022134353A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116760434A (zh) * 2023-06-15 2023-09-15 深圳市泰士特线缆有限公司 通信线缆的智能检测方法、装置、设备及介质
CN118130957A (zh) * 2024-05-08 2024-06-04 吉林大学 一种电气设备自动化巡检系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113360179A (zh) * 2021-06-24 2021-09-07 平安普惠企业管理有限公司 基于Android系统的相机API的适配方法、装置、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107025153A (zh) * 2016-01-29 2017-08-08 阿里巴巴集团控股有限公司 磁盘的故障预测方法和装置
CN109739739A (zh) * 2018-12-28 2019-05-10 中兴通讯股份有限公司 磁盘故障的预测方法、设备及存储介质
CN110347547A (zh) * 2019-05-27 2019-10-18 中国平安人寿保险股份有限公司 基于深度学习的日志异常检测方法、装置、终端及介质
US20200250897A1 (en) * 2019-01-31 2020-08-06 Marketech International Corp. Intelligent prognostics and health management system and method
US20200250584A1 (en) * 2019-01-31 2020-08-06 Marketech International Corp. Modeling method for smart prognostics and health management system and computer program product thereof

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107025153A (zh) * 2016-01-29 2017-08-08 阿里巴巴集团控股有限公司 磁盘的故障预测方法和装置
CN109739739A (zh) * 2018-12-28 2019-05-10 中兴通讯股份有限公司 磁盘故障的预测方法、设备及存储介质
US20200250897A1 (en) * 2019-01-31 2020-08-06 Marketech International Corp. Intelligent prognostics and health management system and method
US20200250584A1 (en) * 2019-01-31 2020-08-06 Marketech International Corp. Modeling method for smart prognostics and health management system and computer program product thereof
CN110347547A (zh) * 2019-05-27 2019-10-18 中国平安人寿保险股份有限公司 基于深度学习的日志异常检测方法、装置、终端及介质

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116760434A (zh) * 2023-06-15 2023-09-15 深圳市泰士特线缆有限公司 通信线缆的智能检测方法、装置、设备及介质
CN116760434B (zh) * 2023-06-15 2024-01-26 深圳市泰士特线缆有限公司 通信线缆的智能检测方法、装置、设备及介质
CN118130957A (zh) * 2024-05-08 2024-06-04 吉林大学 一种电气设备自动化巡检系统

Also Published As

Publication number Publication date
CN112652351A (zh) 2021-04-13

Similar Documents

Publication Publication Date Title
CN103513983B (zh) 用于预测性警报阈值确定工具的方法和系统
Park et al. Forward-looking serial intervals correctly link epidemic growth to reproduction numbers
US8977581B1 (en) Data mining technique with diversity promotion
Häggström Data‐driven confounder selection via Markov and Bayesian networks
JP2021518024A (ja) 機械学習アルゴリズムのためのデータを生成する方法、システム
US10963807B2 (en) Social collaboration in probabilistic prediction
US7409604B2 (en) Determination of related failure events in a multi-node system
WO2022134353A1 (zh) 硬件状态检测方法、装置、计算机设备及存储介质
WO2021068513A1 (zh) 异常对象识别方法、装置、介质及电子设备
CN112882796A (zh) 异常根因分析方法和装置，及存储介质
US11146580B2 (en) Script and command line exploitation detection
CN110679114B (zh) 一种估计数据对象可删除性的方法
JP7294369B2 (ja) 情報処理に用いられる方法、装置、電子機器及びプログラム
George-Williams et al. Extending the survival signature paradigm to complex systems with non-repairable dependent failures
KR20230031889A (ko) 네트워크 토폴로지에서의 이상 탐지
CN109783459A (zh) 从日志中提取数据的方法、装置及计算机可读存储介质
CN114443639A (zh) 处理数据表及自动训练机器学习模型的方法和系统
CN114840531B (zh) 基于血缘关系的数据模型重构方法、装置、设备及介质
US9397921B2 (en) Method and system for signal categorization for monitoring and detecting health changes in a database system
WO2021098265A1 (zh) 缺失信息预测方法、装置、计算机设备及存储介质
CN113192627A (zh) 一种基于病人与疾病二部图的再入院预测方法及系统
US20230236590A1 (en) Method and system for providing maintenance service for recording medium included in electronic device
CN112416896A (zh) 数据异常的报警方法和装置、存储介质、电子装置
Arslan et al. On the distribution modeling of heavy-tailed disk failure lifetime in big data centers
CN117170915A (zh) 数据中心设备故障预测方法、装置和计算机设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21908365

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21908365

Country of ref document: EP

Kind code of ref document: A1