A UEFI-based hard disk monitoring method, device, equipment and medium

Abstract

The present invention relates to the field of storage technology, and in particular to a UEFI-based hard disk monitoring method, device, equipment and medium. The method comprises: utilizing UEFI to perform the following operations: collecting and recording hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters; inputting the short-cycle hard disk parameters into a pre-trained SVM model to obtain a normal or abnormal classification label; inputting the long-cycle hard disk parameters into a pre-trained RNN model to obtain a predicted value; calculating a prediction error rate based on the short-cycle hard disk parameters and the predicted value; and issuing an early warning based on the classification label and the prediction error rate. The solution of the present invention utilizes both SVM and RNN models to analyze sensor parameters, and can immediately detect whether the hard disk has an abnormality, and issue an early warning and prompt countermeasures, which greatly improves the hard disk storage security problem of large-scale data centers.

Description

Technical Field

[0001] The present invention relates to the field of storage technology, and in particular to a UEFI-based hard disk monitoring method, device, equipment and medium. Background Art

[0002] Since the beginning of the 21st century, countries around the world have been vigorously developing their digital economies. my country has also entered an era of large-scale infrastructure construction, and hyperscale data centers have experienced record growth. In this context, data storage security is crucial. As a data storage medium, SAS hard drives, while compatible with SCSI and SATA hard drives, offer superior transmission performance and scalability. They have become the primary hard drive type used in enterprise servers, and the detection and early warning of their normal operation play a vital role in data storage security.

[0003] The current mainstream hard drive status detection method is the method proposed in the last century whereby the hard drive autonomously collects SMART information and analyzes hard drive anomalies based on this information. However, with the increasing emergence of large-scale equipment and large data centers, the requirements for autonomous hard drive information collection are becoming increasingly diverse. For example, the detection of more hard drive parameters and indicators related to time and space environments is also promising. Predicting the hard drive's service life and possible problems based on the collected information is also a promising development. Summary of the Invention

[0004] In view of this, in order to solve and prevent abnormal situations of SAS hard drives used in large-scale data centers, the present invention provides a UEFI-based hard drive monitoring method, device, equipment and medium.

[0005] According to a first aspect of the present invention, a UEFI-based hard disk monitoring method is provided, the method comprising:

[0006] Collect and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters;

[0007] Input the short-cycle hard disk parameters into the pre-trained SVM model to obtain the normal or abnormal classification label;

[0008] Input the long-cycle hard disk parameters into the pre-trained RNN model to obtain the predicted values;

[0009] Calculating a prediction error rate based on the short-cycle hard disk parameter and the predicted value;

[0010] An early warning is issued based on the classification label and the prediction error rate.

[0011] In some embodiments, collecting and recording hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters includes:

[0012] After power-on, UEFI starts and searches for the hard disk;

[0013] Determine whether each hard disk found has a sensor acquisition function;

[0014] In response to the hard disk having a sensing acquisition function, a storage space is reserved on the hard disk, and the read current hard disk sensing parameters or the input hard disk sensing parameters are stored in the reserved storage space to obtain short-cycle hard disk parameters;

[0015] The historical hard disk sensor parameters of a preset number of days before the current day are obtained from the reserved space and processed to obtain the long-period hard disk parameters.

[0016] In some embodiments, obtaining historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and processing them to obtain long-period hard disk parameters includes:

[0017] Get the hard disk sensor parameters for a preset number of days in units of days;

[0018] Calculate the variance and mean of hard disk sensor parameters every day;

[0019] In response to the variance not exceeding the preset variance, taking the mean as the data for the corresponding day;

[0020] In response to the variance exceeding a preset variance, taking the mean and the variance as data for the corresponding day;

[0021] Summarize the data of all preset days to obtain the long-term hard disk parameters.

[0022] In some embodiments, the hard disk sensing parameters include SMART information, humidity, air pressure, instantaneous value of electromagnetic intensity, limit value, and unwritten and written sector status parameters.

[0023] In some embodiments, the short-cycle hard disk parameters also include hard disk secondary parameters calculated based on the hard disk sensor parameters, wherein the hard disk secondary parameters include hard disk service life, motor service life, and head wear degree.

[0024] In some embodiments, the hard disk service life is generated according to the following formula 1:

[0025] TN Servicelife = humidity T humidity +δ temperature T temperature + poweronhours T poweronhours Formula 1;

[0026] Among them, TN Servicelife is the service life parameter, Thumidity is the humidity parameter, Tt emperature is the temperature parameter, T poweronhours is the power-on time parameter, δ humidity is the key proportion of humidity, δ temperature is the critical temperature ratio, δ poweronhours It is the critical proportion of power-on time.

[0027] In some embodiments, calculating the prediction error rate based on the short-cycle hard disk parameter and the predicted value includes:

[0028] Calculating the absolute value of the difference between the short-cycle hard disk parameter and the predicted value;

[0029] The ratio of the absolute value of the difference to the short-cycle hard disk parameter is used as the prediction error rate.

[0030] In some embodiments, issuing an early warning based on the classification label and the prediction error rate includes:

[0031] In response to the classification label being an abnormal label and the prediction error rate being less than a first preset value, a first level warning is issued;

[0032] In response to the classification label being a normal label and the prediction error rate being greater than or equal to a first preset value and less than or equal to a second preset value, a second-level warning is issued;

[0033] In response to the classification label being a normal label and the prediction error rate being greater than a second preset value, a third-level warning is issued;

[0034] In response to the classification label being an abnormal label and the prediction error rate being greater than a second preset value, a fourth-level warning is issued.

[0035] In some embodiments, the method further comprises:

[0036] In response to issuing an alert of any level, determining abnormal parameters based on the output of the pre-trained SVM model and / or the output of the pre-trained RNN model;

[0037] According to the determined abnormal parameters, a processing countermeasure is queried from the countermeasure library.

[0038] According to a second aspect of the present invention, a UEFI-based hard disk monitoring device is provided, in which the following modules are set:

[0039] an acquisition module configured to acquire and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters;

[0040] a classification module configured to input the short-cycle hard disk parameters into a pre-trained SVM model to obtain a normal or abnormal classification label;

[0041] A prediction module configured to input long-cycle hard disk parameters into a pre-trained RNN model to obtain predicted values;

[0042] a calculation module configured to calculate a prediction error rate based on the short-cycle hard disk parameter and the predicted value;

[0043] An early warning module is configured to issue an early warning based on the classification label and the prediction error rate.

[0044] According to a third aspect of the present invention, a computer device is further provided, the computer device comprising:

[0045] at least one processor; and

[0046] The memory stores a computer program that can be run on the processor, and the processor executes the aforementioned UEFI hard disk monitoring method when executing the program.

[0047] According to a fourth aspect of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the aforementioned UEFI hard disk monitoring method is executed.

[0048] The above-mentioned UEFI-based hard disk monitoring method connects to the hard disk through UEFI, collects and stores the hard disk sensor parameters, and then uses the SVM and RNN models to analyze the sensor parameters. It can immediately detect whether there are any abnormalities in the hard disk, and issue early warnings and prompt countermeasures, which greatly improves the hard disk storage security issues in large-scale data centers.

[0049] In addition, the present invention also provides a UEFI-based hard disk monitoring device, a computer device and a computer-readable storage medium, which can also achieve the above technical effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 A flowchart of a UEFI-based hard disk monitoring method provided by one embodiment of the present invention;

[0052] Figure 2 A schematic diagram of UEFI functions using the method of the present invention is provided as another embodiment of the present invention;

[0053] Figure 3 A short-cycle parameter collection flow chart provided in another embodiment of the present invention;

[0054] Figure 4 A schematic diagram of long-period parameter acquisition provided by another embodiment of the present invention;

[0055] Figure 5 A schematic structural diagram of a UEFI-based hard disk monitoring device provided in another embodiment of the present invention;

[0056] Figure 6 This is a diagram of the internal structure of a computer device in another embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0058] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. Subsequent embodiments will not explain this one by one.

[0059] In one embodiment, please refer to Figure 1 As shown, the present invention provides a UEFI-based hard disk monitoring method 100. Specifically, the method includes the following steps:

[0060] Step 101 : Collect and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters.

[0061] In this embodiment, the short-cycle hard disk parameter can be one day's data or one week's data, and the specific number of days can be set according to the prediction accuracy requirements or user needs. The long-cycle hard disk parameter can be one month's data or multiple months' data, and the specific number of months included can be set according to the prediction accuracy requirements or user needs. For example, data for one month or half a year can be collected. In the specific implementation process, of course, users can also use days instead of weeks or months to define long and short cycles. For example, the long cycle can be 100 days, the short cycle can be 5 days, and so on.

[0062] In step 102 , the short-cycle hard disk parameters are input into a pre-trained SVM model to obtain a normal or abnormal classification label; specifically, the abnormal label model may output 0, and the normal label model may output 1.

[0063] In step 103, the long-period hard disk parameters are input into the pre-trained RNN model to obtain a predicted value; specifically, the pre-trained RNN model can predict one day's data in thirty days, one day's data in half a year, and so on.

[0064] Step 104, calculate the prediction error rate based on the short-cycle hard disk parameters and the predicted values; during implementation, if the short cycle is multiple days, the long-cycle hard disk parameter predicted value corresponds to the last day's data of the short-cycle hard disk parameters.

[0065] Step 105: issuing an early warning based on the classification label and the prediction error rate.

[0066] The above-mentioned UEFI-based hard disk monitoring method connects to the hard disk through UEFI, collects and stores the hard disk sensor parameters, and then uses the SVM and RNN models to analyze the sensor parameters. It can immediately detect whether there are any abnormalities in the hard disk, and issue early warnings and prompt countermeasures, which greatly improves the hard disk storage security issues in large-scale data centers.

[0067] In some embodiments, step 101, collecting and recording hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters, includes:

[0068] After power-on, UEFI starts and searches for the hard disk;

[0069] Determine whether each hard disk found has a sensor acquisition function;

[0070] In response to the hard disk having a sensing acquisition function, a storage space is reserved on the hard disk, and the read current hard disk sensing parameters or the input hard disk sensing parameters are stored in the reserved storage space to obtain short-cycle hard disk parameters;

[0071] The historical hard disk sensor parameters of a preset number of days before the current day are obtained from the reserved space and processed to obtain the long-period hard disk parameters.

[0072] In some embodiments, obtaining historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and processing them to obtain long-period hard disk parameters includes:

[0073] Get the hard disk sensor parameters for a preset number of days in units of days;

[0074] Calculate the variance and mean of hard disk sensor parameters every day;

[0075] In response to the variance not exceeding the preset variance, taking the mean as the data for the corresponding day;

[0076] In response to the variance exceeding a preset variance, taking the mean and the variance as data for the corresponding day;

[0077] Summarize the data of all preset days to obtain the long-term hard disk parameters.

[0078] In some embodiments, the hard disk sensing parameters include SMART information, humidity, air pressure, instantaneous value of electromagnetic intensity, limit value, and unwritten and written sector status parameters.

[0079] In some embodiments, the short-cycle hard disk parameters also include hard disk secondary parameters calculated based on the hard disk sensor parameters, wherein the hard disk secondary parameters include hard disk service life, motor service life, and head wear degree.

[0080] In some implementations, the hard disk service life is generated according to the following formula 1:

[0081] TN Servicelife =δ humidity T humidity +δ temperature T temperature +δ poweronhours T poweronhours Formula 1;

[0082] Among them, TN_Servicelife is the service life parameter, T_humidity is the humidity parameter, T_temperature is the temperature parameter, T_poweronhours is the power-on time parameter, δ_humidity is the humidity critical ratio, δ_temperature is the temperature critical ratio, and δ_poweronhours is the power-on time critical ratio.

[0083] It should be noted that in a specific implementation process, the motor service life and the degree of head wear have other related parameters such as unwritten and written sector status parameters, etc., which can be set by referring to the above formula to obtain the corresponding parameters.

[0084] In some embodiments, the aforementioned step 104, calculating the prediction error rate based on the short-cycle hard disk parameter and the predicted value, includes:

[0085] Calculating the absolute value of the difference between the short-cycle hard disk parameter and the predicted value;

[0086] The ratio of the absolute value of the difference to the short-cycle hard disk parameter is used as the prediction error rate.

[0087] In some embodiments, the aforementioned step 105, issuing an early warning based on the classification label and the prediction error rate, includes:

[0088] In response to the classification label being an abnormal label and the prediction error rate being less than a first preset value, a first level warning is issued;

[0089] In response to the classification label being a normal label and the prediction error rate being greater than or equal to a first preset value and less than or equal to a second preset value, a second-level warning is issued;

[0090] In response to the classification label being a normal label and the prediction error rate being greater than a second preset value, a third-level warning is issued;

[0091] In response to the classification label being an abnormal label and the prediction error rate being greater than a second preset value, a fourth-level warning is issued.

[0092] In some embodiments, the method further comprises:

[0093] In response to issuing an alert of any level, determining abnormal parameters based on the output of the pre-trained SVM model and / or the output of the pre-trained RNN model;

[0094] According to the determined abnormal parameters, a processing countermeasure is queried from the countermeasure library.

[0095] In another embodiment, in order to facilitate understanding of the solution of the present invention, please refer to Figure 2 As shown, this embodiment takes the SAS hard disk of a data center as an example to explain the solution of the present invention in detail. This embodiment provides a hard disk monitoring method for UEFI applied to the data center. This method mainly adds the following three functions to UEFI: hard disk sensor parameter collection and recording function, hard disk sensor parameter analysis function, and hard disk sensor parameter decoding, display and early warning function. The following will combine the implementation methods of the above three functions:

[0096] Part 1: Hard disk sensor parameter collection and recording function; Hard disk sensor parameter collection and recording function includes short-term hard disk sensor parameter collection and long-term hard disk sensor parameter collection;

[0097] The process of collecting and recording short-term hard disk sensor parameters is as follows: Figure 3 As shown, after the system is powered on, the firmware starts and executes until the UEFI startup service. It traverses the Protocol, Target, and LUN in sequence according to the Guid of ScsiPassThruProtocol, searches for all SAS hard disks connected to the local machine, and determines the SAS hard disk that has the function of collecting the required sensor parameters.

[0098] When power is supplied to a SAS hard drive, the Passthru Protocol's read / write commands are automatically applied to the drive, allowing the SAS hard drive sensor to read the drive's sensor parameters and then store them in the SAS hard drive's system reserved area. This SAS hard drive's system reserved area should have sufficient space reserved for long-term storage of drive sensor parameters, with 64 bits of data space allocated for each parameter. This SAS hard drive also supports the Passthru Protocol's Logsense and Logselect commands for reading and writing drive sensor parameters. The pagecode is set within the range of 0x30h-0x3eh, and the subpagecode is set within the range of 0x00h-0xfeh. All drive sensor parameters are user-readable and user-writable. Users can update drive sensor parameters during this process, and the location and number of updates to the drive sensor parameter data are recorded in the system reserved area.

[0099] Hard disk sensor parameters include but are not limited to the following information:

[0100] 1.SMART information proposed by Small Form Factor

[0101] 2. External environmental information such as instantaneous value, limit value, average value, etc. of humidity, air pressure, and electromagnetic intensity

[0102] 3. Status parameters of unwritten and written sectors, such as command times, logical block read and write times, etc.

[0103] The hard disk recorded sensor parameters include but are not limited to the following information:

[0104] 1. Hard disk sensor parameters collection

[0105] 2. Hard disk secondary parameters

[0106] 3. Hard disk parameter error record

[0107] 4. Graded warning information and countermeasures

[0108] 5. Hard disk sensor parameter modification record

[0109] During use, SAS hard drives are tested for sensor parameters every two hours, meaning they are tested 12 times a day. Hard drive sensor parameters must be recorded daily for the week preceding the day of use.

[0110] The long-term hard disk sensor parameter recording process is as follows Figure 4 As shown:

[0111] To optimize recorded data size, if the normalized variance of a parameter is less than 0.1 within a day, only the mean value for that day is recorded. If the normalized variance of a drive sensor parameter exceeds 0.1 within a day, the mean ± variance for that day is recorded, and a drive parameter error record is added. Drive sensor parameter data for the six months preceding the day of use is recorded.

[0112] All hard disk sensor parameter records can be read and written by the user. The user can modify the hard disk sensor parameter data in this section, and at the same time record the location and number of hard disk sensor parameter data modifications in the system reserved area.

[0113] Part II: Hard Drive Sensor Parameter Analysis; This is divided into two parts: short-term hard drive sensor parameter analysis and long-term hard drive sensor parameter analysis. First, a sufficient number of SAS hard drives (n>500) were collected under different environments for at least six months to serve as training data for the two models.

[0114] (1) Short-term hard disk sensor parameter analysis

[0115] At the same time, secondary parameters are set up to address the hard drive damage problem caused by multiple hard drive sensor parameters, such as hard drive life, motor life, head wear, etc. Here, it is assumed that the life of SAS hard drives is related to three parameters, namely temperature, humidity, and usage time. The calculation formula for the SAS life parameter is: TN Servicelife =δ humidity T humidity +δ temperature T temperature +δ poweronhours T poweronhours ;

[0116] TN_Servicelife is the service life parameter, T_humidity is the humidity parameter, T_temperature is the temperature parameter, T_poweronhours is the power-on time parameter, δ_humidity is the humidity critical ratio, δ_temperature is the temperature critical ratio, and δ_poweronhours is the power-on time critical ratio. Adjust the critical ratio based on the impact of each parameter on the problem. The greater the impact, the larger the critical ratio.

[0117] To design an SVM model to determine whether a hard drive is in normal use, the collected parameters and secondary parameters are first normalized. All parameters recorded in a single session and their secondary parameters are then set as an input vector. Short-term hard drive sensor parameter analysis uses nearly a week's worth of parameters, or at least 42,000 input vectors. 30,000 input vectors are used as training data to train the SVM model. The SVM uses the Gaussian kernel function: Among them For this model, the penalty factor C is set to 0.01, 0.1, 1, 10, 100, and 1000, and gamma is set to 0.01, 0.02, ..., 1. After training, the model is evaluated using 12,000 input vectors and the parameters C and gamma are determined. The results are evaluated using the F1 score: The precision rate is: ), recall rate: Ultimately, C and gamma should be selected so that the F1 score exceeds 0.95.

[0118] (2) Long-term hard disk sensor parameter analysis

[0119] A unidirectional many-to-one RNN function was designed to determine whether the hard drive was in normal use. Long-term hard drive sensor parameter analysis used nearly six months of parameters. The RNN had at least 180 inputs for each parameter per hard drive. 90% of the data for each parameter was used for training, and the remaining data was used for prediction. The time series length was set to 150; the input data dimension was 2, consisting of the parameter and the number of days since today. The middle layer consisted of two LSTM layers with a batch size of 30 and an output data dimension of 1. After training, the model was evaluated using the prediction data. The error between the predicted and true results should be less than 5%. All hard drive sensor parameter calculation formulas are user-readable and user-writable. Users can update the hard drive sensor parameter calculation formulas in this section, and the location and number of modified data are recorded in a system reserved area.

[0120] Part 3: Hard disk sensor parameter decoding, display, and warning function: When new data is collected, the hard disk sensor parameter decoding, display, and warning function are enabled. First, the data is determined to see if it passes the SVM model established by short-term hard disk sensor parameter analysis. Second, the prediction error e is calculated compared to the RNN model established by long-term hard disk sensor parameter analysis. Then, warnings are issued based on different situations:

[0121] Warning 1: If the data fails to pass the SVM model (i.e., the model classifies a certain parameter as abnormal), the warning light will be on, the warning information will be recorded, and countermeasures will be suggested.

[0122] Warning 2: The data passes the SVM model (i.e., the model classifies a certain parameter as abnormal) or the SVM model (i.e., the model classifies a certain parameter as normal), but the e-value is between 5% and 10%. The warning light is on, the warning information is recorded, and countermeasures are suggested.

[0123] Warning 3: The data passes the SVM model, but the e-value exceeds 10%. The warning light is on, the warning information is recorded, and countermeasures are suggested.

[0124] Warning 4: The data fails the SVM model and the e-value exceeds 10%. The warning light is on, the warning information is recorded, and countermeasures are suggested.

[0125] All hard drive sensor parameters are readable and writable by the user. The user can update the hard drive sensor parameters after taking countermeasures in response to the early warning, and record the location and number of modified data in the system reserved area.

[0126] The UEFI-based hard disk monitoring method of this embodiment has the following beneficial effects: during the UEFI startup process, the SAS hard disk can be searched through UEFI, and the hard disk sensor parameters can be collected and stored. Then, the sensor parameters can be jointly analyzed by single parameters and multiple parameters. Whether there is an abnormality in the hard disk can be detected in the first time, and an early warning and countermeasures can be issued. It has extremely high flexibility and versatility.

[0127] In yet another embodiment, please combine Figure 5 As shown, this embodiment provides a UEFI-based hard disk monitoring device 200, in which the following modules are set:

[0128] The acquisition module 201 is configured to acquire and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters;

[0129] The classification module 202 is configured to input the short-cycle hard disk parameters into a pre-trained SVM model to obtain a normal or abnormal classification label;

[0130] The prediction module 203 is configured to input the long-cycle hard disk parameters into the pre-trained RNN model to obtain a predicted value;

[0131] A calculation module 204 is configured to calculate a prediction error rate based on the short-cycle hard disk parameter and the predicted value;

[0132] The early warning module 205 is configured to issue an early warning based on the classification label and the prediction error rate.

[0133] The above-mentioned UEFI-based hard disk monitoring device connects to the hard disk through UEFI, collects and stores hard disk sensor parameters, and then uses the SVM and RNN models to analyze the sensor parameters. It can immediately detect whether there is any abnormality in the hard disk, and issue an early warning and suggest countermeasures, which greatly improves the hard disk storage security issues in large-scale data centers.

[0134] In some embodiments, the acquisition module 201 is further configured to:

[0135] After power-on, UEFI starts and searches for the hard disk;

[0136] Determine whether each hard disk found has a sensor acquisition function;

[0137] In response to the hard disk having a sensing acquisition function, a storage space is reserved on the hard disk, and the read current hard disk sensing parameters or the input hard disk sensing parameters are stored in the reserved storage space to obtain short-cycle hard disk parameters;

[0138] The historical hard disk sensor parameters of a preset number of days before the current day are obtained from the reserved space and processed to obtain the long-period hard disk parameters.

[0139] In some embodiments, the acquisition module 201 is further configured to:

[0140] Get the hard disk sensor parameters for a preset number of days in units of days;

[0141] Calculate the variance and mean of hard disk sensor parameters every day;

[0142] In response to the variance not exceeding the preset variance, taking the mean as the data for the corresponding day;

[0143] In response to the variance exceeding a preset variance, taking the mean and the variance as data for the corresponding day;

[0144] Summarize the data of all preset days to obtain the long-term hard disk parameters.

[0145] In some embodiments, the hard disk sensing parameters include SMART information, humidity, air pressure, instantaneous value of electromagnetic intensity, limit value, and unwritten and written sector status parameters.

[0146] In some embodiments, the short-cycle hard disk parameters also include hard disk secondary parameters calculated based on the hard disk sensor parameters, wherein the hard disk secondary parameters include hard disk service life, motor service life, and head wear degree.

[0147] In some embodiments, the hard disk service life is generated according to Formula 2;

[0148] TN Servicelife =δ humidity T humidity +δ temperature T temperature +δ poweronhours T poweronhours Formula 2;

[0149] Among them, TN_Servicelife is the service life parameter, T_humidity is the humidity parameter, T_temperature is the temperature parameter, T_poweronhours is the power-on time parameter, δ_humidity is the humidity critical ratio, δ_temperature is the temperature critical ratio, and δ_poweronhours is the power-on time critical ratio.

[0150] In some embodiments, the calculation module 204 is further configured to:

[0151] Calculating the absolute value of the difference between the short-cycle hard disk parameter and the predicted value;

[0152] The ratio of the absolute value of the difference to the short-cycle hard disk parameter is used as the prediction error rate.

[0153] In some embodiments, the early warning module 205 is further configured to:

[0154] In response to the classification label being an abnormal label and the prediction error rate being less than a first preset value, a first level warning is issued;

[0155] If the classification label is a normal label and the prediction error rate is greater than or equal to the first preset value and less than or equal to the second preset value, a level 2 warning is issued;

[0156] If the classification label is a normal label and the prediction error rate is greater than the second preset value, a third-level warning is issued;

[0157] In response to the classification label being an abnormal label and the prediction error rate being greater than a second preset value, a fourth-level warning is issued.

[0158] In some embodiments, the apparatus further comprises a module configured to perform the following steps:

[0159] In response to issuing an alert of any level, determining abnormal parameters based on the output of the pre-trained SVM model and / or the output of the pre-trained RNN model;

[0160] According to the determined abnormal parameters, a processing countermeasure is queried from the countermeasure library.

[0161] It should be noted that the specific definitions of the UEFI hard disk monitoring device can be found in the definitions of the UEFI hard disk monitoring method above and will not be repeated here. The various modules in the above-mentioned UEFI hard disk monitoring device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each of the above modules.

[0162] According to another aspect of the present invention, a computer device is provided. The computer device may be a server. Figure 6As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the UEFI hard disk monitoring method described above is implemented. Specifically, the method includes the following steps:

[0163] Collect and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters;

[0164] Input the short-cycle hard disk parameters into the pre-trained SVM model to obtain the normal or abnormal classification label;

[0165] Input the long-cycle hard disk parameters into the pre-trained RNN model to obtain the predicted values;

[0166] Calculating a prediction error rate based on the short-cycle hard disk parameter and the predicted value;

[0167] An early warning is issued based on the classification label and the prediction error rate.

[0168] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the UEFI hard disk monitoring method described above is implemented. Specifically, the method includes performing the following steps:

[0169] Collect and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters;

[0170] Input the short-cycle hard disk parameters into the pre-trained SVM model to obtain the normal or abnormal classification label;

[0171] Input the long-cycle hard disk parameters into the pre-trained RNN model to obtain the predicted values;

[0172] Calculating a prediction error rate based on the short-cycle hard disk parameter and the predicted value;

[0173] An early warning is issued based on the classification label and the prediction error rate.

[0174] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0175] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0176] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A UEFI-based hard disk monitoring method, characterized in that: The method includes utilizing UEFI to perform the following operations: Collect and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters; Input the short-cycle hard disk parameters into the pre-trained SVM model to obtain the normal or abnormal classification label; Input the long-cycle hard disk parameters into the pre-trained RNN model to obtain the predicted values; Calculating a prediction error rate based on the short-cycle hard disk parameter and the predicted value; Producing an early warning based on the classification label and the prediction error rate; Among them, collecting and recording hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters includes: after power-on, UEFI starts up and traverses to search the hard disk; judging whether each hard disk found has a sensing acquisition function; in response to the hard disk having the sensing acquisition function, reserving storage space on the hard disk, storing the read current hard disk sensor parameters or the input hard disk sensor parameters in the reserved storage space to obtain short-cycle hard disk parameters; obtaining historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and processing them to obtain long-cycle hard disk parameters; wherein, obtaining historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and processing them to obtain long-cycle hard disk parameters includes: obtaining hard disk sensor parameters of a preset number of days in units of days; calculating the variance and mean of the hard disk sensor parameters of each day; in response to the variance not exceeding the preset variance, taking the mean as the data for the corresponding day; in response to the variance exceeding the preset variance, taking the mean and variance as the data for the corresponding day; summarizing the data of all preset days to obtain long-cycle hard disk parameters; Calculating the prediction error rate based on the short-cycle hard disk parameter and the predicted value includes: calculating the absolute value of the difference between the short-cycle hard disk parameter and the predicted value; and taking the ratio of the absolute value of the difference to the short-cycle hard disk parameter as the prediction error rate.

2. The UEFI-based hard disk monitoring method according to claim 1, wherein: The hard disk sensing parameters include SMART information, humidity, air pressure, instantaneous value of electromagnetic intensity, limit value, and unwritten and written sector status parameters.

3. The UEFI-based hard disk monitoring method according to claim 2, wherein: The short-cycle hard disk parameters also include hard disk secondary parameters calculated based on the hard disk sensor parameters, wherein the hard disk secondary parameters include hard disk service life, motor service life, and head wear degree.

4. The UEFI-based hard disk monitoring method according to claim 3, wherein: The hard disk service life is generated according to the following formula 1: Formula 1; in, is the service life parameter, is the humidity parameter, is the temperature parameter, is the power-on time parameter, is the key percentage of humidity, is the critical proportion of temperature, It is the critical proportion of power-on time.

5. The UEFI-based hard disk monitoring method according to claim 1, wherein: Producing an early warning based on the classification label and the prediction error rate, including: In response to the classification label being an abnormal label and the prediction error rate being less than a first preset value, a first level warning is issued; In response to the classification label being a normal label and the prediction error rate being greater than or equal to a first preset value and less than or equal to a second preset value, a second-level warning is issued; In response to the classification label being a normal label and the prediction error rate being greater than a second preset value, a third-level warning is issued; In response to the classification label being an abnormal label and the prediction error rate being greater than a second preset value, a fourth-level warning is issued.

6. The UEFI-based hard disk monitoring method according to claim 5, characterized in that: The method further comprises: In response to issuing an alert of any level, determining abnormal parameters based on the output of the pre-trained SVM model and / or the output of the pre-trained RNN model; According to the determined abnormal parameters, a processing countermeasure is queried from the countermeasure library.

7. A UEFI-based hard disk monitoring device, characterized in that: Set up the following modules in UEFI: an acquisition module configured to acquire and record hard disk parameters to obtain short-cycle hard disk parameters and long-cycle hard disk parameters; a classification module configured to input the short-cycle hard disk parameters into a pre-trained SVM model to obtain a normal or abnormal classification label; A prediction module configured to input long-cycle hard disk parameters into a pre-trained RNN model to obtain predicted values; a calculation module configured to calculate a prediction error rate based on the short-cycle hard disk parameter and the predicted value; An early warning module configured to issue an early warning based on the classification label and the prediction error rate Wherein, the acquisition module is further configured to start UEFI after power-on, traverse and search the hard disk; determine whether each hard disk found has a sensor acquisition function; in response to the hard disk having a sensor acquisition function, reserve storage space on the hard disk, store the read current hard disk sensor parameters or the input hard disk sensor parameters in the reserved storage space to obtain short-cycle hard disk parameters; obtain historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and obtain long-cycle hard disk parameters after processing; wherein, obtain historical hard disk sensor parameters of a preset number of days before the current day from the reserved space and obtain long-cycle hard disk parameters after processing, including: obtaining hard disk sensor parameters of a preset number of days in units of days; calculating the variance and mean of the hard disk sensor parameters of each day; in response to the variance not exceeding the preset variance, taking the mean as the data for the corresponding day; in response to the variance exceeding the preset variance, taking the mean and variance as the data for the corresponding day; summarizing the data of all preset days to obtain long-cycle hard disk parameters; The calculation module is further configured to calculate the absolute value of the difference between the short-cycle hard disk parameter and the predicted value; and use the ratio of the absolute value of the difference to the short-cycle hard disk parameter as the prediction error rate.

8. A computer device, characterized in that: include: at least one processor; as well as A memory storing a computer program executable in the processor, wherein the processor executes the method according to any one of claims 1 to 6 when executing the program.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is performed.

Images