Hard disk failure prediction method, electronic device, storage medium, and program product

Abstract

The application provides a hard disk fault prediction method, electronic equipment, a storage medium and a program product, which can be applied to the technical field of computers. The hard disk fault prediction method comprises the following steps: obtaining a parameter value of a hard disk from a running log of the hard disk; in the case that the parameter value does not exceed a pre-warning threshold corresponding to the parameter value, and it is determined according to a parameter type of the parameter value that trend pre-warning needs to be performed on the hard disk based on a change trend of the parameter value, determining a fault risk degree value of the hard disk according to a change rate of the parameter value in a predetermined time period; and predicting a pre-warning level of the hard disk according to the fault risk degree value.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, specifically to the field of fault processing technology, and more specifically to a hard disk fault prediction method, electronic equipment, storage medium and program product. Background Art

[0002] With the development of information technology, data storage has become increasingly important. As the primary data storage device, the stability and reliability of hard drives directly impact data security and the smooth operation of businesses. A hard drive failure can lead to data loss, business interruption, and other consequences, causing losses to businesses and users.

[0003] Currently, the analysis of hard drive failures is based solely on a single threshold, which results in rigid static rules and a single data dimension. This makes it difficult to cope with the complex and ever-changing situations that may occur in hard drives, making it difficult to make timely and accurate predictions of hard drive failures, thus reducing the accuracy of failure predictions. Summary of the Invention

[0004] In view of the above problems, the present invention provides a hard disk fault prediction method, electronic device, storage medium and program product that improve the accuracy of fault processing.

[0005] One aspect of the present invention provides a hard disk failure prediction method, comprising: obtaining a parameter value of a hard disk from an operation log of the hard disk; when the parameter value does not exceed a warning threshold value corresponding to the parameter value, and it is determined according to the parameter type of the parameter value that a trend warning for the hard disk needs to be performed based on a changing trend of the parameter value, determining a failure risk degree value of the hard disk according to a rate of change of the parameter value in a predetermined period; and predicting a warning level of the hard disk according to the failure risk degree value.

[0006] Another aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0007] Another aspect of the present invention further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the computer program or instructions are executed by a processor.

[0008] Another aspect of the present invention further provides a computer program product, comprising a computer program or instructions, which implement the steps of the above method when the computer program or instructions are executed by a processor.

[0009] According to an embodiment of the present invention, by obtaining hard drive parameter values ​​from the hard drive's operation log, if the parameter values ​​do not exceed a warning threshold and a warning is required for the hard drive based on the parameter value's changing trend, the hard drive's failure risk level is determined based on the parameter value's rate of change within a predetermined period. Furthermore, the hard drive's warning level is predicted based on the failure risk level. By incorporating the parameter value's rate of change within a predetermined period into the failure prediction process, a trend analysis of the parameter values ​​is performed, and a warning is issued for the hard drive based on this trend analysis. This avoids the limitations of untimely and inaccurate hard drive failure assessments caused by alerting the hard drive based on a single threshold, thereby improving the accuracy of hard drive failure prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0011] Figure 1 A diagram showing an application scenario of a hard disk failure prediction method according to an embodiment of the present invention is shown;

[0012] Figure 2 A flowchart of a hard disk failure prediction method according to an embodiment of the present invention is shown;

[0013] Figure 3 Shows an architecture diagram of a hard disk failure prediction system provided by an embodiment of the present invention;

[0014] Figure 4 A flowchart of a hard disk failure prediction method according to another embodiment of the present invention is shown;

[0015] Figure 5 It shows a structural block diagram of a hard disk failure prediction device according to an embodiment of the present invention;

[0016] Figure 6 A block diagram of an electronic device suitable for implementing a hard disk failure prediction method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concept of the present invention.

[0018] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the presence of the features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0019] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0020] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0021] With the increasing diversification of business scenarios, the continuous emergence of business models, and the continuous expansion of innovative businesses, there are extremely high requirements for real-time data processing, accurate assessment of business processing risks, and rapid business response. The data required for real-time data processing, accurate assessment of business processing risks, and rapid business response can all be stored on hard disk devices. A hard disk failure can result in data loss and business interruption, causing losses to business processing.

[0022] Currently, hard drive failure prediction methods have numerous limitations. For example, reliance on manual judgment leads to insufficient accuracy and timeliness in fault prediction. For example, simple threshold judgments make it difficult to provide early warning of failures based on dynamic parameter value trends, resulting in lags. For example, if a hard drive parameter value continues to increase over a short period of time, even if it does not reach the threshold, without early warning and intervention, errors, service interruptions, or even disk failures and downtime may occur. For another example, fault prediction methods based on machine learning and artificial intelligence (AI) can determine the hard drive failure warning time by inputting data into a trained deep belief network model. Based on the warning time and fault alarm level, hard drive failures can be predicted and warned. However, the parameter values ​​input into the deep belief network model lack analysis and warning of core hard drive parameters, resulting in reduced prediction dimensionality and accuracy. Fault prediction methods based on machine learning and AI lack comprehensiveness of parameter values, diversity of fault warning types, and consideration of complex environmental factors.

[0023] In general, current hard disk failure prediction methods have problems such as rigid static rules, single data dimensions, and weak model generalization capabilities. It is difficult to capture trend changes in parameters, and the model's prediction false alarm rate is high, making it difficult to identify failure risks in the early stages of a failure. In view of this, an embodiment of the present invention provides a hard disk failure prediction method that, in addition to taking into account the basic hard disk parameter values, also introduces the underlying core parameter values ​​of the hard disk, and reflects the hard disk basic parameter values ​​and the underlying core parameter values ​​of the hard disk from three perspectives: the core parameters of the entire system or device on which the hard disk is installed, the parameters of user applications, and the user's computer room environment. At the same time, it also combines multiple prediction processes to provide a more accurate, timely, and intelligent automatic fault prediction solution.

[0024] Figure 1 A diagram showing an application scenario of a hard disk failure prediction method according to an embodiment of the present invention is shown.

[0025] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or optical fiber cables, etc.

[0026] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, such as sending a hard drive failure prediction request or a hard drive failure prediction task, or receiving warning level information. In another embodiment, the hard drive failure prediction task can be automatically triggered at a scheduled time without user triggering. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as hard drive failure prediction applications, shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0027] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0028] The server 105 may be a server that provides various services, such as a background management server (for example only) that supports requests sent by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received requests and other data, and feedback the processing results (e.g., obtaining or generating warning levels, web pages, information, or data based on the requests) to the terminal devices.

[0029] It should be noted that the hard disk failure prediction method provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the hard disk failure prediction device provided in the embodiment of the present invention can generally be set in the server 105. The hard disk failure prediction method provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the hard disk failure prediction device provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0030] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0031] The following will be based on Figure 1 The scene described by Figures 2 to 4 A hard disk failure prediction method according to an embodiment of the present invention is described in detail.

[0032] Figure 2 A flowchart of a hard disk failure prediction method according to an embodiment of the present invention is shown.

[0033] like Figure 2 As shown, the hard disk failure prediction of this embodiment includes operations S210 to S230.

[0034] In operation S210 , a parameter value of the hard disk is obtained from an operation log of the hard disk.

[0035] In operation S220, when the parameter value does not exceed the warning threshold corresponding to the parameter value, and it is determined according to the parameter type of the parameter value that a trend warning needs to be issued for the hard disk based on the changing trend of the parameter value, the failure risk level value of the hard disk is determined according to the change rate of the parameter value in a predetermined period.

[0036] In operation S230 , a warning level of the hard disk is predicted according to the failure risk level value.

[0037] The hard drive's operation log can include Self-Monitoring, Analysis, and Reporting Technology (SMART) logs and the hard drive's underlying core logs. Table 1 shows the parameter types of parameter values ​​obtained from the operation log.

[0038] Table 1

[0039]

[0040] Based on Table 1, the SMART log can be used to record the parameter values ​​of basic hard disk operating parameters, such as the number of hard disk boot failures, the number of hard disk boot retries, the number of uncorrectable read errors, the number of uncorrectable errors, the number of hard disk bad blocks, the number of head swings from the start / stop area to the platter (referred to as head load / unload, LUL), the cyclic redundancy check error count (CRC count), hard disk temperature, read error rate, track error rate, platter offset, and other parameter values.

[0041] The hard drive's underlying core log records the values ​​of the drive's core operating parameters, such as the number of IO Error Detection and Correction (IOEDC) errors, workload (which can be the drive's data read and write volume per unit time period (e.g., one year), the drive's 5V supply voltage, the drive's 12V supply voltage, rotational vibration (RV) value, shock value, relative humidity value, flying altitude, and head resistance value.

[0042] As shown in Table 1, the above parameter values ​​can be used for eight types of warnings: hard disk failure warning, business pressure warning, whole-machine storage link warning, whole-machine power supply environment warning, computer room whole-machine heat dissipation environment warning, computer room whole-machine vibration environment warning, computer room whole-machine humidity environment warning, and whole-machine environment or body warning, thus achieving multi-dimensional warnings.

[0043] The parameter values ​​collected by the above operation can be stored in the form of a structured data frame, which can include information such as the machine ID, hard disk device ID, collection timestamp, parameter name, parameter value, etc. In an example, the data frame can be as follows:

[0044] {"machine identification"

[0045] Hard disk device identification

[0046] "Collection timestamp": "****year**month**day**hour**minute**second",

[0047] "Parameter name and parameter value":

[0048] "Number of startup failures": 0,

[0049] "Uncorrectable read error": 5,

[0050] "Temperature": 35,

[0051] ...}.

[0052] For the parameter values ​​in the data frame, you can perform data cleaning preprocessing operations, such as removing outliers and noise, and normalizing the parameter values ​​to ensure data accuracy.

[0053] When removing outliers and noise from parameter values, noise filtering can be employed. For example, for voltage parameters (e.g., 5V and 12V power supply voltages), a three-cycle median filter can be used. For example, for each power supply voltage value, the median of the three preceding values ​​(three values ​​in total) is calculated, replacing the value with the median to eliminate spikes in transient data. For counting parameters (e.g., the number of bad blocks), a sudden change detection method can be configured. For example, if the increment in the number of bad blocks within a single cycle (e.g., one minute) exceeds three times the historical average increment, the parameter value can be marked as suspicious, triggering a manual review process. Parameter values ​​can also be processed for missing data. For example, if a parameter value fails to be collected twice in a row (considered a short-term absence), the most recent valid value can be used to replace it. If it fails to be collected five times in a row (considered a long-term absence), the parameter value is marked as unknown and removed from the fault prediction process to prevent erroneous data from misleading the prediction process.

[0054] The parameter values ​​that have been removed of outliers and noise can be normalized. For example, the parameter values ​​can be normalized according to formula (1) to eliminate the differences in magnitude and value ranges between different parameter values. All parameter values ​​are normalized to map to the interval [0, 1].

[0055] (1)

[0056] Among them, for the parameter value of the same parameter type, is the normalized parameter value, is the parameter value, is the minimum parameter value, is the maximum parameter value.

[0057] The preprocessed parameter values ​​can be used to predict hard drive failures. For example, these parameter values ​​can be divided into a first parameter value set and a second parameter value set. The parameter values ​​in the first parameter value set may only need to be compared with a warning threshold. Once a parameter value exceeds the warning threshold, a hard drive failure is indicated, and an alert is immediately generated. The warning threshold corresponding to the first parameter value set may be a static warning threshold, while the parameter values ​​in the second parameter value set may require a warning based on at least one of the warning threshold and a parameter value change trend. The parameter values ​​in the second parameter value set can be divided into those with a warning threshold and those without a warning threshold. If a parameter value with a warning threshold exceeds the warning threshold, a hard drive failure is indicated, and an alert is immediately generated. If the parameter value does not exceed the warning threshold, a hard drive warning is issued based on the parameter value change trend. Parameter values ​​without a warning threshold require a warning based on the parameter value change trend. The warning threshold corresponding to the second parameter value set may be a dynamic warning threshold. The parameter values, alert methods, and warning thresholds in the first and second parameter sets can be shown in Table 2.

[0058] Table 2

[0059]

[0060] When the parameter value does not exceed the warning threshold and it is determined according to the parameter type of the parameter value that a trend warning needs to be issued for the hard disk based on the changing trend of the parameter value, for each parameter type, the parameter value of the parameter type of the hard disk within a predetermined time period can be obtained, and based on the change rate of the parameter value within the predetermined time period, the trend evaluation value of the change rate can be determined, and based on the trend evaluation value, the failure risk degree value of the hard disk can be obtained.

[0061] The hard drive's failure risk level is matched against the hard drive's warning file to predict the hard drive's warning level. The warning file can contain multiple risk ranges, each corresponding to a different warning level.

[0062] According to an embodiment of the present invention, by obtaining hard drive parameter values ​​from the hard drive's operation log, if the parameter values ​​do not exceed a warning threshold and a warning is required for the hard drive based on the parameter value's changing trend, the hard drive's failure risk level is determined based on the parameter value's rate of change within a predetermined period. Furthermore, the hard drive's warning level is predicted based on the failure risk level. By incorporating the parameter value's rate of change within a predetermined period into the failure prediction process, a trend analysis of the parameter values ​​is performed, and a warning is issued for the hard drive based on this trend analysis. This avoids the limitations of untimely and inaccurate hard drive failure assessments caused by alerting the hard drive based on a single threshold, thereby improving the accuracy of hard drive failure prediction.

[0063] In some embodiments, the parameter type identifier, alarm mode, warning threshold value, and parameter weight of the parameter value may be as shown in Tables 3 to 5. The parameter weight may indicate the degree of influence of the parameter value on the hard disk failure.

[0064] Table 3

[0065]

[0066] Table 4

[0067]

[0068] Table 5

[0069]

[0070] Table 3 includes parameter type identification, alarm mode, warning threshold value of parameter value and parameter weight obtained based on the first parameter value set shown in Table 2. Table 4 and Table 5 are parameter type identification, alarm mode, warning threshold value of parameter value and parameter weight obtained based on the second parameter value set shown in Table 2. Table 4 shows the information of the first type of parameter value in the second parameter value set, and Table 5 shows the information of the second type of parameter value in the second parameter value set. The correlation between the parameter values ​​of the first type of parameter value is greater than or equal to the predetermined correlation, and the correlation between the parameter values ​​of the second type of parameter value is less than the predetermined correlation. The first weight represents the degree of influence of the first type of parameter value on the hard disk failure, and the second weight represents the degree of influence of the second type of parameter value on the hard disk failure. The parameter weight, the first weight and the second weight can be adaptively adjusted according to actual needs.

[0071] Among the preprocessed parameter values, the parameter values ​​corresponding to PI1 to PI5 in Table 3 can be compared with the warning thresholds shown in Table 3, and an alarm message is directly generated if the warning thresholds are exceeded. The parameter values ​​corresponding to PI6 to PI20 in Tables 4 and 5 can be compared with the warning thresholds shown in Tables 4 and 5, and an alarm message is directly generated if the warning thresholds are exceeded. If the warning thresholds are not exceeded or do not exist, an alarm can be issued based on the changing trend of the parameter values.

[0072] In some embodiments, the warning thresholds shown in Tables 4 and 5 may be basic warning thresholds, and the parameter values ​​corresponding to PI6 to PI20 may be compared with dynamic warning thresholds. The dynamic warning thresholds may be obtained by: obtaining an adjusted value of the parameter value based on a characteristic value of the parameter value, a basic warning threshold of the parameter value, and an adjustment coefficient for adjusting the dynamic warning threshold, wherein the characteristic value of the parameter value is determined based on historical parameter values; and obtaining the warning threshold based on the basic warning threshold and the adjustment value of the parameter value.

[0073] The process of obtaining a dynamic warning threshold can be shown as formula (2).

[0074] (2)

[0075] Among them, for each parameter type with a dynamic warning threshold, It is the adjusted dynamic warning threshold, which is used to more accurately determine whether the parameter value exceeds the warning threshold. It can be a basic warning threshold set according to the hard disk standard, such as the thresholds shown in Tables 4 and 5, which are the initial benchmarks of the parameter values. It is the adjustment coefficient, ranging from 0.1 to 0.3, and is used to control the amplitude of dynamic adjustment. The larger it is, the more the dynamic warning threshold is affected by historical data; The smaller it is, the closer the dynamic warning threshold is to the basic warning threshold. The characteristic value of the parameter value, such as the sliding average, is obtained by sliding the historical parameter values ​​(for example, the parameter values ​​in the window within the last 24 hours from the time the threshold is determined). It reflects the average level of the hard disk parameter type in the last 24 hours. is the adjustment value.

[0076] According to an embodiment of the present invention, by setting a dynamic warning threshold and flexibly adjusting the warning threshold according to the actual operation of the hard disk, a more accurate comparison result between the parameter value and the warning threshold can be obtained, thereby avoiding the problem of low false alarm rate caused by false alarms or missed alarms due to individual differences in hard disks or environmental changes, and improving the accuracy of hard disk failure prediction.

[0077] In some embodiments, the characteristic value of the parameter value can change dynamically, and the above method can also include the following process: determining the updated characteristic value of the parameter value at predetermined time intervals, and when the difference between the updated characteristic value of the parameter value and the characteristic value of the parameter value exceeds a predetermined fault tolerance range, updating the dynamic warning threshold based on the shown updated characteristic value, the basic warning threshold and the adjustment coefficient.

[0078] The parameter value of the window of the last 24 hours from the moment when the threshold judgment is performed can be obtained at predetermined intervals, and the average value of the parameter value in the window can be determined as the updated characteristic value of the parameter value. If the difference between the updated characteristic value and the average value of the parameter value determined at the previous predetermined interval is within the predetermined fault tolerance range, it means that the change in the parameter value is not large, and the dynamic threshold value does not need to be updated to save resources. If the difference between the updated characteristic value and the average value of the parameter value determined at the previous predetermined interval exceeds the predetermined fault tolerance range, it means that the change in the parameter value is obvious, and the dynamic warning threshold value needs to be updated.

[0079] In another embodiment, a pre-updated dynamic warning threshold may be obtained based on the updated characteristic value, the basic warning threshold, and the adjustment coefficient. The difference between the pre-updated dynamic warning threshold and the dynamic warning threshold may be compared. If the difference is within a predetermined tolerance range, it indicates that the dynamic warning threshold has not changed significantly, and the dynamic warning threshold may not need to be updated, thereby saving resources. If the difference exceeds the predetermined tolerance range, it indicates that the dynamic warning threshold has changed significantly, and the dynamic warning threshold needs to be updated.

[0080] According to an embodiment of the present invention, by setting a predetermined fault tolerance space and updating the warning threshold when the characteristic value difference of the parameter value or the difference of the warning threshold exceeds the predetermined fault tolerance space, the resource consumption problem caused by frequent adjustment of the warning threshold can be reduced, and resource utilization can be improved while ensuring the accuracy of hard disk prediction.

[0081] Based on the above-mentioned comparison process of the parameter value and the warning threshold, when the parameter value does not exceed the warning threshold and the parameter type of the parameter value is the parameter type shown in Table 4 and Table 5, it is necessary to issue a warning for hard disk failure based on the changing trend of the parameter value to detect hard disk failure in advance.

[0082] In some embodiments, a trend evaluation value for evaluating the rate of change of the parameter value can be obtained based on the rate of change of the parameter value in a predetermined period; a failure risk degree value of the hard disk can be obtained based on the trend evaluation value and the parameter weight of the parameter value, and the parameter weight represents the degree of influence of the parameter value on the failure of the hard disk.

[0083] The predetermined period may include a first period and a second period, the start time of the first period is later than the start time of the second period, and the end time of the first period and the second period are the same, for example, both are the time when the trend analysis is currently being performed. The process of obtaining a trend evaluation value for evaluating the rate of change of the parameter value based on the rate of change of the parameter value in the predetermined period may include the following operations: determining a first rate of change of the parameter value based on the parameter value at the end time, the parameter value at the start time of the first period, and the duration of the first period; determining a second rate of change of the parameter value based on the parameter value at the end time, the parameter value at the start time of the second period, and the duration of the second period; and obtaining a trend evaluation value based on the first rate of change, the second rate of change, and the standard deviation of the parameter value in the first period and the second period.

[0084] The process of obtaining the first rate of change is shown in formula (3), the process of obtaining the second rate of change is shown in formula (4), and the process of obtaining the trend evaluation value is shown in formula (5).

[0085] (3)

[0086] (4)

[0087] (5)

[0088] in, is the first rate of change, is the parameter value at the end time, is the parameter value of the starting time of the first period, the duration of the first period is, for example, 10 minutes, is the parameter value of the starting time of the second period. The length of the first period is, for example, 120 minutes. That is, the first period is 10 minutes before the current trend analysis time, and the second period is 120 minutes before the current trend analysis time. is the trend assessment value, is the standard deviation of the parameter value, which is used as an indicator to measure the trend of change. It can be the standard deviation of the parameter value in the first time period, the standard deviation of the parameter value in the second time period, or the average of the standard deviation of the parameter value in the first time period and the standard deviation of the second time period. Preferably, the standard deviation of the first time period can be selected. Since the first time period is closest to the moment when the trend analysis is currently being performed, the accuracy of the trend evaluation value can be improved. 0.5, 0.3 and 0.2 respectively represent the first rate of change, the second rate of change and the degree of influence of the standard deviation on the trend evaluation value, which can be adaptively adjusted according to actual needs. The trend evaluation value can be between 0 and 10. The higher the trend evaluation value, the more abnormal the change trend.

[0089] According to an embodiment of the present invention, a dual-window trend assessment process is employed, with different first and second time periods set to comprehensively and accurately determine the changing trends of parameter values. The short-term window in the first period can capture sudden anomalies in the changing trend, while the long-term window in the first period can track changes in the gradual trend. The standard deviation is combined to quantify the fluctuations in the changing trend. This allows for earlier identification of subtle parameter trends, preemptive detection of potential fault hazards, and improved fault prediction efficiency and accuracy.

[0090] The trend evaluation value based on the above operation can be combined with the parameter weight to obtain the failure risk degree of the hard disk. For example, the failure risk degree value of the hard disk can be obtained based on the trend evaluation value and the first weight for the first type of parameter value, and the trend evaluation value and the second weight for the second type of parameter value. The first weight represents the influence of the first type of parameter value on the failure of the hard disk, and the second weight represents the influence of the second type of parameter value on the failure of the hard disk. The first weight is greater than the second weight.

[0091] Based on Table 4 and Table 5, the process of obtaining the hard disk failure risk value from the first type of parameter value can be shown as formula (6).

[0092] (6)

[0093] in, is the hard disk failure risk value, is the weighted sum of the first type parameter values, 0.6 is the first weight, is the first type of parameter value The parameter weights for each parameter value, is the first type of parameter value There are eight types of parameter values ​​involved in the calculation process. is the weighted sum of the second type parameter values, 0.4 is the second weight, The second type of parameter value The parameter weights for each parameter value, The second type of parameter value There are seven types of parameter values ​​involved in the calculation process.

[0094] According to an embodiment of the present invention, by collaboratively determining the hard disk failure risk value based on the changing trends of multiple parameter values, an all-round, multi-perspective comprehensive assessment of the hard disk failure risk level can be achieved, thereby improving the prediction accuracy of hard disk failures.

[0095] In some embodiments, the first weight (e.g., 0.6) and the second weight (e.g., 0.4) described above can be dynamically adjusted based on the Pearson correlation coefficient between the parameter values. For any two parameter values ​​of the first or second category, the correlation between the parameter values ​​is updated based on the distribution characteristic data of the parameter values ​​to obtain an updated correlation; if the updated correlation satisfies a trigger condition for triggering a weight adjustment operation, the first weight or the second weight is adjusted to a predetermined weight.

[0096] The correlation between parameters can be the Pearson correlation coefficient. Both the first and second weights can be dynamically adjusted based on the Pearson correlation coefficient between the parameter values. The Pearson correlation coefficient is a statistic that measures the strength of the linear correlation between two parameter value variables. The Pearson correlation coefficient ranges from -1 to 1. The closer the absolute value of the Pearson correlation coefficient is to 1, the stronger the linear correlation between the two parameter values ​​(a positive value indicates a positive correlation, and a negative value indicates a negative correlation). A Pearson correlation coefficient close to 0 indicates a weaker linear correlation between the two parameter values.

[0097] The Pearson coefficient can be obtained based on the distribution characteristic data of the parameter values, such as the mean, standard deviation, and covariance of the parameter values. By updating the Pearson coefficient based on the distribution characteristic data of the parameter values, an updated Pearson coefficient can be obtained, and the updated correlation is the updated Pearson coefficient. The trigger condition for triggering the weight adjustment operation can be a condition that the updated correlation is greater than a predetermined correlation (for example, 0.7). When the updated correlation satisfies the trigger condition for triggering the weight adjustment operation, it indicates that there is a strong positive correlation between the two parameter values, and the first weight or second weight of the first or second category parameter value of the parameter value needs to be increased.

[0098] In some embodiments, when the update correlation of two parameter values ​​is greater than 0.7, it indicates that there is a strong positive correlation between the two parameter values, and the first weight or the second weight can be increased by 20%. If the two parameter values ​​are first-category parameter values, the first weight can be increased by 20% to reach the predetermined weight. If the two parameter values ​​are second-category parameter values, the second weight can be increased by 20% to reach the predetermined weight.

[0099] When the correlation between two parameter values ​​is greater than 0.7, it indicates that the two parameter values ​​are closely related in the fault risk assessment. Increasing the weight can enhance the sensitivity of the joint influence of the two parameter values, making the fault risk assessment results more consistent with the actual correlation degree.

[0100] According to an embodiment of the present invention, by dynamically adjusting the first weight or the second weight according to the correlation between parameter values, the risk assessment of hard disk failure can be made closer to the real-time operation of the hard disk, thereby improving the prediction accuracy of hard disk failure.

[0101] In some embodiments, the warning level of the hard disk can be predicted based on the failure risk obtained according to the above operations. The process may include the following operations: matching the failure risk degree value with the warning file of the hard disk, the warning file including a first risk degree value, a second risk degree value and a third risk degree value in descending order; when the failure risk degree value is greater than or equal to the first risk degree value, the hard disk is predicted to be at the first warning level; when the failure risk degree value is greater than or equal to the second risk degree value and less than the first risk degree value, the hard disk is predicted to be at the second warning level; when the failure risk degree value is greater than or equal to the third risk degree value and less than the second risk degree value, the hard disk is predicted to be at the third warning level; when the failure risk degree value is less than the third risk degree value, no warning is issued for the hard disk.

[0102] Table 6

[0103]

[0104] Hard drive warning files can include information such as warning level, trigger conditions, warning method, and notes, as shown in Table 6. By combining warning thresholds, trend assessment values, and fault risk levels, the warning level can be predicted and recommendations can be made, including replacing hard drives, adjusting service pressure, and troubleshooting the entire system link, power supply, heat dissipation, room heat dissipation, room humidity, room vibration, and fan vibration.

[0105] Continuing with Table 6, when a parameter value in the first parameter value set exceeds the warning threshold, or a parameter value in the second parameter value set that has a warning threshold exceeds the warning threshold, it can indicate that a hard drive failure has occurred, and an alarm message is immediately generated, indicating a red alarm. This level indicates that the hard drive has failed and requires automatic isolation.

[0106] Continuing with Table 6, the first risk level value may be 8, the second risk level value may be 5, and the third risk level value may be 3. If the failure risk level C is greater than or equal to the first risk level value of 8, the hard drive is predicted to be at the first warning level, and a red warning is issued. This level indicates a significant hard drive failure trend, and data backup is recommended. If the failure risk level C is greater than or equal to the second risk level value of 5 and less than the first risk level value of 8, the hard drive is predicted to be at the second warning level, and an orange warning is issued. This level indicates a minor hard drive failure trend, and increased attention is recommended. If the failure risk level C is greater than or equal to the third risk level value of 3 and less than the second risk level value of 5, the hard drive is predicted to be at the third warning level, and a yellow warning is issued. This level indicates a potential hard drive failure trend, and continued attention is recommended. If the failure risk level C is less than the third risk level value of 3, no warning is issued for the hard drive, indicating that the hard drive currently has no failure trend, and only data logging is recommended.

[0107] According to the embodiments of the present invention, by establishing a five-level early warning mechanism and a clear hierarchical response mechanism, operations and maintenance personnel can quickly identify critical issues, prioritize high-risk faults, shorten troubleshooting time, and improve operations and maintenance efficiency. Furthermore, the system includes automatic response capabilities, such as automatically isolating hard drives when a level 1 alarm is triggered. This reduces manual intervention, reduces the workload of operations and maintenance personnel, avoids escalation of faults caused by human delays, and ensures stable operation of the storage system.

[0108] Figure 3 The figure shows an architecture diagram of a hard disk failure prediction system provided according to an embodiment of the present invention.

[0109] like Figure 3 As shown, another aspect of the present invention further provides a hard disk failure prediction system for implementing the above-mentioned hard disk failure prediction method. The hard disk failure prediction system may include a data acquisition module 310, a preprocessing module 320, a threshold detection module 330, a trend analysis module 340, a comprehensive evaluation module 350, and a warning decision module 360. These modules cooperate with each other to implement an intelligent hierarchical warning function.

[0110] based on Figure 3 The hard disk failure prediction system shown in the figure, the present invention also provides another embodiment of a hard disk failure prediction method, such as Figure 4 shown.

[0111] Figure 4 FIG. 4 shows a flow chart of a hard disk failure prediction method according to another embodiment of the present invention. Figure 4 As shown, the hard disk failure prediction method of this embodiment may include operations S401 to S411.

[0112] In operation S401, a hard disk failure prediction solution is deployed. After the server is verified and debugged in the user's computer room, the hard disk failure prediction solution is deployed on the user's end according to the user's needs.

[0113] In operation S402, hard drive logs are collected to obtain raw parameter values. After the user-side hard drive prediction solution is deployed, hard drive operation logs are automatically collected. Data is collected from each hard drive connected to the hard drive failure prediction system at a fixed one-minute interval. Multi-threading technology enables simultaneous and parallel processing of data collection tasks for multiple hard drives, improving collection efficiency and meeting the monitoring needs of large-scale hard drive clusters. The collected data is packaged and stored in a unified format of {machine ID, device ID, timestamp, parameter name, raw parameter value}. The device ID uniquely identifies each hard drive, facilitating accurate mapping during data processing and analysis; the timestamp accurately records the moment of data collection, providing a temporal reference for trend analysis and anomaly detection; the parameter name represents the corresponding hard drive operating indicator; and the raw parameter value represents the collected, unprocessed real data. This operation can be performed by the data collection module 310. Twenty hard drive log parameters (PI1-PI20), including threshold alarm parameters, trend warning parameters, and associated warning parameters, are collected at a one-minute interval to comprehensively obtain hard drive operating status information.

[0114] In operation S403, the original parameter values ​​are preprocessed. The collected original parameter values ​​are cleaned, denoised and normalized to ensure that the data quality used by the trend analysis module 340 is reliable. The cleaning operation removes duplicate values ​​and invalid data in the data; the median filter algorithm is used for denoising to remove random noise in the data; the normalization process maps parameter data of different ranges to the [0, 1] interval to facilitate analysis and calculation. This operation can be completed by the preprocessing module 320. The 3σ principle is used to eliminate abnormal data, linear interpolation is used for single-cycle missing, and continuous missing >3 cycles are marked as invalid. The data is then normalized to the [0, 1] interval using a formula to improve the data quality and availability of the parameter values.

[0115] In operation S404 , detection is performed according to the early warning threshold.

[0116] The basic threshold is set according to the hard disk standard and is dynamically adjusted through the sliding average of historical data (window = 24 hours). The dynamic warning threshold adjustment formula is shown in formula (2), which makes the threshold more consistent with the actual operation of the hard disk.

[0117] In operation S405, it is determined whether the parameter value exceeds the threshold value. If the parameter value exceeds the warning threshold value, operation S408 is performed. If the parameter value does not exceed the warning threshold value, operation S406 is performed.

[0118] Real-time detection is performed based on the preset warning thresholds of each parameter value. For any parameter value with a warning threshold, when the collected parameter value exceeds the corresponding warning threshold range, an alarm signal is immediately sent to the warning decision module 360. For parameter values ​​with thresholds and requiring fault prediction based on the reference value change trend, the corresponding warning threshold needs to be adjusted and determined with reference to formula (2). When it is determined that the current parameter value exceeds the dynamic warning threshold, a corresponding warning is directly issued through the warning decision module 360. Operations S404 and S405 can be completed by the threshold detection module 330.

[0119] In operation S406, a trend analysis is performed on the parameter values ​​to obtain a trend evaluation value. Through dual-window evaluation and fluctuation index calculation, the changing trend of the hard disk parameters is deeply analyzed. By calculating the average change rate and cumulative change rate of the trend for the short-term window of 10 minutes and the long-term window of 2 hours, and correcting it through the fluctuation index, a relatively reasonable trend evaluation value is given. This operation can be completed by the trend analysis module 340. This operation can refer to formulas (3) to (5). The dual-window trend analysis process, by combining multiple parameters and different time windows for analysis and evaluation, avoids overlap with traditional algorithms such as current simple trend analysis, thereby improving the efficiency of trend analysis.

[0120] In operation S407, a comprehensive assessment of the failure risk is performed based on the trend assessment value. A comprehensive assessment of the failure risk is performed based on the trend changes of the parameter value itself and the changes in the parameter values ​​associated with the parameter value, achieving a multi-parameter collaborative assessment of the hard disk risk. The hard disk parameters are divided into two categories: the first category of strongly associated parameter values ​​(8 items) and the second category of reference-associated parameter values ​​(7 items), totaling 15 parameter values. The trend assessment values ​​of these 15 parameter values ​​are accumulated according to their respective parameter weights, and combined with the first and second weights to obtain a failure risk level value that comprehensively reflects the overall risk level of the hard disk, providing a basis for the early warning decision module 360. This operation can be completed by the comprehensive assessment module 350. This operation can refer to formula (6).

[0121] In operation S408, a warning decision is made. Based on the results of the threshold detection module 330, trend analysis module 340, and comprehensive assessment module 350, the warning levels are divided into five levels: direct warning, first warning level, second warning level, third warning level, and no warning. When the parameter value exceeds the warning threshold, a direct warning is issued. When the failure risk level is greater than 8, a first-level warning (red) is issued, indicating a significant failure trend. It is recommended to back up data, prepare for a hard drive replacement, and refer to reasonable recommendations. When the failure risk level is between 5 and 8, a second-level warning (orange) is issued, indicating a mild failure trend. Increased attention is paid to the hard drive and reasonable recommendations are being referred to. When the failure risk level is between 3 and 5, a third-level warning (yellow) is issued, indicating a failure trend. Continued attention is being paid and reasonable recommendations are being referred to. When the failure risk level is less than 3, only data is recorded without issuing a warning. This process can be completed by the warning decision module 360.

[0122] In operation S409, the user is notified to process. When the warning decision module 360 ​​issues an alarm or performs a first-level warning, a second-level warning, or a third-level warning, the user is promptly notified to process the corresponding action. After notifying the user to process the action, the process can be terminated or the execution can continue with operation S410 or operation S411.

[0123] In operation S410, the re-judgment and analysis of the faulty hard disk is continuously followed up and analyzed. By analyzing the parameters of the faulty hard disk, missed reports and false reports of hard disk parameters are identified, and data statistics are prepared.

[0124] In operation S411, the warning thresholds and parameter weights of the parameter values ​​are optimized. Through in-depth analysis and research of the faulty hard disk device, deficiencies in the warning solution are promptly identified and the warning thresholds and parameter weights are adjusted and optimized in real time, continuously improving the accuracy of fault prediction. The adjusted warning thresholds can be used in the detection process based on the warning thresholds in operation S404. The adjusted parameter weights can also be used in the process of comprehensively assessing fault risks based on trend assessment values.

[0125] According to the embodiments of the present invention, a multi-dimensional evaluation system, transcending the single-threshold evaluation model, is established. A multi-dimensional evaluation model encompasses static and dynamic warning thresholds, trend warnings, warnings based on strongly correlated first-class parameter values, and reference-related second-class parameter values. This allows for a comprehensive, multi-perspective assessment of the hard drive's operating status. This highly accurate fault prediction and efficient operation and maintenance management achieves the goal of foreseeing failures, improving system stability, reducing manual intervention, saving costs, and increasing user satisfaction.

[0126] It should be noted that, unless it is clearly stated that there is a sequence of execution between different operations shown in the flowchart in the embodiments of the present invention, or there is a sequence of execution between different operations in technical implementation, otherwise, the execution order of multiple operations may not be prioritized, and multiple operations may also be executed simultaneously.

[0127] Based on the above hard disk failure prediction method, the present invention also provides a hard disk failure prediction device. Figure 5 The device is described in detail.

[0128] Figure 5 A structural block diagram of a hard disk failure prediction device according to an embodiment of the present invention is shown.

[0129] like Figure 5 As shown, the hard disk failure prediction device 500 of this embodiment includes an acquisition module 510 , a first determination module 520 and a prediction module 530 .

[0130] The acquisition module 510 is used to acquire the parameter value of the hard disk from the operation log of the hard disk.

[0131] The first determination module 520 is used to determine the failure risk level value of the hard disk based on the rate of change of the parameter value in a predetermined period when the parameter value does not exceed the warning threshold corresponding to the parameter value and it is determined according to the parameter type of the parameter value that a trend warning needs to be issued for the hard disk based on the changing trend of the parameter value.

[0132] The prediction module 530 is used to predict the warning level of the hard disk according to the failure risk level value.

[0133] In some embodiments, the first determining module 520 may include a first determining unit and a second determining unit.

[0134] The first determining unit is configured to obtain a trend evaluation value for evaluating a change rate of the parameter value according to a change rate of the parameter value in a predetermined period of time.

[0135] The second determining unit is configured to obtain a hard disk failure risk value according to the trend evaluation value and a parameter weight of the parameter value, wherein the parameter weight represents the degree of influence of the parameter value on the hard disk failure.

[0136] In some embodiments, the first determining unit may include a first determining subunit, a second determining subunit, and a third determining subunit.

[0137] The first determining subunit is configured to determine a first change rate of the parameter value based on the parameter value at the end time, the parameter value at the start time of the first time period, and the duration of the first time period.

[0138] The second determining subunit is configured to determine a second change rate of the parameter value based on the parameter value at the end time, the parameter value at the start time of the second time period, and the duration of the second time period.

[0139] The third determining subunit is configured to obtain a trend evaluation value based on the first change rate, the second change rate, and the standard deviation of the parameter value in the first time period and the second time period.

[0140] In some embodiments, the hard disk failure prediction device may further include a second determination module.

[0141] The second determination module is used to determine whether the parameter value is a first-category parameter value or a second-category parameter value according to the parameter identifier of the parameter value, the correlation between the parameter values ​​of the first-category parameter value is greater than or equal to the predetermined correlation, and the correlation between the parameter values ​​of the second-category parameter value is less than the predetermined correlation.

[0142] In some embodiments, the second determining unit may include a fourth determining subunit.

[0143] The fourth determination subunit is used to obtain the failure risk degree value of the hard disk based on the trend evaluation value and the first weight for the first type of parameter value, and the trend evaluation value and the second weight for the second type of parameter value, wherein the first weight represents the degree of influence of the first type of parameter value on the failure of the hard disk, and the second weight represents the degree of influence of the second type of parameter value on the failure of the hard disk, and the first weight is greater than the second weight.

[0144] In some embodiments, the hard disk failure prediction device may further include an updating module and an adjusting module.

[0145] For any two parameter values ​​of the first or second category:

[0146] The updating module is used to update the correlation between the parameter values ​​according to the distribution characteristic data of the parameter values ​​to obtain the updated correlation.

[0147] The adjustment module is configured to adjust the first weight or the second weight to a predetermined weight when the updated correlation satisfies a triggering condition for triggering a weight adjustment operation.

[0148] In some embodiments, the hard disk failure prediction device may further include a third determination module and a fourth determination module.

[0149] The third determination module is used to obtain the adjusted value of the parameter value according to the characteristic value of the parameter value, the basic warning threshold of the parameter value, and the adjustment coefficient for adjusting the warning threshold. The characteristic value of the parameter value is determined based on the historical parameter value.

[0150] The fourth determining module is used to obtain the warning threshold value according to the basic warning threshold value and the adjustment value of the parameter value.

[0151] In some embodiments, the prediction module 530 may include a matching unit, a first prediction unit, a second prediction unit, a third prediction unit, and a fourth prediction unit.

[0152] The matching unit is used to match the fault risk level value with a warning file of the hard disk, where the warning file includes a first risk level value, a second risk level value, and a third risk level value in descending order.

[0153] The first prediction unit is configured to predict that the hard disk is at a first warning level when the failure risk level value is greater than or equal to a first risk level value.

[0154] The second prediction unit is configured to predict that the hard disk is at a second warning level when the failure risk level value is greater than or equal to the second risk level value and less than the first risk level value.

[0155] The third prediction unit is configured to predict that the hard disk is at a third warning level when the failure risk level value is greater than or equal to the third risk level value and less than the second risk level value.

[0156] The fourth prediction unit is configured to not issue an early warning to the hard disk when the failure risk level value is less than the third risk level value.

[0157] According to embodiments of the present invention, any multiple modules among the acquisition module 510, the first determination module 520, and the prediction module 530 may be combined into a single module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. According to embodiments of the present invention, at least one of the acquisition module 510, the first determination module 520, and the prediction module 530 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware through any other reasonable means of circuit integration or packaging, or may be implemented in any one of software, hardware, and firmware, or any appropriate combination of these. Alternatively, at least one of the acquisition module 510, the first determination module 520, and the prediction module 530 may be at least partially implemented as a computer program module that, when executed, performs the corresponding functionality.

[0158] Figure 6 A block diagram of an electronic device suitable for implementing a hard disk failure prediction method according to an embodiment of the present invention is shown.

[0159] like Figure 6As shown, an electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 602 or programs loaded from a storage unit 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or related chipsets and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0160] Various programs and data required for the operation of the electronic device 600 are stored in the RAM 603. The processor 601, ROM 602, and RAM 603 are connected to each other via a bus 604. The processor 601 executes the programs in the ROM 602 and / or RAM 603 to perform various operations according to the method flow of the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than the ROM 602 and RAM 603. The processor 601 may also execute the programs stored in the one or more memories to perform various operations according to the method flow of the embodiment of the present invention.

[0161] According to an embodiment of the present invention, electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to bus 604. Electronic device 600 may also include one or more of the following components connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 608 including a hard disk; and a communication section 609 including a network interface card such as a LAN card or modem. Communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. Removable media 611, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 610 as needed, so that computer programs read from the removable media can be installed into storage section 608 as needed.

[0162] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0163] According to an embodiment of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, and may include, for example, but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, a computer-readable storage medium may include the ROM 602 and / or RAM 603 described above, and / or one or more memories other than ROM 602 and RAM 603.

[0164] The embodiments of the present invention further include a computer program product, which includes a computer program containing program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to cause the computer system to implement the method provided by the embodiments of the present invention.

[0165] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when the computer program is executed by the processor 601. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0166] In one embodiment, the computer program may be stored on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal on a network medium, downloaded and installed via the communication portion 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0167] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609 and / or installed from a removable medium 611. When the computer program is executed by the processor 601, the above-described functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the systems, devices, means, modules, units, etc. described above can be implemented by computer program modules.

[0168] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0169] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0170] It will be understood by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or couplings fall within the scope of the present invention.

[0171] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A hard disk failure prediction method, characterized in that: The method comprises: Obtaining parameter values ​​of the hard disk from the operation log of the hard disk; If the parameter value does not exceed the warning threshold corresponding to the parameter value, and it is determined based on the parameter type of the parameter value that a trend warning needs to be issued for the hard disk based on the change trend of the parameter value, determining the failure risk level of the hard disk based on the change rate of the parameter value in a predetermined period of time; Predicting a warning level of the hard disk according to the failure risk level value; The method further comprises: determining, according to the parameter identifier of the parameter value, that the parameter value is a first-category parameter value or a second-category parameter value, wherein a correlation between parameter values ​​of the first-category parameter value is greater than or equal to a predetermined correlation, and a correlation between parameter values ​​of the second-category parameter value is less than the predetermined correlation; For any two parameter values ​​of the first category parameter value or the second category parameter value: based on the distribution characteristic data of the parameter values, the correlation between the parameter values ​​is updated to obtain an updated correlation; when the updated correlation meets the trigger condition for triggering the weight adjustment operation, the first weight or the second weight is adjusted to a predetermined weight, the first weight represents the degree of influence of the first category parameter value on the failure of the hard disk, the second weight represents the degree of influence of the second category parameter value on the failure of the hard disk, and the first weight is greater than the second weight.

2. The method according to claim 1, characterized in that Determining the failure risk level of the hard disk according to the rate of change of the parameter value in a predetermined period includes: Obtaining a trend evaluation value for evaluating the rate of change of the parameter value according to the rate of change of the parameter value during the predetermined period; A failure risk level value of the hard disk is obtained according to the trend evaluation value and the parameter weight of the parameter value, wherein the parameter weight represents the influence of the parameter value on the failure of the hard disk.

3. The method according to claim 2, characterized in that The predetermined time period includes a first time period and a second time period, the start time of the first time period is later than the start time of the second time period, and the end time of the first time period and the end time of the second time period are the same; Obtaining a trend evaluation value for evaluating the rate of change of the parameter value according to the rate of change of the parameter value during the predetermined period includes: determining a first rate of change of the parameter value based on the parameter value at the end time, the parameter value at the start time of the first time period, and the duration of the first time period; determining a second rate of change of the parameter value based on the parameter value at the end time, the parameter value at the start time of the second period, and the duration of the second period; The trend evaluation value is obtained based on the first change rate, the second change rate, and the standard deviation of the parameter value in the first time period and the second time period.

4. The method according to claim 2, characterized in that Obtaining the failure risk level of the hard disk according to the trend evaluation value and the parameter weight of the parameter value includes: A failure risk level value of the hard disk is obtained based on the trend evaluation value and the first weight for the first type of parameter value, and the trend evaluation value and the second weight for the second type of parameter value.

5. The method according to claim 1, wherein The warning threshold is determined as follows: obtaining an adjusted value of the parameter value according to a characteristic value of the parameter value, a basic warning threshold of the parameter value, and an adjustment coefficient for adjusting the warning threshold, wherein the characteristic value of the parameter value is determined based on a historical parameter value; The warning threshold is obtained according to the basic warning threshold of the parameter value and the adjustment value.

6. The method according to claim 1, characterized in that The predicting of the warning level of the hard disk according to the failure risk value includes: Matching the fault risk level value with a warning file of the hard disk, the warning file including a first risk level value, a second risk level value, and a third risk level value in descending order; If the failure risk level is greater than or equal to the first risk level, predicting that the hard disk is at a first warning level; When the failure risk level value is greater than or equal to the second risk level value and less than the first risk level value, predicting that the hard disk is at a second warning level; When the failure risk level value is greater than or equal to the third risk level value and less than the second risk level value, predicting that the hard disk is at a third warning level; When the failure risk level value is less than the third risk level value, no warning is issued for the hard disk.

7. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images