CN117215812A - Hard disk fault prediction method, device, equipment and storage medium - Google Patents

Hard disk fault prediction method, device, equipment and storage medium Download PDF

Info

Publication number
CN117215812A
CN117215812A CN202210605693.1A CN202210605693A CN117215812A CN 117215812 A CN117215812 A CN 117215812A CN 202210605693 A CN202210605693 A CN 202210605693A CN 117215812 A CN117215812 A CN 117215812A
Authority
CN
China
Prior art keywords
hard disk
user
fault prediction
fault
sample set
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210605693.1A
Other languages
Chinese (zh)
Inventor
姬云飞
黄科
张小兵
周希锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Huawei Technology Co Ltd
Original Assignee
Chengdu Huawei Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Huawei Technology Co Ltd filed Critical Chengdu Huawei Technology Co Ltd
Priority to CN202210605693.1A priority Critical patent/CN117215812A/en
Publication of CN117215812A publication Critical patent/CN117215812A/en
Pending legal-status Critical Current

Links

Classifications

    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Debugging And Monitoring (AREA)

Abstract

The application provides a hard disk fault prediction method, a hard disk fault prediction device, hard disk fault prediction equipment and a storage medium. In an embodiment, feature data of a hard disk of a user is obtained, wherein the feature data is data of factors affecting hard disk faults; inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk, and obtaining fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk or not; determining a demand reference value corresponding to a user, wherein the demand reference value is used for indicating the acceptable degree of the user to the number of hard disks with fault risks; and adjusting the fault prediction results of the plurality of hard disks according to the demand reference value. Therefore, through the technical scheme of the application, the fault prediction result can be adjusted according to the acceptable degree of different users to the number of the hard disks with faults, so that the fault prediction result meets the requirements of the users, has individuation, and improves the experience of the users.

Description

Hard disk fault prediction method, device, equipment and storage medium
Technical Field
The present application relates to the field of multimedia technologies, and in particular, to a method, an apparatus, a device, and a storage medium for predicting a hard disk failure.
Background
The hard disk is used as a basic component of the data storage system, and the health condition of the hard disk has great influence on the reliability and usability of the whole storage system. In the hardware problem distribution ratio of the production environment, the fault problem caused by the hard disk accounts for 14 percent. If the possible risk state of the hard disk can be predicted in advance before the hard disk fails or the failure alarm occurs, the passive fault tolerance is converted into active fault tolerance, and the data loss of a user and the economic loss caused by the data loss can be avoided to a great extent.
Currently, in order to avoid data loss of a user, a hard disk risk prediction technology is generally adopted to alarm that a hard disk may malfunction in a subsequent period of time. In the related art, a hard disk failure prediction model is generally trained by collecting a feature data set of a hard disk, so that a hard disk failure result is predicted using the hard disk failure prediction model. However, only a single hard disk feature data is used for training the hard disk fault prediction model, so that the hard disk risk faults predicted by the hard disk fault prediction model cannot meet the requirements of users, and the experience of the users is reduced.
Disclosure of Invention
The embodiment of the application provides a hard disk fault prediction method, a device, equipment and a storage medium, which can adjust fault prediction results according to the acceptable degree of different users to the number of hard disks with faults, so that the fault prediction results more meet the requirements of the users, have individuation and improve the experience of the users.
In a first aspect, an embodiment of the present application provides a hard disk failure prediction method, which is applied to a server, and includes:
acquiring characteristic data of a hard disk of a user, wherein the characteristic data are data of factors affecting the hard disk fault;
inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk to obtain fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk;
determining a demand reference value corresponding to the user, wherein the demand reference value is used for indicating the acceptability degree of the user on the number of hard disks with fault risks;
according to the demand reference value, adjusting fault prediction results of a plurality of hard disks;
And sending a plurality of failure prediction results after hard disk adjustment to a client.
According to the embodiment of the application, the hard disk fault prediction result predicted by using the hard disk fault prediction model is adjusted based on the demand reference value, so that the adjusted fault prediction result can better meet the demands of users. The requirement reference value can represent the acceptable degree of the user to the number of the hard disks with fault risks in the plurality of the hard disks owned by the user, for example, the user is sensitive to the cost of hard disk replacement, i.e. the acceptable degree of the user to the number of the hard disks with fault risks is low; for another example, in order to avoid that the hard disk failure affects the production efficiency, etc., the sensitivity of the user to the hard disk failure is high, that is, the user has a high acceptable level for the number of hard disks with failure risk. Therefore, the fault prediction result is adjusted according to the acceptable degree of the number of the hard disks with faults by different users, so that the fault prediction result meets the requirements of the users, the individuation is realized, and the experience degree of the users is improved.
In a possible implementation manner, the fault prediction information further includes a probability value for indicating that the hard disk has a fault risk, and the adjusting, according to the requirement reference value, the fault prediction results of the plurality of hard disks includes:
According to the demand reference value and a preset threshold value, adjusting probability values of a plurality of hard disks;
and generating a plurality of fault prediction results after hard disk adjustment according to the probability values after hard disk adjustment and the threshold values.
In a possible implementation manner, the fault prediction information further includes a probability value for indicating that the hard disk has a fault risk, and the adjusting, according to the requirement reference value, the fault prediction results of the plurality of hard disks includes:
adjusting the threshold according to the demand reference value;
and generating a fault prediction result after the hard disk is adjusted according to the probability value of the hard disk and the adjusted threshold value.
In one possible implementation manner, the determining the requirement reference value corresponding to the user includes:
receiving a preference type of the user sent by the client;
and determining the demand reference value according to the preference type.
In this embodiment, the user may configure, according to his own needs, the preference type to which his own belongs through the client, so that the flexibility of configuring his own preference type by the user may be improved.
In one possible implementation, the determining a preference type to which the user belongs includes:
Acquiring user behavior data of the user using a hard disk;
classifying the users according to the user behavior data, and determining the preference type of the users.
In this embodiment, the accuracy of determining the preference type of the user can be improved by determining the preference type to which the user belongs by the user using the behavior data in the hard disk process.
In one possible implementation, the method further includes:
converting the probability value of the hard disk into a health value of the hard disk according to a preset probability conversion value;
and sending the health value of the hard disk to the client so that a user can determine the health of the hard disk.
In this embodiment, the probability value of the hard disk may be converted into the health value of the hard disk, so that the user may determine the health of the hard disk more intuitively. The user can compare the health degree values of the hard disk at the current time with the health degree values of the hard disk at the current time through the plurality of historical time hard disk health degree values stored by the client, so that the user can know the change condition of the health degree of the hard disk at the time, and the use experience of the user is improved.
In one possible implementation, the method further includes:
obtaining health values of the hard disk at a plurality of moments in a historical time period;
Generating health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in a historical time period and the health degree values of the hard disk;
the sending the health value of the hard disk to the client so as to enable the user to determine the health of the hard disk comprises the following steps:
and sending the health degree information of the plurality of hard disks to the client so as to enable the client to determine a change trend chart of the health degree of the hard disks along with time.
In this embodiment, the server may send the health value of the hard disk at multiple times and the health value of the current time in the historical time period to the client according to the summary, so that the client generates a trend chart of the change of the hard disk with time, so that the user can more intuitively observe the change of the health value of the hard disk with time, and the experience of the user is improved.
In one possible implementation manner, the server stores a history sample set for training the hard disk failure prediction model, and the method further includes:
acquiring a feedback sample set fed back by the user, wherein each feedback sample in the feedback sample set comprises characteristic data of a target hard disk and tag information used for indicating whether the target hard disk is faulty, and the target hard disk is a hard disk in which fault prediction results in a plurality of hard disks are inconsistent with the tag information;
Performing statistical analysis on characteristic data of a plurality of hard disks in the history sample set and characteristic data of a plurality of target hard disks in the feedback sample set, and determining difference information between the history sample set and the feedback sample set;
determining a target training sample set from the historical sample set and the feedback sample set according to the difference information;
and optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
In this embodiment, by performing statistical analysis on the characteristic data of the hard disk fed back by the user and the characteristic data of each hard disk in the history sample set, it is able to determine difference information between the history sample set and the feedback sample set, for example, whether the model of the hard disk in the history sample set, the manufacturer, and the like include the feedback sample, the model of the middle hard disk, and the manufacturer of the hard disk, so as to determine the sample for optimizing the hard disk failure prediction model from the history sample set and the feedback sample set, thereby improving the accuracy of the hard disk failure prediction model.
In a second aspect, an embodiment of the present application provides a hard disk failure prediction apparatus, which is applied to a server, including:
The system comprises an acquisition module, a storage module and a control module, wherein the acquisition module is used for acquiring characteristic data of a hard disk of a user, wherein the characteristic data is data of factors affecting the hard disk fault;
the prediction module is used for inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk to obtain fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk;
the determining module is used for determining a demand reference value corresponding to the user, wherein the demand reference value is used for indicating the acceptability degree of the user on the number of hard disks with fault risks;
the adjusting module is used for adjusting the fault prediction results of the plurality of hard disks according to the demand reference value;
and the sending module is used for sending the failure prediction results after the hard disk adjustment to the client.
According to the embodiment of the application, the hard disk fault prediction result predicted by using the hard disk fault prediction model is adjusted based on the demand reference value, so that the adjusted fault prediction result can better meet the demands of users. The requirement reference value can represent the acceptable degree of the user to the number of the hard disks with fault risks in the plurality of the hard disks owned by the user, for example, the user is sensitive to the cost of hard disk replacement, i.e. the acceptable degree of the user to the number of the hard disks with fault risks is low; for another example, in order to avoid that the hard disk failure affects the production efficiency, etc., the sensitivity of the user to the hard disk failure is high, that is, the user has a high acceptable level for the number of hard disks with failure risk. Therefore, the fault prediction result is adjusted according to the acceptable degree of the number of the hard disks with faults by different users, so that the fault prediction result meets the requirements of the users, the individuation is realized, and the experience degree of the users is improved.
In a possible implementation manner, the fault prediction information further includes a probability value for indicating that the hard disk has a fault risk, and the adjustment module is configured to:
according to the demand reference value and a preset threshold value, adjusting probability values of a plurality of hard disks;
and generating a plurality of fault prediction results after hard disk adjustment according to the probability values after hard disk adjustment and the threshold values.
In a possible implementation manner, the fault prediction information further includes a probability value for indicating that the hard disk has a fault risk, and the adjustment module is configured to:
adjusting the threshold according to the demand reference value;
and generating a fault prediction result after the hard disk is adjusted according to the probability value of the hard disk and the adjusted threshold value.
In one possible implementation, the determining module is configured to:
receiving a preference type of the user sent by the client;
and determining the demand reference value according to the preference type.
In this embodiment, the user may configure, according to his own needs, the preference type to which his own belongs through the client, so that the flexibility of configuring his own preference type by the user may be improved.
In one possible implementation, the determining module is configured to:
acquiring user behavior data of the user using a hard disk;
classifying the users according to the user behavior data, and determining the preference type of the users.
In this embodiment, the accuracy of determining the preference type of the user can be improved by determining the preference type to which the user belongs by the user using the behavior data in the hard disk process.
In one possible implementation, the apparatus further includes:
the conversion module is used for converting the probability value of the hard disk into the health value of the hard disk according to a preset probability conversion value;
the sending module is further configured to send the health value of the hard disk to the client, so that a user determines the health of the hard disk.
In this embodiment, the probability value of the hard disk may be converted into the health value of the hard disk, so that the user may determine the health of the hard disk more intuitively. The user can compare the health degree values of the hard disk at the current time with the health degree values of the hard disk at the current time through the plurality of historical time hard disk health degree values stored by the client, so that the user can know the change condition of the health degree of the hard disk at the time, and the use experience of the user is improved.
In one possible implementation, the apparatus further includes:
the acquisition module is used for acquiring health values of the hard disk at a plurality of moments in a historical time period;
the generation module is used for generating the health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in the historical time period and the health degree values of the hard disk;
the sending module is used for sending the health degree information of the plurality of hard disks to the client so as to enable the client to determine a change trend chart of the health degree of the hard disks along with time.
In this embodiment, the server may send the health value of the hard disk at multiple times and the health value of the current time in the historical time period to the client according to the summary, so that the client generates a trend chart of the change of the hard disk with time, so that the user can more intuitively observe the change of the health value of the hard disk with time, and the experience of the user is improved.
In one possible implementation manner, the server stores a history sample set for training the hard disk failure prediction model, and the apparatus further includes:
the system comprises an acquisition module, a feedback sample set and a storage module, wherein the acquisition module is used for acquiring a feedback sample set fed back by a user, each feedback sample in the feedback sample set comprises characteristic data of a target hard disk and label information used for indicating whether the target hard disk is faulty, and the target hard disk is a hard disk with fault prediction results inconsistent with the label information in a plurality of hard disks;
The determining module is further configured to perform statistical analysis on characteristic data of a plurality of hard disks in the historical sample set and characteristic data of a plurality of target hard disks in the feedback sample set, and determine difference information between the historical sample set and the feedback sample set; determining a target training sample set from the historical sample set and the feedback sample set according to the difference information;
and the optimization module is used for optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
In this embodiment, by performing statistical analysis on the characteristic data of the hard disk fed back by the user and the characteristic data of each hard disk in the history sample set, it is able to determine difference information between the history sample set and the feedback sample set, for example, whether the model of the hard disk in the history sample set, the manufacturer, and the like include the feedback sample, the model of the middle hard disk, and the manufacturer of the hard disk, so as to determine the sample for optimizing the hard disk failure prediction model from the history sample set and the feedback sample set, thereby improving the accuracy of the hard disk failure prediction model.
In a third aspect, an embodiment of the present application provides a hard disk failure prediction system, where the system may include a server and a client, where the system is configured to perform the method provided in the first aspect and any one of the possible implementation manners of the first aspect.
In a fourth aspect, an embodiment of the present application provides a computer apparatus, including: at least one memory for storing a program; at least one processor for executing a memory-stored program, the processor being adapted to perform the method provided in the first aspect and any one of the possible implementations of the first aspect when the memory-stored program is executed.
In a fifth aspect, embodiments of the present application provide a computer storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the method provided in the first aspect and any one of the possible implementations of the first aspect.
In an eighth aspect, embodiments of the present application provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method provided in the first aspect and any one of the possible implementations of the first aspect.
Drawings
FIG. 1 is a system architecture diagram of a hard disk failure prediction system according to an embodiment of the present application;
FIG. 2 is a system architecture diagram of another hard disk failure prediction system according to an embodiment of the present application;
FIG. 3 is a schematic flow chart of a hard disk failure prediction method according to an embodiment of the present application;
FIG. 4 is a schematic diagram of an interface of a computer device according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a correspondence relationship between preference types and requirement reference values according to an embodiment of the present application;
FIG. 6 is a schematic diagram of an interface of another computer device according to an embodiment of the present application;
FIG. 7 is a schematic diagram of a decision tree for determining preference types provided by an embodiment of the present application;
fig. 8 is a flowchart of a health degree information generating method according to an embodiment of the present application;
FIG. 9 is a flow chart of another method for health degree according to an embodiment of the present application;
FIG. 10 is a schematic diagram of a health conversion function according to an embodiment of the present application;
FIG. 11 is a schematic diagram of a health degree variation trend provided by an embodiment of the present application;
FIG. 12 is a flowchart of another health degree information generating method according to an embodiment of the present application;
FIG. 13 is a flowchart of a method for optimizing a hard disk failure prediction model according to an embodiment of the present application;
fig. 14 is a schematic structural diagram of a hard disk failure prediction apparatus according to an embodiment of the present application;
FIG. 15 is a schematic diagram of a computer device according to an embodiment of the present application;
Fig. 16 is a schematic structural diagram of a server according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be described below with reference to the accompanying drawings.
In describing embodiments of the present application, words such as "exemplary," "such as" or "for example" are used to mean serving as an example, instance, illustration, or description. Any embodiment or design described herein as "exemplary," "such as" or "for example" is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "such as" or "for example," etc., is intended to present related concepts in a concrete fashion.
In the description of the embodiments of the present application, the term "and/or" is merely an association relationship describing an association object, and indicates that three relationships may exist, for example, a and/or B may indicate: a alone, B alone, and both A and B. In addition, unless otherwise indicated, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of terminals means two or more terminals.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or the inclusion of technical features that are indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically noted.
The hard disk is used as a basic component of the data storage system, and the health condition of the hard disk has great influence on the reliability and usability of the whole storage system. In the hardware problem distribution ratio of the production environment, the fault problem caused by the hard disk accounts for 14 percent. If the possible risk state of the hard disk can be predicted in advance before the hard disk fails or the failure alarm occurs, the passive fault tolerance is converted into active fault tolerance, and the data loss of a user and the economic loss caused by the data loss can be avoided to a great extent.
As shown in fig. 1, to avoid data loss, the client 12 typically sends a request message to the server 11 to request a failure risk prediction for the hard disk in the data storage system 121, so as to determine whether the hard disk will fail in a subsequent period of time.
Currently, in the related art, a hard disk fault prediction model is usually trained by collecting a characteristic data set of a hard disk. The prediction result of the hard disk fault prediction model can give a certain prompt to the user so that the user can decide whether to replace the hard disk. However, when the prediction result of the hard disk failure prediction model cannot meet the expectations of the user, the user experience is reduced. For example, the prediction result obtained by the hard disk fault prediction model shortens the replacement period of the hard disk, which not only causes that the user needs to consume manpower and physical force to replace the hard disk, but also may cause waste of resources. For another example, the user has a high sensitivity to the hard disk failure, but the hard disk failure prediction model predicts that the hard disk is not replaced, so that the user will not replace the hard disk. But the use of the device is not smooth at the subsequent time, resulting in a low production efficiency for the user, etc.
According to the method, the hard disk fault prediction model is trained by using only single hard disk characteristic data, so that the hard disk fault prediction model can not be used for predicting hard disk risk faults and can not meet the requirements of users, and the experience of the users is reduced.
Therefore, the embodiment of the application adjusts the hard disk fault prediction result predicted by using the hard disk fault prediction model based on the demand reference value, so that the adjusted fault prediction result can more meet the demands of users. The demand reference value can represent the acceptable degree of the user to the number of the hard disks with fault risks in the plurality of the hard disks, so that the fault prediction result can be more accordant with the demands of the user, has individuation, and improves the experience of the user.
Based on the above inventive concept, a hard disk failure prediction system provided by an embodiment of the present application will be described in detail.
Fig. 2 is a schematic architecture diagram of a hard disk failure prediction system 200 according to an embodiment of the present application. As shown, the hard disk failure prediction system 200 includes a server 21 and a client 22. The communication manner between the client 21 and the server 22 may be an Https secure transmission channel. Here, the server 21 is configured to adjust, according to the requirement reference value, the failure prediction result of the hard disk predicted by using the hard disk failure prediction model, so that the adjusted failure prediction result can more conform to the requirement of the user, so that the model failure prediction result has individuality, and the experience of the user is improved. The client 22 side includes a data storage system 221, and the client may be regarded as a management platform of the data storage system 221, and is configured to collect characteristic data of each hard disk in the data storage system 221, so as to send a request message to the server 21, so as to request the server 22 to predict that the hard disk is capable of having a fault risk.
In one example, the service end related to the embodiment of the present application may be used to provide a cloud service, which may be a server or a super terminal that may establish a communication connection with a computer device and may provide a hard disk failure prediction function, an operation function, and/or a storage function for the computer device. The server related in the embodiment of the present application may be a hardware server, or may be embedded in a virtualized environment, for example, the server related in the embodiment of the present application may be a virtual machine executing on a hardware server including one or more other virtual machines.
In one example, the client may be a computer device, a virtual machine executing on a computer device that includes at least one other virtual machine, or an application. The computer equipment related in the embodiment of the application can be a mobile phone, a tablet personal computer, a host computer, a desktop computer and the like. Exemplary embodiments of computer devices to which embodiments of the present application relate include, but are not limited to, devices on which iOS, android, windows, hong System (Harmony OS) or other operating systems are mounted. The embodiment of the application does not particularly limit the type of the computer equipment. In one example, the computer device may have a sensor therein for collecting data. By way of example, the sensor may be a sensor for monitoring the health or movement of the user, such as a speed sensor, an optical heart rate sensor, a position sensor, or the like.
The hard disk failure prediction method according to the embodiment of the present application will be described in detail based on the hard disk failure prediction system described in fig. 2.
Fig. 3 is a flow chart of a hard disk failure prediction method according to an embodiment of the present application. The method may be performed by any apparatus, platform or cluster of devices having computing, processing capabilities. For convenience of description, the execution subject of the method may be referred to as a server, and the server may start to execute the method under a triggering operation of the server or the user.
As shown in fig. 3, the hard disk failure prediction method provided by the embodiment of the application includes the following steps.
S301, acquiring characteristic data of a hard disk of a user, wherein the characteristic data are data of factors influencing hard disk faults.
The user may be an enterprise, for example, an enterprise having a large-scale data center, which may include a plurality of hard disks for data caching or storage, i.e., the user has a plurality of hard disks. In order to ensure the working efficiency and the like, the client requests the server to predict whether the hard disk has fault risks. When a user needs to detect whether the hard disk is at risk of failure or not, the client can send request information to the server, wherein the request information is used for indicating whether the hard disk is at risk of failure or not to predict.
In some embodiments, the request information may include characteristic data of the hard disk. In other embodiments, after receiving the request information, the server may send a data acquisition request to the client, so as to acquire the characteristic data of the hard disk from the database used by the client to store the characteristic data of the hard disk. The characteristic data of the hard disk are data of factors influencing the hard disk faults, such as the running speed of the hard disk, the type of the hard disk, the manufacturer of the hard disk, the delivery date of the hard disk, the running time of the hard disk, the capacity of the hard disk and the like.
For example, the characteristic data of the hard disk may be stored in the database of the client in the form of an information table, as shown in table one:
list one
Running speedDegree of Type(s) Delivery date Duration/hour of operation capacity/T
Hard disk 1 XX A 2019.01.02 —— 500
Hard disk 2 XX B 2020.11.02 —— 500
…… XX C 2022.05.13 —— 200
S302: and inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk, and obtaining fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk.
The hard disk fault prediction model is obtained through training of a historical sample set. Wherein each sample in the history sample set includes characteristic data of the hard disk and tag information for indicating whether the hard disk is at risk of failure. Here, in order to ensure universality of the hard disk failure prediction model, the hard disks corresponding to the plurality of samples in the history sample set may be hard disks of different manufacturers, different models and different functions.
And inputting the characteristic data of the hard disk into a hard disk fault prediction model to predict and obtain the fault prediction information of the hard disk. The failure prediction information of the hard disk may include information capable of indicating whether the hard disk stores a failure risk, that is, a failure prediction result of the hard disk. For example, if the failure prediction result of the hard disk is "yes" or "1", it indicates that there is a failure risk of the hard disk. The result of the hard disk failure prediction is "no" or "0", which indicates that the hard disk is at risk of failure.
In some embodiments, the hard disk failure prediction result is obtained by classifying the probability value of the hard disk failure risk predicted by the hard disk failure prediction model. Therefore, the probability value indicating that the hard disk has a fault risk may reflect the health degree of the hard disk, and thus, the fault prediction information of the hard disk may further include the probability value to analyze the health degree of the hard disk.
S303: and determining a demand reference value corresponding to the user, wherein the demand reference value is used for indicating the acceptability degree of the user on the number of the hard disks with the fault risk.
Different users have different demands, and the acceptable degree of the detected number of the hard disks with fault risks is different for the users with different demands. For example, a user pays attention to the health of a hard disk, that is, is sensitive to a hard disk that is at risk of failure (hereinafter, such a user is referred to as a risk-sensitive user for convenience of description), and then the number of detected hard disks that are at risk of failure is relatively high. For focusing on the cost of replacing hard disks (hereinafter such users are referred to as cost-sensitive users for convenience of description), the number of hard disks at risk of failure detected by the cost-sensitive users may be relatively low. A user who is not concerned about the health of the hard disk and the cost of replacing the hard disk (hereinafter, such a user is referred to as a general type user for convenience of description), the number of detected hard disks at risk of failure by the general type user is moderate. It should be noted that the foregoing is merely an illustration of a preference type of a user, and the preference type of the user is not particularly limited by the present application.
For example, the user's demand reference value may be a target threshold value of the prediction probability and/or the detected number of hard disks at which there is a risk of failure, or the like. The range of the target threshold may be 0-1, the larger the value, the more stringent the criteria for determining that a hard disk is a hard disk that is at risk of failure. The more the number of hard disks that are detected at risk of failure, the more stringent the criteria for determining that a hard disk is at risk of failure. The target threshold for risk-sensitive users may be higher relative to other types of users and/or the number of hard disks at risk of failure may be higher relative to other types of users. The threshold of the prediction probability corresponding to the normal user may be lower than the target threshold of the risk sensitive user, and/or the number of the hard disks with fault risk may be lower than the number of the hard disks with fault risk of the risk sensitive user.
In order to be able to meet the demands of different users, the users of different demand types all correspond to different demand reference values, wherein the demand reference values can indicate the acceptable degree of the user to the detected number of the hard disks with fault risks. The user's demand reference value may be determined in various manners, for example, the user may configure the demand reference value according to the user's own demand through the client; the user can also select a hard disk failure prediction mode or a preference type of the user through the client according to the self demand, and the server determines a demand reference value of the user according to the hard disk failure prediction mode or the preference type of the user.
In some embodiments, as shown in FIG. 4, the user may configure the user-desired reference value via a client-side display 40. The client may send request information to the server, where the request information may include a requirement reference value. For example, the user confirms the detection of the hard disk failure risk through the display screen 40, and then the display screen 40 displays an interface for parameter configuration, which displays the required parameters that the user can configure, such as the target threshold and the detected number. For the convenience of user filling, the parameter configuration interface may also display an interpretation of each required parameter, such as a target threshold value range, a detected number of value ranges, and the like. Wherein the user can fill in the target threshold value and the detected number according to the sensitivity of the user to risk, the sensitivity of the user to cost and the like. The client can send the requirement parameter value carrying and request information to the server. The server receives the request information and analyzes the request information to determine the demand reference value.
Therefore, the user configures the demand parameter value according to the self demand, so that the failure prediction result of the hard disk better meets the demand of the user, and the experience of the user is improved.
In other embodiments, the server may be preconfigured with a plurality of requirement parameter values. Users of different preference types correspond to different values of demand parameters. The server may have stored therein a correspondence between the demand parameter values and the preference types. As shown in fig. 5, the risk-sensitive type corresponds to the demand parameter value μ 1, the cost-sensitive type corresponds to the demand parameter value μ 2, and the normal type corresponds to the demand parameter value μ 3. It should be noted that the value μ1 and μ3 of the demand parameter are only an example, and the present application is not limited to the value of the demand parameter.
The user may pre-configure the corresponding preference type through the client, and the client sends the preference type of the user to the server, so that the server determines the value of the requirement parameter corresponding to the user according to the preference type of the user. Specifically, the client may carry the preference type of the user in the request information, and send the request information to the server. The server side analyzes the request information, so that the preference type of the user is determined.
For example, as shown in fig. 6, the user confirms the detection of the hard disk failure risk through the display screen 60, and then the display screen 40 displays an interface of preference selection. The interface for selecting the preference may further include an explanation (not shown in the figure) for different preference types, for example, the risk sensitivity may be interpreted as that the detection rate of the hard disk with the fault risk is high, which may result in a high false alarm rate. The normal type can be interpreted as a balanced detection rate and a false positive rate. Cost-sensitive may be interpreted as a low false positive rate, possibly resulting in a partially insignificant risk not being detected. The client can carry the preference type selected by the user in the request information and send the request information to the server.
In addition, the selection of the preference type can be replaced by the selection of the detection mode, and different required parameter values corresponding to different detection modes can be replaced. Specific procedures can be found in the selection procedure of preference types and will not be described in detail here.
For example, the server may analyze the behavior data of the user to determine the preference type of the user. The behavior data of the user may refer to data corresponding to behavior in the process of the user using the hard disk, for example, the number of times the user accesses the hard disk (monthly/weekly, etc.), the stay time of each access to the hard disk, the access frequency (daily/weekly, etc.), the content of interest when accessing the hard disk, the duration of the user using the hard disk weekly/monthly/daily, etc. The server may classify behavior data of the user, so as to determine a preference type to which the user belongs.
For example, as shown in fig. 7, the server may classify behavior data of the user through a decision tree classifier. When the number of times the user accesses the hard disk per month is greater than n, the user belongs to preference type 1. When the number of times that the user accesses the hard disk per month is less than or equal to n, the time length of the user using the hard disk is greater than m, and the stay time length of the user accessing the hard disk each time is less than k, the user belongs to preference type 2. When the number of times the user accesses the hard disk per month is less than or equal to n, the duration of the user using the hard disk is less than or equal to m, and the content of interest when the user accesses the hard disk is h, the user belongs to preference type 3. Here, fig. 7 illustrates only a classification flow of behavior data of the user by the decision tree classifier, and the specific classification flow is not particularly limited.
For another example, the server may also generate a user portrait via the user's behavioral data, thereby determining a user's preference type based on the user portrait. It should be noted that the foregoing is merely an example of a manner of determining the preference type of the user, and the manner of determining the preference type of the user is not particularly limited in the present application.
S304: and adjusting the fault prediction results of the plurality of hard disks according to the demand reference value.
Different users have different demands, so that the fault prediction results of the plurality of hard disks can be adjusted according to the demand reference values corresponding to the users, so that the final fault prediction results are more in line with the demands of the users, and the experience of the users is improved. Here, the failure prediction result of the hard disk may also be adjusted in conjunction with a probability value for indicating that the hard disk is at risk of failure. Wherein the failure prediction information further comprises a probability value for indicating that the hard disk is at risk of failure.
In some embodiments, the failure prediction result of the hard disk may be adjusted directly according to the demand reference value. For example, the demand reference value is a target threshold value and a detected number. The number of the hard disks with the fault risk can be compared with the number of the hard disks with the fault risk, and the fault prediction result of the hard disks, namely the adjusted fault prediction result, can be regenerated according to the comparison result, the target threshold value and the probability value of the hard disks. For example, the user has 100 hard disks, 10 of the 100 hard disks are hard disks with a risk of failure, and the detected number is 18. I.e. the number of hard disks at risk of failure is smaller than the detected number. The target threshold is 0.59 and the preset threshold is 0.62. And regenerating a fault prediction result of the hard disk according to the target threshold value 0.59 and the probability value of the hard disk. Wherein, the hard disk which is finally determined to have fault risks has 16.
In other embodiments, the probability values of the plurality of hard disks may be adjusted according to the demand reference value and a preset threshold value. Illustratively, the user is a cost sensitive user, who owns 100 hard disks, and the number of hard disks detected for which there is a risk of failure per month is 5. The failure detection result including 8 hard disks among the failure detection results of 100 hard disks is yes. Since 8 is greater than 5, the probability value of the hard disk may be adjusted according to a preset threshold value such that the number of hard disks having a probability value greater than the threshold value is less than or equal to 5.
In yet another embodiment, the preset threshold may be adjusted according to the demand reference value, and a new failure prediction result, that is, an adjusted failure prediction result, is generated according to the adjusted threshold and the probability value of the hard disk. For example, the demand reference value includes a target threshold value, a preset threshold value may be adjusted to the target threshold value, and then a failure prediction result of the hard disk is regenerated according to the adjusted threshold value and the probability value of the hard disk.
Therefore, the fault prediction results of the plurality of hard disks can be adjusted according to the requirements of the users, so that the fault prediction results more meet the requirements of the users, and the experience of the users is improved.
In order to enable the user to determine the health degree of the hard disk more intuitively, the probability value of the hard disk may be converted. As shown in fig. 8, a probability value corresponding to the hard disk is determined according to the failure prediction information of the hard disk. And then, converting the probability value corresponding to the hard disk to obtain the health degree of the hard disk. And then, the health degree of the hard disk at a plurality of moments is plotted according to time sequence, and a health degree change trend graph of the hard disk is generated, so that a user can more intuitively determine the health degree of the hard disk.
Specifically, as shown in fig. 9, the method for generating health degree information provided by the embodiment of the present application may include:
s901: and converting the probability value of the hard disk into the health value of the hard disk according to the preset probability conversion value.
The probability transition threshold may be used to determine whether the hard disk is a healthy hard disk or a risky hard disk, e.g., if the health value of the hard disk is greater than or equal to 60, then the hard disk is healthy, and if the health of the hard disk is less than 60, then the hard disk is risky.
Illustratively, the health value H may satisfy the following formula (1), and the formula (1) function image is shown in fig. 10:
where P1 represents a probability transition threshold and P2 represents a probability value of the hard disk.
S902: and sending the health value of the hard disk to the client so that the user can determine the health of the hard disk.
The health degree value of the hard disk is sent to the client, and the client can generate a time-dependent change trend graph of the health degree of the hard disk with time according to the health degree values of the hard disk stored by the client at a plurality of moments in a historical time period, as shown in fig. 11, wherein "-" indicates a health degree transformation trend of the hard disk in 100 days, calculated forward from the current moment, for example, -80 indicates 80 days calculated forward from the current moment. Therefore, the user can more intuitively determine the current health degree of the hard disk and the health degree change trend of the hard disk.
In this embodiment, the probability value of the hard disk may be converted into the health value of the hard disk, so that the user may determine the health of the hard disk more intuitively. The user can compare the health degree values of the hard disk at the current time with the health degree values of the hard disk at the current time through the plurality of historical time hard disk health degree values stored by the client, so that the user can know the change condition of the health degree of the hard disk at the time, and the use experience of the user is improved.
In some embodiments, as shown in fig. 12, the method for generating health degree information according to the embodiment of the present application may further include the following steps:
S1201: and obtaining health degree values of the hard disk at a plurality of moments in a historical time period.
The server may store health values for the hard disk at multiple times during the historical time period. The server side can call the health degree values of the hard disk at a plurality of moments in the calendar history time period to generate a change trend chart of the health degree of the hard disk along with time.
S1202: and generating health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in the historical time period and the health degree values of the hard disk.
S1203: and sending the health degree information of the plurality of hard disks to the client so that the client can determine a time-dependent change trend chart of the health degree of the hard disks.
The health degree information of the hard disk can be a time-dependent change trend chart of the health degree of the hard disk, and the server side can send the time-dependent change trend chart of the health degree of the hard disk to the client side. The health degree information of the hard disk can also be health degree values corresponding to each moment of the hard disk and each moment, the server side sends the health degree information to the client side, and the client side generates a change trend chart of the health degree of the hard disk along with time according to the health degree values of the hard disk at each moment.
In this embodiment, the server may send the health value of the hard disk at multiple times and the health value of the current time in the historical time period to the client according to the summary, so that the client generates a trend chart of the change of the hard disk with time, so that the user can more intuitively observe the change of the health value of the hard disk with time, and the experience of the user is improved.
In order to ensure the accuracy of the hard disk fault prediction model, the hard disk fault prediction model can be optimized. Specifically, the server stores a history sample set for training a hard disk failure prediction model. As shown in fig. 13, the optimization method of the hard disk failure prediction model provided by the embodiment of the present application may include:
s1301: and acquiring a feedback sample set fed back by the user, wherein each feedback sample in the feedback sample set comprises characteristic data of a target hard disk and tag information used for indicating whether the target hard disk fails, and the target hard disk is a hard disk with failure prediction results inconsistent with the tag information in a plurality of hard disks.
After receiving the fault prediction result of the hard disk, the user can observe the hard disk and record the health state of the hard disk. When the health state of the hard disk is observed, the hard disk with the failure prediction result being not reported or being wrongly reported can be fed back. The feedback sample comprises characteristic data of the target hard disk, such as the model number of the hard disk, the production date of the hard disk and the like, and tag information of each target hard disk, wherein the tag information is used for indicating whether the hard disk fails or not. Here, the target hard disk is a hard disk whose failure prediction result is inconsistent with the tag information, for example, the failure prediction result of the hard disk is yes, that is, there is a failure risk, but after one month, the failure prediction result of the hard disk is a false alarm result, and the hard disk is the target hard disk. For another example, if the failure prediction result of the hard disk is "no", that is, there is no failure risk, but if the hard disk fails after half a month, the failure prediction result of the hard disk is a missing report result. The hard disks corresponding to the false alarm and missing alarm results can be regarded as the hard disks with failure prediction results inconsistent with the label information, and the hard disks are target hard disks.
S1302: and carrying out statistical analysis on the characteristic data of the plurality of hard disks in the historical sample set and the characteristic data of the plurality of target hard disks in the feedback sample set, and determining difference information between the historical sample set and the feedback sample set.
The historical sample set may be stored at the server. The server side can call the characteristic data of the plurality of hard disks in the history sample set, and perform statistical analysis on the characteristic data of the plurality of hard disks in the history sample set and the characteristic data of the plurality of target hard disks in the feedback sample set, so as to determine difference information between the history sample set and the feedback sample set. For example, it may be analyzed whether the model of the hard disk in the history sample set contains the model of the target hard disk in the feedback sample set. For another example, the hard disk in the history sample set and the target hard disk in the feedback sample set may be analyzed for differences in distribution of the characteristic data, and the like.
S1303: and determining a target training sample set from the historical sample set and the feedback sample set according to the difference information.
According to the difference information, a target training sample set for optimizing the hard disk failure prediction model can be determined. And if the difference between the feedback sample sets of the historical sample set is large, taking the historical sample set and the feedback sample set as target training sample sets. For example, if the model of the hard disk in the history sample set does not include the model of the target hard disk, both the history sample set and the feedback sample set are used as the target training sample set. If the difference between the feedback sample sets of the historical sample sets is smaller, the feedback sample sets are all used as target training sample sets
S1304: and optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
And if the target training sample set comprises a history sample set and a feedback sample set, retraining the hard disk fault prediction model. And if the target training sample set only comprises the feedback sample set, performing optimization training on the hard disk fault prediction model by using the feedback sample set.
In this embodiment, by performing statistical analysis on the characteristic data of the hard disk fed back by the user and the characteristic data of each hard disk in the history sample set, it is able to determine difference information between the history sample set and the feedback sample set, for example, whether the model of the hard disk in the history sample set, the manufacturer, and the like include the feedback sample, the model of the middle hard disk, and the manufacturer of the hard disk, so as to determine the sample for optimizing the hard disk failure prediction model from the history sample set and the feedback sample set, thereby improving the accuracy of the hard disk failure prediction model.
Fig. 14 is a schematic structural diagram of a hard disk failure prediction apparatus 1400 according to an embodiment of the present application. The hard disk failure prediction device can be applied to a server. As shown in fig. 14, the hard disk failure prediction apparatus 1400 provided by the embodiment of the present application may include:
An obtaining module 1401, configured to obtain feature data of a hard disk of a user, where the feature data is data of factors affecting a hard disk failure;
the prediction module 1402 is configured to input feature data of the hard disk into a hard disk failure prediction model to perform failure prediction of the hard disk, so as to obtain failure prediction information of the hard disk, where the failure prediction information includes failure prediction results of the hard disk, and the failure prediction results are used to indicate whether the hard disk is a hard disk with failure risk;
a determining module 1403, configured to determine a demand reference value corresponding to a user, where the demand reference value is used to indicate an acceptable degree of the user to the number of hard disks that have a risk of failure;
the adjusting module 1404 is configured to adjust failure prediction results of the plurality of hard disks according to the demand reference value;
a sending module 1405, configured to send the failure prediction results after the adjustment of the plurality of hard disks to the client.
In one possible implementation, the fault prediction information further includes a probability value for indicating that the hard disk is at risk of a fault, and the adjusting module 1404 is configured to:
according to the demand reference value and a preset threshold value, adjusting probability values of a plurality of hard disks;
and generating fault prediction results after the adjustment of the plurality of hard disks according to the probability values and the threshold values after the adjustment of the plurality of hard disks.
According to the embodiment of the application, the hard disk fault prediction result predicted by using the hard disk fault prediction model is adjusted based on the demand reference value, so that the adjusted fault prediction result can better meet the demands of users. The requirement reference value can represent the acceptable degree of the user to the number of the hard disks with fault risks in the plurality of the hard disks owned by the user, for example, the user is sensitive to the cost of hard disk replacement, i.e. the acceptable degree of the user to the number of the hard disks with fault risks is low; for another example, in order to avoid that the hard disk failure affects the production efficiency, etc., the sensitivity of the user to the hard disk failure is high, that is, the user has a high acceptable level for the number of hard disks with failure risk. Therefore, the fault prediction result is adjusted according to the acceptable degree of the number of the hard disks with faults by different users, so that the fault prediction result meets the requirements of the users, the individuation is realized, and the experience degree of the users is improved.
In one possible implementation, the fault prediction information further includes a probability value for indicating that the hard disk is at risk of a fault, and the adjusting module 1404 is configured to:
adjusting a threshold according to the demand reference value;
And generating a fault prediction result after hard disk adjustment according to the probability value of the hard disk and the adjusted threshold value.
In one possible implementation, the determining module 1403 is configured to:
receiving a preference type of a user sent by a client;
according to the preference type, a demand reference value is determined.
In one possible implementation, the determining module 1403 is configured to:
acquiring user behavior data of a user using a hard disk;
classifying the users according to the user behavior data, and determining the preference type of the users.
In one possible implementation, the apparatus further includes:
the conversion module is used for converting the probability value of the hard disk into the health value of the hard disk according to a preset probability conversion value;
the sending module 1405 is further configured to send the health value of the hard disk to the client, so that the user determines the health of the hard disk.
In one possible implementation, the apparatus further includes:
an obtaining module 1401, configured to obtain health values of the hard disk at a plurality of moments in a historical time period;
the generation module is used for generating health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in a historical time period and the health degree values of the hard disk;
the sending module 1405 is configured to send the health information of the plurality of hard disks to the client, so that the client determines a trend chart of health of the hard disks over time.
In one possible implementation manner, the server stores a history sample set for training a hard disk failure prediction model, and the apparatus further includes:
an obtaining module 1401, configured to obtain a feedback sample set fed back by a user, where each feedback sample in the feedback sample set includes feature data of a target hard disk and tag information for indicating whether the target hard disk is faulty, and the target hard disk is a hard disk in which a fault prediction result in multiple hard disks is inconsistent with the tag information;
the determining module 1404 is further configured to perform statistical analysis on the characteristic data of the plurality of hard disks in the history sample set and the characteristic data of the plurality of target hard disks in the feedback sample set, and determine difference information between the history sample set and the feedback sample set; determining a target training sample set from the historical sample set and the feedback sample set according to the difference information;
and the optimization module is used for optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
As shown in FIG. 15, the present application also provides a computer device 1500. The computer device 1500 may be any of the computer devices mentioned in connection with the embodiments corresponding to fig. 1-14 above. The computer device 1500 may include: processor 1510, external memory interface 1520, internal memory 1515, USB interface 1530, charge management module 1540, power management module 1541, battery 1542, antenna 1, antenna 2, mobile communication module 1550, wireless communication module 1560, and the like.
It should be understood that the architecture shown in the exemplary embodiments of the present application is not intended to limit the computer device 1500 in any particular manner. Computer device 1500 may include more or fewer components than shown, or may combine certain components, or split certain components, or a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
The processor 1510 may be an ARM, X86, MIPS, or other architecture processor. The processor 1510 may include one or more processing units, such as: an application processor (application processor, AP), a modem processor, a GPU, an ISP, a transmitter, a video codec, a DSP, a baseband processor and/or an NPU, etc. Wherein the different processing units may be separate devices or may be integrated in one or more processors.
The transmitter can generate an operation transmitting signal according to the instruction operation code and the time sequence signal to finish the instruction fetching and the instruction executing.
A memory may also be provided in the processor 1510 for storing instructions and data. In some embodiments of the application, the memory in processor 1510 is a cache memory. The memory may hold instructions or data that the processor 1510 has just used or recycled. If the processor 1510 needs to reuse the instruction or data, it may be called directly from memory. Repeated accesses are avoided, reducing the latency of the processor 1510 and thus improving the efficiency of the system.
In some embodiments of the application, the processor 1510 may include one or more interfaces. The interfaces may include an I2C interface, an I2S interface, a PCM interface, a UART interface, a MIPI, a GPIO interface, a SIM interface, and/or a USB interface, among others.
The charge management module 1540 is for receiving charge input from a charger. The charger can be a wireless charger or a wired charger.
In some wired charging embodiments, the charge management module 1540 may receive the charging input of the wired charger through the USB interface 1530. In some wireless charging embodiments, the charge management module 1540 may receive wireless charging input through a wireless charging coil of the computer device 1500. The computer device 1500 may also be powered by the power management module 1541 while the battery 1542 is charged by the charge management module 1540.
The wireless communication function of the computer device 1500 can be implemented by an antenna 1, an antenna 2, a mobile communication module 1550, a wireless communication module 1560, a modem processor, a baseband processor, and the like.
The mobile communication module 1550 may provide a solution for wireless communication including 2G/3G/4G/5G/6G, etc., as applied on the computer device 1500. The wireless communication module 1560 may provide a solution for wireless communications including WLAN, BT, GNSS, FM, NFC, zigbee and IR, etc., as applied to the computer device 1500. The WLAN may be, for example, a Wi-Fi network.
External memory interface 1520 may be used to connect external memory cards, such as Micro SD cards, to enable expansion of the memory capabilities of computer device 1500. The external memory card communicates with the processor 1510 through an external memory interface 1520 to implement data storage functions.
The internal memory 1515 may be used to store computer-executable program code that includes instructions. The internal memory 1515 may include a stored program area and a stored data area. The storage program area may store an operating system, an application program required for at least one function, and the like. The storage data area may store data created during use of the computer device 1500, and the like. In addition, the internal memory 1515 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, UFS, and the like. The processor 1510 executes various functional applications of the computer device 1500 and data processing by executing instructions stored in the internal memory 1515 and/or stored in a memory provided within the processor.
The computer device 1500 provided in the present application may implement any one of the methods or any one of the apparatuses described in fig. 1 to 14 related to the computer device, and specific implementation manner may refer to corresponding descriptions in fig. 1 to 14, which are not repeated herein.
As shown in fig. 16, the present application also provides a server 1600. The server 1600 may be any of the servers mentioned in connection with the corresponding embodiments of fig. 1-14. The server 1600 may include: processor 1610, external memory interface 1620, internal memory 1616, USB interface 1630, charge management module 1640, power management module 1641, battery 1642, antenna 1, antenna 2, mobile communication module 1650, wireless communication module 1660, and the like.
It should be understood that the illustrated structure of the embodiment of the present application does not constitute a specific limitation on the server 1600. Server 1600 may include more or fewer components than shown, or may combine certain components, or split certain components, or a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
The processor 1610 may be an ARM, X86, MIPS, etc. architecture processor. Processor 1610 may include one or more processing units such as: an application processor (application processor, AP), a modem processor, a GPU, an ISP, a transmitter, a video codec, a DSP, a baseband processor and/or an NPU, etc. Wherein the different processing units may be separate devices or may be integrated in one or more processors.
The transmitter can generate an operation transmitting signal according to the instruction operation code and the time sequence signal to finish the instruction fetching and the instruction executing.
A memory may also be provided in processor 1610 for storing instructions and data. In some embodiments of the application, the memory in processor 1610 is a cache memory. The memory may hold instructions or data that is just used or recycled by the processor 1610. If the processor 1610 needs to reuse the instruction or data, it may be called directly from memory. Repeated accesses are avoided, reducing the latency of the processor 1610, and thus improving the efficiency of the system.
In some embodiments of the application, processor 1610 may include one or more interfaces. The interfaces may include an I2C interface, an I2S interface, a PCM interface, a UART interface, a MIPI, a GPIO interface, a SIM interface, and/or a USB interface, among others.
The charge management module 1640 is used to receive charge input from a charger. The charger can be a wireless charger or a wired charger.
In some wired charging embodiments, the charge management module 1640 may receive a charging input of the wired charger through the USB interface 1630. In some wireless charging embodiments, the charge management module 1640 may receive wireless charging input through a wireless charging coil of the server 1600. While charge management module 1640 charges battery 1642, server 1600 may also be powered through power management module 1641.
The wireless communication functions of the server 1600 may be implemented by an antenna 1, an antenna 2, a mobile communication module 1650, a wireless communication module 1660, a modem processor, a baseband processor, and the like.
The mobile communication module 1650 may provide a solution for wireless communication including 2G/3G/4G/5G/6G, etc., applied on the server 1600. The wireless communication module 1660 may provide a solution for wireless communication including WLAN, BT, GNSS, FM, NFC, zigbee and IR, etc., applied on the server 1600. The WLAN may be, for example, a Wi-Fi network.
The external memory interface 1620 may be used to connect an external memory card, such as a Micro SD card, to realize the memory capability of the extension server 1600. The external memory card communicates with the processor 1610 through an external memory interface 1620 to implement data storage functions.
The internal memory 1616 may be used to store computer-executable program code that includes instructions. The internal memory 1616 may include a storage program area and a storage data area. The storage program area may store an operating system, an application program required for at least one function, and the like. The storage data area may store data created during use of the server 1600, etc. In addition, internal memory 1616 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, UFS, and the like. Processor 1610 performs various functional applications and data processing of server 1600 by executing instructions stored in internal memory 1616, and/or instructions stored in a memory disposed in the processor.
The server 1600 provided by the present application may implement any one of the methods or any one of the apparatuses related to the server described in fig. 1 to 14, and specific implementation manner may refer to corresponding descriptions in fig. 1 to 14, which are not repeated herein.
In addition, in combination with the above embodiment, the embodiment of the application also provides a computer storage medium for implementation. The computer storage medium has stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement a method of determining a port attribute or a method of forwarding data in any of the above embodiments.
In addition, in combination with the above embodiment, the embodiment of the present application further provides a computer program product containing instructions, which when executed on a computer, cause the computer to execute any one of the method for determining the port attribute or the method for forwarding data.
The functional blocks shown in the above block diagrams may be implemented in hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the application are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine readable medium or transmitted over a transmission medium or a communication link by a data signal carried in a carrier wave. A "machine-readable medium" may include any medium that can store or transfer information. Examples of machine-readable media include electronic circuitry, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio Frequency (RF) links, and the like. The code segments may be downloaded via computer networks such as the internet, intranets, etc.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted across a computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). Computer readable storage media can be any available media that can be accessed by a computer or data storage devices, such as servers, data centers, etc., that contain an integration of one or more available media. Usable media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid State Disks (SSDs)), among others.
It should also be noted that the exemplary embodiments mentioned in this disclosure describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, or may be performed in a different order from the order in the embodiments, or several steps may be performed simultaneously.
Aspects of the present application are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions/acts specified in the flowchart and/or block diagram block or blocks. Such a processor may be, but is not limited to being, a general purpose processor, a special purpose processor, an application specific processor, or a field programmable logic circuit. It will also be understood that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware which performs the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Those of ordinary skill would further appreciate that the elements and algorithm steps of the embodiments described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the various illustrative components and steps have been described above generally in terms of function in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The present application is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present application are intended to be included in the scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims (19)

1. The hard disk fault prediction method is characterized by being applied to a server and comprising the following steps of:
Acquiring characteristic data of a hard disk of a user, wherein the characteristic data are data of factors affecting the hard disk fault;
inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk to obtain fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk;
determining a demand reference value corresponding to the user, wherein the demand reference value is used for indicating the acceptability degree of the user on the number of hard disks with fault risks;
according to the demand reference value, adjusting fault prediction results of a plurality of hard disks;
and sending a plurality of failure prediction results after hard disk adjustment to a client.
2. The method of claim 1, wherein the failure prediction information further includes a probability value indicating that the hard disk is at risk of failure, and wherein adjusting the failure prediction results of the plurality of hard disks according to the demand reference value includes:
according to the demand reference value and a preset threshold value, adjusting probability values of a plurality of hard disks;
And generating a plurality of fault prediction results after hard disk adjustment according to the probability values after hard disk adjustment and the threshold values.
3. The method of claim 1, wherein the failure prediction information further includes a probability value indicating that the hard disk is at risk of failure, and wherein adjusting the failure prediction results of the plurality of hard disks according to the demand reference value includes:
according to the demand reference value, a preset threshold value is adjusted;
and generating a fault prediction result after the hard disk is adjusted according to the probability value of the hard disk and the adjusted threshold value.
4. The method of claim 1, wherein the determining the user's corresponding demand reference value comprises:
determining a preference type to which the user belongs;
and determining the demand reference value according to the preference type.
5. The method of claim 4, wherein the determining the type of preference to which the user belongs comprises:
acquiring user behavior data of the user using a hard disk;
classifying the users according to the user behavior data, and determining the preference type of the users.
6. The method according to claim 1, wherein the method further comprises:
Converting the probability value of the hard disk into a health value of the hard disk according to a preset probability conversion value;
and sending the health value of the hard disk to the client so that a user can determine the health of the hard disk.
7. The method of claim 6, wherein the method further comprises:
obtaining health values of the hard disk at a plurality of moments in a historical time period;
generating health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in a historical time period and the health degree values of the hard disk;
the sending the health value of the hard disk to the client so as to enable the user to determine the health of the hard disk comprises the following steps:
and sending the health degree information of the plurality of hard disks to the client so that the client can determine a change trend chart of the health degree of the hard disks along with time.
8. The method of any of claims 1-7, wherein the server stores a set of historical samples for training the hard disk failure prediction model, the method further comprising:
acquiring a feedback sample set fed back by the user, wherein each feedback sample in the feedback sample set comprises characteristic data of a target hard disk and tag information used for indicating whether the target hard disk fails, and the target hard disk is a hard disk with failure prediction results inconsistent with the tag information in a plurality of hard disks;
Carrying out statistical analysis on the characteristic data of a plurality of hard disks in the history sample set and the characteristic data of a plurality of target hard disks in the feedback sample set, and determining difference information between the history sample set and the feedback sample set;
determining a target training sample set from the historical sample set and the feedback sample set according to the difference information;
and optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
9. The hard disk fault prediction device is characterized by being applied to a server and comprising:
the system comprises an acquisition module, a storage module and a control module, wherein the acquisition module is used for acquiring characteristic data of a hard disk of a user, wherein the characteristic data is data of factors affecting the hard disk fault;
the prediction module is used for inputting the characteristic data of the hard disk into a hard disk fault prediction model to perform fault prediction of the hard disk to obtain fault prediction information of the hard disk, wherein the fault prediction information comprises a fault prediction result of the hard disk, and the fault prediction result is used for indicating whether the hard disk is a hard disk with fault risk;
the determining module is used for determining a demand reference value corresponding to the user, wherein the demand reference value is used for indicating the acceptability of the user on the number of hard disks with fault risks;
The adjusting module is used for adjusting the fault prediction results of the plurality of hard disks according to the demand reference value;
and the sending module is used for sending the failure prediction results after the hard disk adjustment to the client.
10. The apparatus of claim 9, wherein the failure prediction information further comprises a probability value indicating that the hard disk is at risk of failure, the adjustment module to:
according to the demand reference value and a preset threshold value, adjusting probability values of a plurality of hard disks;
and generating a plurality of fault prediction results after hard disk adjustment according to the probability values after hard disk adjustment and the threshold values.
11. The apparatus of claim 9, wherein the failure prediction information further comprises a probability value indicating that the hard disk is at risk of failure, the adjustment module to:
according to the demand reference value, a preset threshold value is adjusted;
and generating a fault prediction result after the hard disk is adjusted according to the probability value of the hard disk and the adjusted threshold value.
12. The apparatus of claim 9, wherein the determining module is configured to:
determining a preference type to which the user belongs;
And determining the demand reference value according to the preference type.
13. The apparatus of claim 12, wherein the determining module is configured to:
acquiring user behavior data of the user using a hard disk;
classifying the users according to the user behavior data, and determining the preference type of the users.
14. The apparatus of claim 9, wherein the apparatus further comprises:
the conversion module is used for converting the probability value of the hard disk into the health value of the hard disk according to a preset probability conversion value;
the sending module is further configured to send the health value of the hard disk to the client, so that a user determines the health of the hard disk.
15. The apparatus of claim 14, wherein the apparatus further comprises:
the acquisition module is used for acquiring health values of the hard disk at a plurality of moments in a historical time period;
the generation module is used for generating the health degree information of the hard disk according to the health degree values of the hard disk at a plurality of moments in the historical time period and the health degree values of the hard disk;
the sending module is used for sending the health degree information of the plurality of hard disks to the client so as to enable the client to determine a change trend chart of the health degree of the hard disks along with time.
16. The apparatus of any one of claims 9-15, wherein the server stores a set of historical samples for training the hard disk failure prediction model, the apparatus further comprising:
the system comprises an acquisition module, a feedback sample set and a storage module, wherein the acquisition module is used for acquiring a feedback sample set fed back by a user, each feedback sample in the feedback sample set comprises characteristic data of a target hard disk and label information used for indicating whether the target hard disk is faulty, and the target hard disk is a hard disk with fault prediction results inconsistent with the label information in a plurality of hard disks;
the determining module is further configured to perform statistical analysis on characteristic data of a plurality of hard disks in the historical sample set and characteristic data of a plurality of target hard disks in the feedback sample set, and determine difference information between the historical sample set and the feedback sample set; determining a target training sample set from the historical sample set and the feedback sample set according to the difference information;
and the optimization module is used for optimizing the hard disk fault prediction model according to the target training sample set to obtain an optimized hard disk fault prediction model.
17. A hard disk failure prediction system, characterized in that the system comprises a client and a server, wherein the system is configured to perform the method according to any of claims 1-8.
18. A computer device, comprising:
at least one memory for storing a program;
at least one processor for executing the memory-stored program, which processor is adapted to perform the method according to any of claims 1-8, when the memory-stored program is executed.
19. A computer storage medium having instructions stored therein which, when executed on a computer, cause the computer to perform the method of any of claims 1-8.
CN202210605693.1A 2022-05-31 2022-05-31 Hard disk fault prediction method, device, equipment and storage medium Pending CN117215812A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210605693.1A CN117215812A (en) 2022-05-31 2022-05-31 Hard disk fault prediction method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210605693.1A CN117215812A (en) 2022-05-31 2022-05-31 Hard disk fault prediction method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117215812A true CN117215812A (en) 2023-12-12

Family

ID=89044929

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210605693.1A Pending CN117215812A (en) 2022-05-31 2022-05-31 Hard disk fault prediction method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117215812A (en)

Similar Documents

Publication Publication Date Title
JP7451479B2 (en) Systems and methods for collecting, tracking, and storing system performance and event data about computing devices
US11321210B2 (en) System and method for cloud-device collaborative real-time user experience and performance abnormality detection
US20220351081A1 (en) Model training method and apparatus
EP3654186A1 (en) Automated identification of device status and resulting dynamic modification of device operations
US10719769B2 (en) Systems and methods for generating and communicating application recommendations at uninstall time
CN108268366B (en) Test case execution method and device
CN111491552B (en) Sensor channels grouped by context for healthcare monitoring
KR20160132394A (en) Behavioral analysis for securing peripheral devices
US11722371B2 (en) Utilizing unstructured data in self-organized networks
CN110650503A (en) Network access method, device, system and computer readable storage medium
US11095735B2 (en) Configuration of event data communication in computer networks
CN117215812A (en) Hard disk fault prediction method, device, equipment and storage medium
US20220303227A1 (en) Facilitating identification of background browsing traffic in browsing history data in advanced networks
JP2022536048A (en) Diagnostic test prioritization based on accumulated diagnostic reports
CN115565215B (en) Face recognition algorithm switching method and device and storage medium
CN117097635B (en) Method, device, storage medium and equipment for calling link sampling
CN115640081A (en) Business case decision method and device, electronic equipment and storage medium
WO2024110160A1 (en) Ml model transfer and update between ue and network
WO2022241335A1 (en) Systems for controlling acquisition of test data from devices
CN116670654A (en) Device diagnosis and troubleshooting employing a multi-layer architecture

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination