Information processing device, information processing method and program

Abstract

To provide highly convenient failure prediction and a related service thereof for users of a storage medium.SOLUTION: A smart information acquisition part 731 in a server 1 for performing failure prediction diagnosis before a failure of a storage device SD, acquires from the storage device SD, values of N (N is an integer value of 2 or more) items of smart information, etc. representing a state of the storage device SD. A comprehensive state calculation part 732 calculates a comprehensive state value representing a comprehensive state of the storage device SD from the values of the N items. An item diagnosis part 733 determines an item diagnosis result according to values of at least some of the N items, a first index set for each item and a first determination method. A comprehensive diagnosis part 734 determines a comprehensive diagnosis result according to a comprehensive state value, a second index, and a second determination method. A final diagnosis result determination part 735 determines a failure prediction diagnosis result of the storage device SD on the basis of the item diagnosis result and the comprehensive diagnosis result.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Conventionally, storage media such as hard disk drives have been equipped with a function for performing self-diagnosis on the storage media itself and acquiring the resulting SMART information (SMART function: Self-Monitoring, Analysis and Reporting Technology function). There is also technology for diagnosing hard disk failures using the SMART information (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-173809 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the prior art including the above-mentioned Patent Document 1, the occurrence of a failure was merely identified when a predetermined parameter included in the smart information exceeded a predetermined value. Furthermore, even if the failure was detected when it was still minor, that is, at an early stage, once a storage medium failed, the failure progressed rapidly, and it was often the case that the data stored on the storage medium could not be read even if an attempt was made to copy the data. Thus, there has been a demand for fault prediction and related services that are highly convenient for users of storage media.

[0005] The present invention has been made in view of the above circumstances, and aims to provide highly convenient failure prediction and related services for users of storage media. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing device according to one aspect of the present invention comprises: An information processing device that performs a failure prediction diagnosis before a failure occurs on a storage medium, smart information acquisition means for acquiring values ​​of N items (N is an integer value of 2 or more) of smart information items indicating the state of the storage medium from the storage medium to be predicted for failure; a comprehensive status calculation means for calculating a comprehensive status value indicating a comprehensive status of the target storage medium from the values ​​of the N items acquired from the target storage medium according to a predetermined calculation method using the N items as parameters; an item diagnosis means for comparing at least some of the N item values ​​acquired from the target storage medium with a first index set for each item, and determining an item diagnosis result related to failure prediction according to a predetermined first judgment method using the comparison result; a comprehensive diagnosis means for comparing the comprehensive state value with a second index and determining a comprehensive diagnosis result related to failure prediction according to a predetermined second judgment method using the result of the comparison; a failure prediction diagnosis means for determining a failure prediction diagnosis result for the target storage medium based on the item diagnosis result and the overall diagnosis result; Equipped with.

[0007] An information processing method and a program according to one aspect of the present invention are information processing methods and programs corresponding to the information processing devices according to the above-described one aspect of the present invention, respectively. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide highly convenient failure prediction and related services to users of storage media. [Brief explanation of the drawings]

[0009] [Figure 1]1 is a diagram showing an example of a configuration of an information processing system including a server according to a first embodiment of the information processing device of the present invention. [Figure 2] 2 is a block diagram showing an example of a hardware configuration of a server in the information processing system of FIG. 1. FIG. [Figure 3] 2 is a block diagram showing an example of the hardware configuration of a user terminal in the information processing system of FIG. 1. FIG. [Figure 4] 4 is a diagram illustrating an example of a functional configuration of an information processing system including the server and user terminals illustrated in FIGS. 2 and 3. FIG. [Figure 5] 5 is a diagram showing an example of conditions for failure prediction executed by a server having the functional configuration of FIG. 4. [Figure 6] 5 is a diagram illustrating an outline of a method for calculating a total state value by the server in FIG. 4. FIG. [Figure 7] 5 is a diagram showing an example of a failure prediction result generated by the server of FIG. 4 and presented to a user via a user terminal. FIG. [Figure 8] 8 is a diagram showing an example of a failure prediction result generated by the server of FIG. 4 and presented to a user via a user terminal, and is a diagram showing an example different from that of FIG. 7. FIG. [Figure 9] FIG. 5 is a diagram illustrating an example of a proactive response of the present service, which is realized by a server having the functional configuration of FIG. [Figure 10] FIG. 5 is a diagram illustrating the characteristics of a failure prediction technique using a server having the functional configuration of FIG. 4. [Figure 11] 5 is a diagram showing an example of a screen including the results of a failure prediction diagnosis performed on a user terminal by a server having the functional configuration of FIG. 4. FIG. [Figure 12] FIG. 10 is a diagram showing an example of an outline of the present service of the second embodiment using a user terminal according to an embodiment of the information processing device of the present invention. [Figure 13] FIG. 10 is a diagram illustrating the overall flow of the present service according to the second embodiment. [Figure 14] FIG. 10 is a diagram illustrating a basic commercial flow of the present service according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] First, a brief description will be given of a service (hereinafter referred to as "this service") to which an information processing system including a server according to an embodiment of the information processing device of the present invention is applied. This service predicts failures in storage devices used in information processing devices, such as hard disk drives and solid state drives. In other words, this service acquires smart information from a storage device installed in an information processing device (hereinafter referred to as a "user terminal") managed by the user, performs a failure prediction diagnosis of the storage device based on the smart information, and notifies the user of the results of the failure prediction diagnosis, etc.

[0011] Here, SMART information refers to information about the status of a storage device that can be obtained using the SMART function, such as the product specifications of the storage device, self-diagnosis results, and information about the extent of use. Specifically, for example, the SMART information includes information on items such as the frequency of various errors occurring in the storage device, temperature, cumulative usage time, number of startups, and usage information on bad sectors, pending sectors, and uncorrected sectors. For example, the SMART information in this example also includes information on increased torque to compensate for deterioration of the spindle motor or bearings. For example, the SMART information may also include parameters such as total LBAs written (number of sectors) / total LBAs read. These items are merely examples used in this example, and the SMART information includes information on a variety of items related to the status of the storage device. Furthermore, if a virtual server is configured in the user terminal, information necessary for configuring the virtual server may also be included in the smart information. By combining this information contained in smart information, this service is able to perform highly accurate failure prediction diagnosis.

[0012] Furthermore, the failure prediction diagnosis of this service employs multiple levels set from the perspective of the probability of a storage device failing within a predetermined time period after the prediction (e.g., from the perspective of failure probability). That is, of these multiple levels, the predetermined level that corresponds to the storage device being diagnosed is determined as the failure prediction diagnosis result. This allows the user to take action for the storage device in accordance with the predetermined level. In the following example, the storage device is a hard disk drive, and a predetermined level out of five levels is determined as the failure prediction diagnosis result.

[0013] An information processing system including a server according to an embodiment of an information processing device of the present invention, and first and second embodiments to which the information processing system is applied will be described below in order. [First embodiment]

[0014] FIG. 1 is a diagram showing an example of the configuration of an information processing system including a server according to a first embodiment of an information processing device of the present invention. The information processing system shown in Figure 1 is configured to include a server 1 managed by a service provider of this service and user terminals 2-1 to 2-n used by n users (n is any integer value greater than or equal to 1). The server 1 and each of the user terminals 2-1 to 2-n are connected to one another via a predetermined network NW such as the Internet. Here, it is assumed that each of the user terminals 2-1 to 2-n includes a storage device SD-1 to SD-n, as shown in FIG. The user terminal 2 has installed therein dedicated application software (for example, a failure prediction program, which will be described later) that allows the user U to receive the service.

[0015] In the following, when there is no need to distinguish between the user terminals 2-1 to 2-n individually, they will be collectively referred to as the "user terminal 2." Furthermore, when referring to the user terminal 2, the storage devices SD-1 to SD-n will be collectively referred to as the "storage device SD."

[0016] First, as a premise, even before providing this service, the service provider of this service has been providing a data recovery service for recovering data from failed storage devices SD in response to user requests. Therefore, the service provider was provided by the user with a storage device SD that had already failed or was immediately before the failure (e.g., malfunctioning), and was able to copy data from that storage device SD. Therefore, the service provider possesses data such as the cause of the storage device SD's failure and its smart information at the time. Furthermore, as a result of repairing the storage device SD to copy data from it, the service provider also possesses the smart information of the storage device SD after the failure, which is normally unreadable when the storage device SD is installed in the user terminal 2. Such information as the model number of the failed storage device SD, the cause of the failure, smart information, etc. is stored and managed in a learning data DB 81 provided in the server 1 as learning data to be used in the learning process described later.

[0017] In a learning process described later, the server 1 uses the learning data stored in the learning data DB 81 to generate or update a model for performing a failure prediction diagnosis of the storage device SD, and stores and manages the model in the learning data DB 81.

[0018] In the example of the first embodiment, the smart information of the storage device SD acquired by each user terminal 2 is transmitted to the server 1. The server 1 performs an inference process using the transmitted smart information and the model stored in the model DB 82, thereby performing a failure prediction diagnosis of the storage device SD.

[0019] Here, the smart information may include values ​​for more than 100 parameters (items), including the temperature of the storage device SD, vibration during writing, number of times the power is turned on / off, number of communication errors, rate of occurrence of write errors, average number of erases and maximum number of erases, depending on the manufacturer and model of the storage device SD. As will be explained in more detail later, the failure prediction diagnosis in this service obtains the parameter values ​​of approximately 300 items contained in the smart information of the storage device SD being diagnosed. The acquired values ​​of each item are then compared with the first index set for each item, and a diagnostic result (hereinafter referred to as "item diagnostic result") from the perspective of each item (micro perspective) is obtained according to a predetermined first judgment method that uses the results of the comparison. In other words, the first index and the predetermined first judgment method that uses the results of the comparison with the first index are one of the models used in this service. In other words, the first index and the first judgment method are updated daily through a learning process. Furthermore, a value indicating the overall condition of the storage device SD to be diagnosed (hereinafter referred to as the "overall condition value") is obtained from the values ​​of each of these acquired items, and this overall condition value is compared with the second index. A diagnosis result (hereinafter referred to as the "overall diagnosis result") from the perspective of the overall condition (macro perspective) of the storage device SD to be diagnosed is obtained according to a predetermined second judgment method that uses the results of the comparison. In other words, the second index and the predetermined second judgment method that uses the results of the comparison with the second index are one of the models used in this service. In other words, the second index and the second judgment method are updated daily through a learning process. Then, based on the micro diagnosis result and the macro diagnosis result, the failure prediction diagnosis result of the storage device SD is determined as the final result. The failure prediction diagnosis results determined in this way make it possible to diagnose failures in the storage device SD in advance with a high degree of accuracy.

[0020] At this time, the smart information and the like of the storage device SD transmitted to the server 1 is stored in the learning data DB 81 and managed. The server 1 executes the learning process as appropriate to update the model (the first index and the first judgment method, and at least a part of the second index and the second judgment method) for performing failure diagnosis and prediction of the storage device SD stored in the model DB 82. As a result, the accuracy of failure prediction and diagnosis improves day by day.

[0021] FIG. 2 is a block diagram showing an example of a hardware configuration of a server in the information processing system of FIG.

[0022] The server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0023] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13 . The RAM 13 also stores information necessary for the CPU 11 to execute various processes as needed.

[0024] The CPU 11, ROM 12, and RAM 13 are connected to one another via a bus 14. An input / output interface 15 is also connected to this bus 14. An input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20 are connected to the input / output interface 15.

[0025] The input unit 16 is made up of various hardware such as a keyboard and a mouse, and is used to input various types of information. The output unit 17 is composed of various hardware components such as a liquid crystal display and a speaker, and outputs various types of information. The storage unit 18 is configured with a hard disk, a DRAM (Dynamic Random Access Memory), etc., and stores various types of information. The communication unit 19 controls communication with other devices (for example, the user terminal 2 in the example of FIG. 1) via a network NW including the Internet.

[0026] The drive 20 is provided as needed. Removable media 31, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 20. The program read from the removable media 31 by the drive 20 is installed in the storage unit 18 as needed. The removable media 31 can also store various pieces of information stored in the storage unit 18 in the same way as the storage unit 18.

[0027] FIG. 3 is a block diagram showing an example of a hardware configuration of a user terminal in the information processing system of FIG.

[0028] The user terminal 2 includes a CPU 41 , a ROM 42 , a RAM 43 , a bus 44 , an input / output interface 45 , an input unit 46 , an output unit 47 , a storage unit 48 , a communication unit 49 , and a drive 50 .

[0029] The CPU 41 executes various processes according to a program recorded in the ROM 42 or a program loaded from the storage unit 48 into the RAM 43 . The RAM 43 also stores information necessary for the CPU 41 to execute various processes as needed.

[0030] The CPU 41, ROM 42, and RAM 43 are connected to one another via a bus 44. An input / output interface 45 is also connected to this bus 44. An input unit 46, an output unit 47, a storage unit 48, a communication unit 49, and a drive 50 are connected to the input / output interface 45.

[0031] The input unit 46 is made up of various hardware such as a keyboard and a mouse, and is used to input various types of information. The output unit 47 is configured with various hardware components such as a display unit D including a liquid crystal display and a speaker, and outputs various information. The storage unit 48 is composed of a storage device SD, DRAM, etc., and stores various information. It is assumed here that the storage device SD is equipped with the above-mentioned SMART function.

[0032] The communication unit 49 controls communication with other devices (for example, the server 1 in the example of FIG. 1) via a network NW including the Internet.

[0033] The drive 50 is provided as needed. Removable media 61, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is appropriately attached to the drive 50. The program read from the removable media 61 by the drive 50 is installed in the storage unit 48 as needed. The removable media 61 can also store various pieces of information stored in the storage unit 48 in the same way as the storage unit 48.

[0034] The server 1 and the user terminal 2 configured in this manner are capable of executing various processes through cooperation between the various hardware and software components.

[0035] For example, the server 1 and the user terminal 2 have a functional configuration as shown in FIG. 4 when executing various processes including the failure prediction diagnosis process and the learning process. The failure prediction diagnosis process refers to a series of processes in which the server 1 acquires smart information from the user terminal 2, performs a failure prediction diagnosis based on the smart information, and presents the failure prediction diagnosis results to the user terminal 2. The learning process refers to a process in which the server 1 executes machine learning to generate or update a model (first index and first judgment method, and second index and second judgment method) used in the failure prediction process.

[0036] When the failure prediction diagnosis process is executed in the server 1, the CPU 11 functions as a learning data management unit 72, a failure prediction diagnosis unit 73, and a failure prediction diagnosis result presentation control unit 74.

[0037] When a failure prediction diagnosis process is executed, the learning data management unit 72 obtains a model (first index and first judgment method, and second index and second judgment method) from the learning data DB 81 and provides it to the failure prediction diagnosis unit 73.

[0038] In the failure prediction diagnosis unit 73, when the failure prediction diagnosis process is executed, a smart information acquisition unit 731, a comprehensive state calculation unit 732, an item diagnosis unit 733, a comprehensive diagnosis unit 734, and a final diagnosis result determination unit 735 function.

[0039] Here, a smart monitoring unit 91 is functioning in the CPU 41 of the user terminal 2 that is the target of the failure prediction diagnosis process. The smart monitoring unit 91 acquires smart information from the storage device SD installed in the user terminal 2 and transmits it to the server 1. Then, the smart information acquisition unit 731 of the server 1 acquires the values ​​of N items (N is an integer value of 2 or more) from among the multiple items of the smart information transmitted from the user terminal 2.

[0040] The overall status calculation unit 732 calculates an overall status value indicating the overall status of the storage device SD of the user terminal 2 from the values ​​of the N items acquired by the smart information acquisition unit 731 according to a predetermined calculation method that uses the N items of smart information as parameters.

[0041] The item diagnosis unit 733 compares the values ​​of at least some of the N items acquired by the smart information acquisition unit 731 with a first index set for each item, and determines the item diagnosis results related to failure prediction according to a predetermined first judgment method using the results of the comparison.

[0042] The final diagnostic result determination unit 735 determines a failure prediction diagnostic result as the final diagnostic result for the storage device SD of the user terminal 2 based on the item diagnostic result determined by the item diagnostic unit 733 and the overall diagnostic result determined by the overall diagnostic unit 734.

[0043] The failure prediction diagnosis result presentation control unit 74 executes control to present the failure prediction diagnosis result, which is the final diagnosis result of the failure prediction diagnosis unit 73, to the user via the user terminal 2. That is, the failure prediction diagnosis result presentation control unit 74 transmits the failure prediction diagnosis result, which is the final diagnosis result of the failure prediction diagnosis unit 73, to the user terminal 2 via the communication unit 19. Then, the failure prediction diagnosis result presentation unit 92 functions in the CPU 41 of the user terminal 2. That is, the failure prediction diagnosis result presentation unit 92 receives the failure prediction diagnosis result via the communication unit 49 and displays it on the display unit D, thereby presenting it to the user.

[0044] Here, as a specific example of the failure prediction diagnosis result, as will be described later with reference to Figures 7 and 8, either a first type that recommends backing up data from the storage device SD of the user terminal 2, or a second type level that does not require backing up data from the storage device SD, is determined. Furthermore, the first level is divided into M levels (M is an integer value of 2 or more, and M=2 in the examples of FIGS. 7 and 8). The second level is divided into L levels (L is an integer value of 2 or more independent of M, and L=3 in the examples of FIGS. 7 and 8). In this case, the failure prediction diagnosis unit 73 (final diagnosis result determination unit 735) determines a predetermined level from among (M+L) levels (levels 1 to 5 in the examples of Figures 7 and 8) as the failure prediction diagnosis result. The failure prediction diagnosis result presentation control unit 74 executes control to present the predetermined level to the user via the user terminal 2 as the failure prediction diagnosis result, which is the final diagnosis result of the failure prediction diagnosis unit 73 . Furthermore, when the predetermined level is the first level, the failure prediction diagnosis result presentation control unit 74 executes control to output an alert from the user terminal 2 equipped with the storage device SD. Specific examples of the alert will be described later with reference to FIGS. 8 and 11.

[0045] When the learning process is executed in the server 1, the CPU 11 functions as a learning unit 71 and a learning data management unit 72. The learning section 71 includes an item diagnostic learning section 711 and a comprehensive diagnostic learning section 712 .

[0046] As described above, smart information and the like obtained from a plurality of storage devices whose faulty or normal states are known is stored as learning data in the learning data DB 81. Here, the learning data also includes failure prediction diagnosis results previously output from the failure prediction diagnosis unit 73 (failure prediction diagnosis results for various user terminals 2). Therefore, the learning data management unit 72 extracts learning data from the learning data DB 81 and provides it to the learning unit 71 .

[0047] The item diagnosis learning unit 711 executes predetermined machine learning using the learning data, and uses the results to generate or update at least a part of the first index and the first judgment hand (a part of the model). The comprehensive diagnostic learning unit 712 executes predetermined machine learning using the learning data, and uses the results to generate or update at least a part of the second index and the second judgment move (a part of the model).

[0048] The learning data management unit 72 stores the models (the first index and the first judgment method, and the second index and the second judgment method) generated or updated by the learning unit 71 in the learning data DB 81.

[0049] The outline and configuration of the information processing system including the server 1 of the first embodiment have been described above using Figures 1 to 4. Below, Figures 5 and 6 are diagrams showing examples of the first index and first determination method, and the second index and second determination method in the information processing system of the above-mentioned first embodiment. FIG. 5 is a diagram showing an example of conditions for failure prediction executed by a server having the functional configuration of FIG.

[0050] The table in Figure 5 shows an example of a conditional expression for evaluating each of the five levels. The conditional expressions in the table of Figure 5 use thresholds generated or updated by the learning process of server 1 (e.g., learning process using a machine learning algorithm) to determine the conditions for smart information that correspond to each level. Then, for the storage device SD to be judged, the conditional expressions are applied to the smart values ​​acquired from that storage device SD in order starting from level 5. If the result of applying the conditional expression is a match, the level of that conditional expression is output as the result of the failure prediction diagnosis.

[0051] Specifically, for example, the conditional expression is applied when the SMART information of the storage device SD shows that the worst value of the number of uncorrectable sectors (ID198) is 200, the worst value of the temperature (ID194) is 70, the worst value of the read error rate (ID1) is 50, the number of reallocated sectors (ID5) is 30, the number of times the spindle motor has rotated / stopped (ID4) is 38, and the number of sectors waiting for replacement processing (ID197) is 0. Furthermore, as will be described in detail later with reference to Fig. 7, the overall evaluation value of the storage device SD (for example, the area S in Fig. 7) is also included in the conditional expression for each level. It is assumed that the overall evaluation value of the storage device SD is 70 (for example, the area Sa in Fig. 7 is 70). Such a condition for the overall evaluation value is also included in the conditional expression.

[0052] As a result, the condition in the second line (confirmation) of conditional expression 2 is met, namely, the worst value of the number of uncorrectable sectors (ID198) is less than 227, the worst value of the temperature (ID194) is 65 or more, the worst value of the read error rate (ID1) is 57 or less, the number of reallocated sectors (ID5) is 23 or more, the number of times the spindle motor has rotated / stopped (ID4) is 36 or more, and the number of sectors awaiting replacement processing (ID197) is less than 1. In addition, the condition that the overall evaluation value (area SA in FIG. 7) is 50 or more in the second line of the conditional expression is met. As a result, the overall evaluation of the storage device SD is determined to be at the second level.

[0053] 5 corresponds to the overall evaluation value. The numerical value to be compared with Sa (for example, 60 in the first line of the fifth level) is an example of a second index. The symbol between Sa and the numerical value to be compared with Sa (for example, "=>" (the left side is greater than or equal to the right side) in the first line of the fifth level) is an example of a second judgment method.

[0054] Furthermore, sa in FIG. 5 corresponds to the numerical value of a certain item included in the smart information. The numerical value to be compared with sa (for example, 50 in the first line of the fifth level) is an example of a first index. Furthermore, the symbol between sa and the numerical value to be compared (for example, ">" (the left side is greater than the right side) in the first line of the fifth level) is an example of a first determination method. As shown in FIG. 5, a first index and a first determination method are defined for each of a plurality of items. In the model DB 82 of the first embodiment, a set of determination formulas for determining whether or not a subject falls into each of these levels is stored as a model.

[0055] Here, the smart information that can be acquired from the storage device SD differs depending on the manufacturer, model number, manufacturing date, etc. Therefore, different conditional expressions are used as appropriate depending on the manufacturer, model number, manufacturing date, etc. of the storage device SD. The example in FIG. 5 is a conditional expression relating to a storage device SD with a certain manufacturer, model number, and manufacturing date.

[0056] In the explanation of Figure 5, the overall evaluation is based on determining the level of the storage device SD when the smart information of the target storage device SD matches the judgment formula for each level. However, each formula in the example of Figure 5 is generated or updated based on the following concept.

[0057] FIG. 6 is a diagram for explaining an outline of a method for calculating the overall state value by the server in FIG. For convenience of explanation, FIG. 6(A) is limited to the number of items of smart information N=6, and explains an outline of a method for calculating a comprehensive state value from the values ​​of six items of smart information. In the example of Figure 6(A), six items of smart information are assigned at equal intervals around the circumference of a specified circle, and axes from the center of the specified circle to the circumference are assigned to each of the six items, and the coordinates of each axis (numbers from 0 to 5) are a radar chart showing the value of the assigned item. That is, in the radar chart of the example of FIG. 6(A), six items are assigned clockwise from the top: temperature, read / write amount, number of bad sectors, usage time, number of startups, and startup time. 4 plots the values ​​of the six items acquired from the user terminal 2 at the corresponding coordinates on the assigned axis. In the radar chart of the example in FIG. 6(A), clockwise from the top, "3" is plotted as the temperature value, "4" is plotted as the read / write amount value, "5" is plotted as the number of bad sectors value, "2" is plotted as the usage time value, "1" is plotted as the number of startups value, and "5" is plotted as the time required for startup. Thus, in the example of Figure 6, the value of item N is not the raw data of each item (for example, temperature is 65 degrees), but is normalized to be one of the values ​​between 0 and 5 based on the first index and first judgment method used in item diagnosis. In this case, the overall state calculation unit 732 in FIG. 4 calculates the area S of the closed curve formed by connecting the points of the six plotted items as the overall state value, as shown in FIG. 6(A). That is, in the example of Figure 6(A), for example, the area SA is used as the second index, and the second judgment method is a method of comparing the area S of the closed curve formed by connecting each of the points of the six items plotted on the radar chart of Figure 6(A) with the area SA as the second index, and determining the overall judgment result based on the comparison result.

[0058] In the example of FIG. 6(A), for the sake of convenience of explanation, only N=6 items out of the many items of smart information are reflected in the radar chart. However, as mentioned above, smart information can contain over 1000 items, including differences in the manufacturer and model of the storage device SD. Therefore, in reality, as shown in Figure 6(B), it becomes a polygon with N = many vertices (the more N, the closer the polygon becomes to a circle).

[0059] FIG. 7 is a diagram showing an example of a failure prediction result generated by the server of FIG. 4 and presented to the user via the user terminal. In the example of Figure 7, as an example of a failure prediction result, information identifying the storage device SD (here, the string "Drive 01" indicating that it is the first drive) is shown, along with the string "The drive is operating normally," indicating that it is at the third level. That is, in the example of Figure 7, the failure prediction diagnosis unit 73 (final diagnosis result determination unit 735) of Figure 4 determines, as the failure prediction diagnosis result, either a first level that recommends backing up data from the storage device SD of the user terminal 2, or a second level that does not require backing up data from the storage device SD. The first level is divided into two stages: the fifth level (danger level) and the fourth level (caution level). The second level is the normal level, but is further divided into three levels, from level 1 to level 3. Here, level 1 is the same level as a new product. Level 2 is the same level as a product that has been used for 1 to 3 years. Level 3 is the same level as a product that has been used for more than 3 years. That is, the failure prediction diagnostic means, the failure prediction diagnostic unit 73 (final diagnostic result determination unit 735) in Figure 4, determines a predetermined level out of five levels (in the same figure, the third level out of the normal levels) as the failure prediction diagnostic result. As shown in Figure 7, among the failure prediction results for the user, the results indicating soundness are presented in three levels as levels of aging deterioration that do not correspond to a failure and show no signs of failure under normal conditions. This allows the user to understand the deterioration over time and the possibility that a failure prediction will one day be presented as a risk.

[0060] FIG. 8 is a diagram showing an example of a failure prediction result generated by the server of FIG. 4 and presented to a user via a user terminal, and is a diagram showing an example different from that of FIG.

[0061] FIG. 8 is a diagram showing an example of a screen on which a result indicating unsoundness is output from the failure prediction results for the user by the server having the functional configuration of FIG. In the screen shown in Figure 8(A), as an example of a failure prediction, information identifying the storage device SD (here, the string "Drive 01" indicating that it is the first drive) is displayed, along with the string "There is a possibility of failure, so please make a backup just in case," indicating that it is at the fourth level. That is, in the example of Figure 8 (A), the failure prediction diagnosis unit 73 (final diagnosis result determination unit 735) of Figure 4 diagnoses the failure prediction diagnosis result as being at level 4 (caution level) of the first type level which recommends backing up data from the storage device SD of the user terminal 2. That is, the failure prediction and diagnosis means, the failure prediction and diagnosis unit 73 (final diagnosis result determination unit 735) in FIG. 4, determines the fourth level as a predetermined level out of the five levels as the failure prediction and diagnosis result. As a result, as shown in the example of FIG. 8(A), a message is presented to the user indicating that the storage device SD has been determined to be at the fourth level.

[0062] FIG. 8 is a diagram showing an example of a screen on which a result indicating unsoundness is output from the failure prediction results for the user by the server having the functional configuration of FIG. In the screen shown in Figure 8(B), as an example of a failure prediction, information identifying the storage device SD (here, the string "Drive 01" indicating that it is the first drive) is displayed, along with the string "There is a possibility of failure at any moment, so please make a backup immediately," indicating that it is at the fifth level. That is, in the example of Figure 8 (B), the failure prediction diagnosis unit 73 (final diagnosis result determination unit 735) of Figure 4 diagnoses the failure prediction diagnosis result as being at the fifth level (risk level) of the first type level, which recommends backing up data from the storage device SD of the user terminal 2. That is, the failure prediction and diagnosis means, the failure prediction and diagnosis unit 73 (final diagnosis result determination unit 735) in FIG. 4, determines the fifth level as a predetermined level out of the five levels as the failure prediction and diagnosis result. As a result, as shown in the example of FIG. 8(B), a message is presented to the user indicating that the storage device SD has been determined to be at the fifth level.

[0063] As shown in Figures 8(A) and 8(B), among the failure prediction results for the user, results that indicate an unhealthy state are presented in two levels as backup recommendations that correspond to future failures (or that already correspond to failures). As a result, the user is recommended to make a backup based on the failure prediction, and can take appropriate action to make a backup.

[0064] FIG. 9 is a diagram illustrating an example of a proactive response of this service, which is realized by a server having the functional configuration of FIG. In the conventional approach, a failure occurs after normal operation, as shown in Figure 9. In other words, users usually recognize a failure when the device stops operating normally. Specifically, when a user attempts to read data stored on the storage device SD provided in the user terminal 2, the operating system of the user terminal 2 will attempt to read the data multiple times, and if the reading still fails, a warning or the like will usually be displayed. In other words, the user will notice the malfunction at the point when the data reading has already failed.

[0065] In such a situation, it is difficult for the user to fix the problem by himself. That is, disassembly and repair of SD storage devices, especially hard disk drives, can only be performed in a clean room. Also, repairs must be performed using healthy parts from the same type of SD storage device. For example, if data stored in a storage device SD cannot be read because it is logically (bit-wise) corrupted, it is necessary to read the data as is, check the bit sequence of the corrupted data, and repair the corrupted bit sequence. This is something that users cannot deal with.

[0066] Therefore, users who are unable to repair their own devices will request data recovery from a data recovery company when a malfunction occurs. As a result, the data recovery company will carry out data recovery. This type of work may take more than one week if the user terminal 2 is a home personal computer, or more than three weeks if the user terminal 2 is a business server or the like and the data is distributed and stored across multiple storage devices SD (for example, when a RAID system is used).

[0067] As mentioned above, the service provider of this service has been offering data recovery services for recovering data from failed storage devices (SD) at the request of users since before the launch of this service. The service provider of this service is currently the largest in the industry and has achieved one of the highest recovery rates. However, even if a data recovery request is made immediately after a failure occurs, the recovery rate is only 95.2%, not 100%. Furthermore, the reality is that data recovery requires a considerable amount of money. This explains why recovering data from a failed storage device (SD) is so difficult.

[0068] In contrast, when failure prediction is performed in the first embodiment, failure prediction is performed during the period of normal operation before the occurrence of a failure as in the conventional system. Therefore, data copying can be performed during the period of normal operation. As a result, data recovery work is not required, and the data migration rate (data backup completion rate) is close to 100%. This means that data problems such as being unable to read data can be prevented. In other words, the failure prediction provided by this service can reduce data problems to zero. The reason we say it's close to 100% is because, strictly speaking, the data migration rate is never 100%. In other words, there is a possibility that a very small number of bad sectors or minor logical errors have occurred in areas of the storage device SD that are not being accessed. Such bad sectors in areas that are not being accessed are not counted in the SMART information. For example, even if the condition for a normal state is that there are zero bad sectors, there is a non-zero chance that there may be undetected bad sectors and data migration will fail. For example, there is also the possibility that a completely new failure may occur during data migration. For this reason, a 100% data migration rate cannot be achieved.

[0069] FIG. 10 is a diagram showing the characteristics of a failure prediction technique using a server having the functional configuration of FIG. The failure prediction method shown in FIG. 10(A) makes a judgment by individually using each item included in the smart information, without using the above-mentioned overall evaluation value. 10(A), the failure level is predicted to be low (second level) if the worst case number of unrecoverable sectors is less than 10 and the worst case temperature is equal to or greater than 30 and less than 60. Therefore, for example, if the worst case number of unrecoverable sectors is 5 and the worst case temperature is 50, these conditions are met and the failure level (degree of failure) is determined to be low (second level). Here, in the example shown in Fig. 10(A), only the second level and the fourth level exist. This indicates that in the example shown in Fig. 10(A), as a result of failure prediction, no signs of failure are observed, and only the second level exists as a level at which backup of a backup is unnecessary, and the fourth level 4 exists as a level at which backup is recommended. In this way, the user is notified of the level determined based solely on the presence or absence of signs of a malfunction.

[0070] In contrast to this, the failure prediction method shown in FIG. 10(B) makes a judgment by individually using the above-mentioned comprehensive evaluation value and each item included in the smart information. That is, in the example shown in FIG. 10(B), if the overall evaluation value satisfies the predetermined condition, and furthermore, the worst value of the number of unrecoverable sectors is less than 10, and the worst value of the temperature is between 0 and less than 30, the failure degree is predicted as a first level (low). Also, if the overall evaluation value satisfies the predetermined condition, and furthermore, the worst value of the number of unrecoverable sectors is less than 10, and the worst value of the temperature is between 30 and 45, the failure degree is predicted as a second level (low). Also, if the overall evaluation value satisfies the predetermined condition, and furthermore, the worst value of the number of unrecoverable sectors is less than 10, and the worst value of the temperature is between 45 and 60, the failure degree is predicted as a third level (low). Also, if the overall evaluation value satisfies the predetermined condition, and furthermore, the worst value of the number of unrecoverable sectors is less than 10, and the worst value of the temperature is between 60 and 60, the failure degree is predicted as a fourth level (medium). In this way, the failure prediction method shown in Figure 10(B) makes a judgment using the above-mentioned overall evaluation value and each item included in the smart information individually, so that what would have simply been judged as a low failure level (second level) in the failure prediction method shown in Figure 10(A) can be judged as a first level to a third level. As a result, notifications are given to the user in stages, from level 1 to level 3. Therefore, when level 3 is notified, the user is notified that level 4 is approaching, that is, that the failure prediction result indicates that the storage device SD is likely to fail and that a backup is recommended. This allows the user to prepare in advance for backups, such as by preparing software and hardware for backing up the storage device SD.

[0071] FIG. 11 is a diagram showing an example of a screen including the results of a failure prediction diagnosis performed on a user terminal by a server having the functional configuration of FIG. As shown in FIG. 11, the failure diagnosis prediction result is output by being displayed on a display D, which is an example of the output unit 147 included in the user terminal 2, for example. The failure prediction diagnosis results will display messages such as "SMART information warning," "The storage device you are using is at a high risk of failure. Please back it up immediately," and "You can also request support from the mass retailer XX Company YY store where you purchased the device." In this way, the user terminal 2 outputs to the user the results of the failure prediction diagnosis by the server 1. This allows the user to complete the backup before the storage device SD fails, which is why a 100% data migration rate is achieved as shown in Figure 9.

[0072] As mentioned above, instead of backing up the data themselves, users can request support from the retailer where they purchased the data. This service can also manage and provide information for support at such retailers. This point will be described in detail in the second embodiment below. [Second embodiment]

[0073] In the first embodiment, the failure prediction diagnosis process was executed in the server 1. That is, in the first embodiment, communication was required to transmit the smart information acquired from the storage device SD to be diagnosed from the user terminal 2 to the server 1, and communication was required to transmit the failure diagnosis prediction result from the server 1 to the user terminal 2. In contrast to this, in the second embodiment, the failure prediction and diagnosis processing is executed in the user terminal 2. That is, in the second embodiment, communication between the user terminal 2 and the server 1 accompanying the failure prediction and diagnosis processing is not required.

[0074] FIG. 12 is a diagram showing an example of an outline of the present service of the second embodiment using a user terminal according to an embodiment of the information processing device of the present invention.

[0075] As shown in FIG. 12, a USB memory storing a program (hereinafter referred to as a "failure prediction program") for causing a computer (here, the CPU 41 of the user terminal 2) to execute the failure prediction diagnostic process is provided by provider T to seller B. Here, the provider T is a provider of the present service (the present service of the second embodiment) that causes the user terminal 2 to execute the failure prediction and diagnosis process. The USB memory is provided by provider T to seller B via a third party as necessary.

[0076] Seller B is, for example, a mass retailer that sells a variety of information processing devices BP, and sells the information processing device BP equipped with a predetermined storage device SD together with a USB memory storing a failure prediction program as a set.

[0077] A user U purchases from a seller B an information processing device BP equipped with a predetermined storage device SD, together with a USB memory storing a failure prediction program. The user U inserts the USB into the information processing device BP and installs the failure prediction program into the information processing device BP, which then becomes the user terminal 2 of the second embodiment. That is, the hardware configuration of the user terminal 2 of the second embodiment is the same as that of Fig. 3. Furthermore, although not shown, the functional configuration of the user terminal 2 of the second embodiment is such that in addition to the smart monitoring unit 91 and the failure prediction diagnosis result presentation unit 92 of Fig. 4, the CPU 41 further functions as a failure prediction diagnosis unit 73 and a failure prediction diagnosis result presentation control unit 74, and models (first index and first judgment method, and second index and second judgment method) are stored in the storage unit 48. As a result, the user terminal 2 will be able to constantly execute the failure prediction diagnosis process.

[0078] Here, if the failure prediction diagnosis result indicates that support is required (for example, the first level described in Figures 6 and 7 of the first embodiment, i.e., the fourth or fifth level), a predetermined alert (see, for example, Figure 11) is displayed on the display unit D of the user terminal 2. In the second embodiment, a user terminal 2 equipped with a storage device SD when such an alert is output can function as a medium to indicate to seller B that user U has the right to receive support from seller B.

[0079] In other words, when user U shows seller B the user terminal 2 equipped with the storage device SD with the alert output, seller B can provide various types of support to user U, assuming that user U has the right to receive support. Here, "user U shows user terminal 2 equipped with storage device SD with an alert output" means that the user terminal 2 is directly brought to seller B, or a photograph of the user terminal 2 with the alert displayed is brought, or seller B's terminal is notified by communication in a predetermined manner, or any other method is used to show that the user terminal 2 is supported by seller B. In addition, support is a broad concept that includes not only inspecting and repairing the user terminal 2 itself, but also things that can be decided freely by the seller B, such as selling a new information processing device BP to the user U in exchange for the user terminal 2. In this way, by providing the service of the second embodiment, the user U will go to the seller B to receive support, which has the effect of encouraging the seller B to have the user U visit their store.

[0080] Furthermore, the failure prediction program can cause the user terminal 2 to exhibit a function of executing a control process that outputs an alert even when an error exists in the program itself. In the second embodiment, a user terminal 2 equipped with a storage device SD, in a state in which an alert indicating that there is an error in such a failure prediction program itself, can function as a medium indicating to provider T that user U has the right to receive support from provider T.

[0081] In other words, when user U shows provider T a user terminal 2 equipped with a storage device SD that has an alert output indicating that there is an error in the failure prediction program itself, provider T can provide various types of support to user U, assuming that user U has the right to receive support. Here, "user U shows user terminal 2 equipped with storage device SD with an alert output" means that the actual user terminal 2 is brought directly to provider T, or the actual user terminal 2 is brought to seller B after a contract has been signed between provider T and seller B, or a photo of the user terminal 2 with an alert displayed is brought to provider T or seller B, or the terminal of provider T or seller B is notified by communication in a specified manner, or the user terminal 2 is shown in any way so that it is clear that it is supported by provider T. Furthermore, the support referred to here is for alerts regarding the failure prediction program, and therefore includes repair, replacement, or resale of the failure prediction program. Depending on the nature of the malfunction in the failure prediction program, it may be possible to deal with only the failure prediction program that generated the alert, or it may affect all of the other failure prediction programs. If the problem affects all of the other failure prediction programs, support may include seller B collecting inventory from stores, or provider T reporting the malfunction on seller B's website.

[0082] FIG. 13 is a diagram illustrating the overall flow of the present service according to the second embodiment. FIG. 14 is a diagram illustrating a basic commercial flow of the present service according to the second embodiment. 13, the USB is manufactured by a USB manufacturer, and third parties P1 and P2 are involved in the process from provider T to seller B. However, these are merely examples, and for example, provider T can perform the processes (tasks) of third parties P1 and P2 to exclude third parties P1 and P2, or the processes (tasks) of third parties P1 and P2 can be shared among one or three or more third parties.

[0083] In step S1 of FIGS. 13 and 14, a USB manufacturer manufactures a USB memory and sends it to a provider T.

[0084] In step S2 in FIGS. 13 and 14, provider T issues a license key. In step S21 of FIG. 14, the USB memory and the license key are shared and sent to a third party P1.

[0085] 13 and 14, the third party P1 performs kitting on the USB memory. For example, the third party P1 transfers an installer for a failure prediction program to the USB memory. In step S31 of FIG. 14, the third party P1 sends the USB memory (license key) to the third party P2.

[0086] 13 and 14, the third party P2 performs packaging on the USB memory (license key). For example, the third party P2 packages the USB memory (license key) into a product state. Here, packaging refers to work such as printing and packing the package. In step S41 of FIG. 14, the packaged USB memory (license key) is sent to seller B as a product.

[0087] In step S5 of FIGS. 13 and 14, seller B sells a USB memory (license key) as a product to user U. In step S6 of FIG. 13, the user U purchases a USB memory (license key) as a product from the seller B. In step S7 of FIG. 13, seller B and user U make a payment at the register for the product, which is a USB memory (license key). In step S8 of Fig. 13, seller B performs POS linkage. That is, in step S9 of Fig. 13, seller B manages sales information and stock confirmation of USB memory (license key), and in step S19 of Fig. 13, provider T manages sales information and stock confirmation of USB memory (license key). In step S10 of Fig. 13, seller B places an additional order for the manufacture of a USB memory to a USB manufacturer in response to the need for an additional order. For the party that placed this additional order, the processes of steps S1 to S41 of Fig. 13 and Fig. 14 described above are repeatedly executed, and the USB memory (license key) as the additionally ordered product is sent to seller B.

[0088] On the other hand, on the user U side, the processes from step S11 onward in FIG. 13 are executed. That is, in step S11 in FIGS. 13 and 14, the user U inserts the USB memory into the user terminal 2 and installs the failure prediction program. In step S12 of FIG. 13, the user U inputs the license key into the user terminal 2. The user terminal 2 checks the consistency in step S13 of FIG. 13, and performs an activation process on the failure prediction program in step S14 of FIG. As a result, in the CPU 41 of the user terminal 2, in addition to the smart monitoring unit 91 and the failure prediction diagnosis result presentation unit 92 of Figure 4, the failure prediction diagnosis unit 73 and the failure prediction diagnosis result presentation control unit 74 also function, and the model (first index and first judgment method, and second index and second judgment method) is stored in the memory unit 48. As a result, the user terminal 2 will be able to constantly execute the failure prediction diagnosis process. Therefore, in step S15 of FIG. 13, the user U starts to use the user terminal 2 equipped with the storage device SD.

[0089] 14, the user terminal 2 runs a failure prediction program in a resident state. That is, the user terminal 2 executes a failure prediction diagnosis process as appropriate. In step S23 of FIG. 14, the user terminal 2 determines whether or not a malfunction has occurred. If no malfunction has occurred, that is, if the failure prediction diagnosis result is any of the first to third levels and the failure prediction program itself is normal, the result is determined to be NO in step S23 of FIG. 14, and the process returns to step S22 of FIG. 14, and the subsequent processes are repeatedly executed. On the other hand, if a malfunction occurs, i.e., if the malfunction prediction diagnosis result is either the fourth or fifth level, or if a malfunction occurs in the malfunction prediction program itself, a predetermined alert is displayed on the user terminal 2, and processing proceeds to step S16 in FIG. 14.

[0090] Here, for example, it is assumed that the failure prediction diagnosis result is level 4 or level 5, so that the determination in step S23 of Fig. 13 is YES (a malfunction) and an alert is issued. In this case, as described above, the user terminal 2 equipped with the storage device SD in a state in which an alert regarding the predicted failure of the storage device SD has been output functions as a medium indicating to the seller B that the user U has the right to receive support from the seller B. That is, when seller B is shown a user terminal 2 equipped with a storage device SD in which an alert regarding a predicted failure of the storage device SD has been output, seller B will provide support (respond to the alert) as step S16 in Figures 13 and 14. In this case, in step S17 of FIG. 13, the third party P1 can cooperate as the seller B to respond to the alert (support response). In step S24 of Fig. 14, it is determined whether the problem has been resolved. If the problem has not been resolved, the determination in step S24 of Fig. 14 is NO, and the process returns to step S16 of Fig. 14, where further support measures are taken. On the other hand, if the malfunction has been resolved, a determination of YES is made in step S24 of Fig. 14, the process returns to step S22, and the subsequent processes are repeated. Here, if the method of resolving the malfunction is to repair the user terminal 2 itself in which the malfunction has occurred and then use it again by the user, then in step S22 of Fig. 14, the user terminal 2 itself runs a resident malfunction prediction program. On the other hand, if the method of reselling a new user terminal 2 is used as the method of resolving the malfunction, then in step S22 of Fig. 14, the resold user terminal 2 runs a resident malfunction prediction program.

[0091] Also, for example, assume that the failure prediction program is defective, so that the determination in step S23 of Fig. 13 is YES (defective), and an alert is issued. In this case, as described above, the user terminal 2 equipped with the storage device SD in a state in which an alert indicating that the failure prediction program is defective is outputted functions as a medium indicating to the provider T that the user U has the right to receive support from the provider T. That is, when provider T is shown a user terminal 2 equipped with a storage device SD in which an alert indicating that the failure prediction program is malfunctioning has been output, provider T will provide support (respond to the alert) at step S18 in FIG. 13 (corresponding to step S16 in the example of FIG. 14). Then, in step S24 of Fig. 14, it is determined whether the problem has been resolved. If the problem has not been resolved, the determination in step S24 of Fig. 14 is NO, and the process returns to step S16 of Fig. 14, where further support measures are taken. On the other hand, if the problem has been resolved, the determination in step S24 of FIG. 14 is YES, the process returns to step S22, and the subsequent processes are repeatedly executed.

[0092] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. are included in the present invention within the scope that can achieve the object of the present invention. In the above embodiment, failure prediction is performed using SMART information as information related to the storage device SD, but the present invention is not limited to this. Specifically, for example, as described above, configuration information of a file system or RAID system constructed using the storage device SD may also be used for learning and failure prediction diagnosis. Also, for example, firmware information regarding bad sectors may also be used for learning and failure prediction diagnosis. That is, information indicating that the handling conditions for bad sectors differ depending on the firmware of the storage device SD and its version may be used for learning and failure prediction diagnosis.

[0093] Furthermore, apart from the above-mentioned smart information, terminal information and data exchange information may be used as information relating to the user terminal 2. Terminal information refers to information other than smart information that is related to the state of the user terminal 2. Specifically, for example, terminal information may include information such as the usage rate of the CPU 41 of the user terminal 2, the usage rate of the RAM 43 of the user terminal 2, and the number of input / output operations per second of the communication unit 49 of the user terminal 2. Here, these are merely examples, and terminal information may include all kinds of information related to the state of the user terminal 2 other than smart information. The data transfer information refers to information relating to the transfer of data between the storage device SD provided in the user terminal 2 and other hardware devices (such as the CPU 41) provided in the user terminal 2. Specifically, for example, the data transfer information may include information such as the number of input / output operations per second, throughput, and latency information relating to the input and output of data to and from the storage device SD. These are merely examples, and the data transfer information may include any and all information relating to the transfer of data between the user terminal 2 and the storage device SD provided in the user terminal 2. The server 1 can also calculate a highly accurate failure probability using the smart information, terminal information, and data transfer information. In this way, the server 1 can perform failure prediction diagnosis using "smart information, etc." that may include not only smart information but also terminal information and data exchange information.

[0094] In the second embodiment described above, the support provided by seller B and provider T can be freely decided by seller B, such as inspecting and repairing the user terminal 2 itself, or selling a new information processing device BP to user U in exchange for the user terminal 2, but can also include the following: That is, for example, seller B can provide warranty support for user terminal 2 if the user terminal 2 is within the warranty period. Furthermore, for example, if the user terminal 2 is out of warranty and data can be copied from the storage device SD, seller B can introduce and sell another storage device (for example, an external hard disk drive) to the user.

[0095] Furthermore, for example, seller B can purchase the user terminal 2 as a second-hand information processing terminal. At this time, if the user terminal 2 is not in a state where a failure is predicted (for example, a state where the above-mentioned fourth or fifth level alert has been issued), the alert of this service can be used as information for appraisal, such as purchasing the user terminal 2 at a lower price than when the user terminal 2 is in a healthy state (backup is not recommended, and backup is not necessary (first to third levels)). Furthermore, at this time, the user U can expect to receive a higher purchase price by selling the user terminal 2 to seller B or the like as a used information processing terminal while the user terminal 2 is in a healthy state (backup is not recommended and backup is not necessary (levels 1 to 3)).

[0096] Furthermore, for example, provider T or seller B may grant the failure prediction program the right to support data recovery for the storage device SD. This allows provider T to recover the data if, for example, user U requests support in a fourth-level state and the storage device SD fails during data backup. This allows provider T to recover the data from the storage device SD when the probability of failure is lower, thereby further increasing the data recovery rate.

[0097] Also, for example, seller B can provide provider T with a storage device SD that was removed as a result of support. By receiving storage devices SD of levels 1 to 5, provider T can use them as parts to repair other storage devices SD, use them for research into rare cases of failure, and collect data on the relationship between the failure state and smart information, etc., and use it as learning data. This is expected to improve the accuracy of failure prediction diagnosis. As mentioned above, the premise is that the service provider of this service (Provider T) has been providing a data recovery service to recover data from failed storage devices SD in response to requests from users before providing this service. Therefore, the service provider (Provider T) has the technology to recover data from storage devices SD and the related technology to analyze firmware for storage devices SD. This allows for the efficient collection of learning data, etc.

[0098] In the above description of the second embodiment, a USB memory storing a failure prediction program is provided from provider T to seller B, but this is not particularly limited. For example, the failure prediction program may be provided on any removable media. Furthermore, the failure prediction program may be downloaded via the Internet.

[0099] Furthermore, for example, the above-described series of processes can be executed by hardware or software. In other words, the functional configuration of FIG. 4 is merely an example and is not particularly limited. That is, it is sufficient if the information processing system is provided with a function that can execute the above-described series of processes as a whole, and the functional blocks and databases used to realize these functions are not limited to the example of FIG. 4. Furthermore, the locations of the functional blocks are not particularly limited to those of FIG. 4 and may be arbitrary. For example, the functional block of server 1 may be transferred to user terminal 2, etc. Furthermore, the functional block of user terminal 2 may be transferred to server 1, etc. Furthermore, server 1 and user terminal 2 may be the same hardware.

[0100] Furthermore, for example, when a series of processes is executed by software, the programs that make up the software are installed into a computer or the like from a network or a recording medium. The computer may be a computer built on dedicated hardware. The computer may also be a computer that can execute various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0101] Furthermore, for example, the recording medium containing such a program may be configured not only as a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to the user, but also as a recording medium that is provided to the user in a state that is pre-installed in the device main body.

[0102] In this specification, the steps of describing a program to be recorded on a recording medium include not only processes that are performed chronologically in accordance with the order, but also processes that are not necessarily performed chronologically but are performed in parallel or individually. In addition, in this specification, the term "system" refers to an overall device that is made up of a plurality of devices, a plurality of means, etc.

[0103] In other words, the information processing device to which the present invention is applied can take various forms having the following configurations.

[0104] That is, the information processing device to which the present invention is applied (the server 1 in the first embodiment, and the user terminal 2 in the second embodiment) is An information processing device that performs a failure prediction diagnosis before a failure occurs on a storage medium (for example, the storage device SD of FIG. 1), A smart information acquisition means (for example, the smart information acquisition unit 731 in FIG. 4) for acquiring values ​​of N items (N is an integer value of 2 or more) from the target storage medium for failure prediction out of a plurality of items of smart information indicating the state of the storage medium (for example, items such as temperature, read / write amount, number of bad sectors, usage time, number of startups, and startup time in FIG. 6); a comprehensive state calculation means (for example, the comprehensive state calculation unit 732 in FIG. 4) that calculates a comprehensive state value (for example, the area S in FIG. 6) indicating the comprehensive state of the target storage medium from the values ​​of the N items acquired from the target storage medium according to a predetermined calculation method that uses the N items as parameters; an item diagnosis means (e.g., the item diagnosis unit 733 in FIG. 4) for comparing at least some of the values ​​of the N items acquired from the target storage medium with a first index set for each item (e.g., in the description of FIG. 5, an index in which the worst value of the number of unrecoverable sectors is 227), and determining an item diagnosis result related to failure prediction according to a predetermined first judgment method using the result of the comparison (e.g., in the description of FIG. 5, a judgment method in which the worst value of the number of unrecoverable sectors is "less than" the first index); a comprehensive diagnosis means (e.g., the comprehensive diagnosis unit 734 in FIG. 4) for comparing the comprehensive state value with a second index (e.g., the area S is 50) and determining a comprehensive diagnosis result related to failure prediction according to a predetermined second judgment method using the result of the comparison (e.g., a judgment index that the area S is "less than" the second index); a failure prediction diagnostic means (for example, the final diagnostic result determination unit 735 in FIG. 4) that determines a failure prediction diagnostic result for the target storage medium based on the item diagnostic result and the overall diagnostic result; and, It is enough to have this. This makes it possible to provide highly convenient failure prediction and related services for users of storage media.

[0105] The failure prediction diagnostic means can determine, as the result of the failure prediction diagnostic, either a first level that recommends backing up data from the target storage medium, or a second level that does not require backing up data from the target storage medium.

[0106] The device may further include an alert output control means that, when the failure prediction diagnosis result for the target storage medium is determined to be at the first level, controls the output of an alert from the device in which the target storage medium is installed.

[0107] The first level is divided into M levels (M is an integer value of 2 or more), The second level is divided into L levels (L is an integer value of 2 or more independent of M), The failure prediction and diagnosis means can determine a predetermined level out of (M+L) levels as the failure prediction and diagnosis result.

[0108] The N items are allocated at equal intervals on the circumference of a predetermined circle, an axis from the center of the predetermined circle to the circumference is allocated to each of the N items, and the coordinates of the axis indicate the values ​​of the allocated items; The overall state calculation means can calculate the area of ​​the closed curve formed by connecting each of the plotted points of the N items when each of the values ​​of the N items obtained from the target storage medium is plotted on the corresponding coordinate of the assigned axis as the overall state value.

[0109] At least one of the first index for a predetermined item and the first determination method for the predetermined item can be set in consideration of a correlation with other items.

[0110] A learning means may be provided that generates or updates at least some of the first index, the first judgment method, the second index, and the second judgment method based on the results of predetermined machine learning using smart information obtained from multiple storage media whose faulty or normal states are known.

[0111] Furthermore, the information processing device to which the present invention is applied can take various forms having the following configurations.

[0112] That is, a system to which the present invention is applied (for example, a support system for the service shown in FIG. 12) is as follows: an information processing device (for example, the user terminal 2 in FIG. 12) equipped with a predetermined storage medium that is sold to a user by a seller and used by the user; a program (e.g., the failure prediction and diagnosis program of FIG. 12) that, when installed in the information processing device having the predetermined storage medium mounted therein, causes the information processing device to perform a function of executing a control process for predicting a failure of the predetermined storage medium and outputting an alert when the failure prediction and diagnosis result indicates that support is required, and that is transferred from the seller to the user; Equipped with It is sufficient to have the information processing device equipped with the specified storage medium in a state in which the alert is output function as a medium that indicates to the seller that the user has the right to receive the support. This makes it possible to provide highly convenient failure prediction and related services for users of storage media.

[0113] The program is provided by the provider to the seller, the program causes the information processing device to perform a function of executing a control process to output a first alert when a failure prediction diagnosis result indicates that support is required, and to output a second alert when an error exists in the program itself; causing the information processing device in which the predetermined storage medium is installed, in a state in which the first alert is output, to function as a medium indicating to the seller that the user has the right to receive the support from the seller; The information processing device equipped with the specified storage medium when the second alert is output can function as a medium that indicates to the provider that the user has the right to receive support from the provider.

[0114] The program is stored in a removable storage medium (e.g., a USB memory in FIG. 12), and is provided to the distributor together with a license key for the program issued by the provider, and is then transferred from the distributor to the user; the user installs the removable storage medium into the information processing device in which the predetermined storage medium is installed, thereby installing the program into the information processing device; By inputting the license key into the information processing device, the information processing device can be caused to generate the function for executing the control process.

[0115] The program, as the control processing, a smart information acquisition step of acquiring, from the predetermined storage medium, values ​​of N items (N is an integer value of 2 or more) among a plurality of items of smart information indicating the state of the predetermined storage medium; a comprehensive status calculation step of calculating a comprehensive status value indicating a comprehensive status of the predetermined storage medium from the values ​​of the N items acquired from the predetermined storage medium according to a predetermined calculation method using the N items as parameters; an item diagnosis step of comparing values ​​of at least some of the N items acquired from the predetermined storage medium with a first index set for each item, and determining an item diagnosis result related to failure prediction according to a predetermined first judgment method using the result of the comparison; a comprehensive diagnosis step of comparing the comprehensive state value with a second index and determining a comprehensive diagnosis result related to failure prediction according to a predetermined second judgment method using the result of the comparison; a failure prediction diagnosis step of determining a failure prediction diagnosis result for the predetermined storage medium based on the item diagnosis result and the overall diagnosis result; an alert output control step of executing the failure prediction control for the predetermined storage medium; The program may be a program that causes the information processing device to perform a function of executing a control process including the above. [Explanation of symbols]

[0116] 1 Server, 2 User terminal, 11 CPU, 18 Storage unit, 20 Drive, 31 Removable media, 41 CPU, 42 ROM, 48 Storage unit, 50 Drive, 61 Removable media, 71 Learning unit, 72 Learning data management unit, 73 Failure prediction and diagnosis unit, 74 Failure prediction and diagnosis result provider display control unit, 81... learning data DB, 82... model DB, 91... smart monitoring unit, 92... fault prediction diagnosis result presentation unit, 147... output unit, 711... item diagnosis learning unit, 712... overall diagnosis learning unit, 731... smart information acquisition unit, 732... overall state calculation unit, 733... item diagnosis unit, 734... overall diagnosis unit, 735... final diagnosis result determination unit

Claims

1. A second information processing device connected via a network to a plurality of first information processing devices each managed by a plurality of users; each of the plurality of first information processing devices; In an information processing system including: Each of the plurality of first information processing devices Storage media is installed, a smart information transmission control means for executing control to acquire, from the storage medium installed in the first information processing device, values ​​of N items (N is an integer value of 2 or more) of the smart information, etc., from the storage medium installed in the first information processing device, and to transmit the values ​​to the second information processing device via the network, the smart information being a group of information that includes at least smart information indicating the status of the storage medium and may include information regarding the status of the first information processing device that uses the storage medium and information regarding the exchange of data between the storage medium and other hardware devices installed in the first information processing device; The second information processing device smart information reception control means for executing control to receive the smart information etc. transmitted from each of the plurality of first information processing devices; a comprehensive status calculation means for calculating a comprehensive status value indicating a comprehensive status of the storage medium installed in each of the plurality of first information processing devices from the values ​​of the N items received from each of the plurality of first information processing devices in accordance with a predetermined calculation method using the N items as parameters; an item diagnosis means for comparing at least some of the values ​​of the N items received from each of the plurality of first information processing devices with a first index set for each item, and determining an item diagnosis result related to failure prediction according to a predetermined first judgment method using the result of the comparison; a comprehensive diagnosis means for comparing the comprehensive state value with a second index and determining a comprehensive diagnosis result related to failure prediction according to a predetermined second judgment method using the result of the comparison; a failure prediction and diagnosis means for determining a failure prediction and diagnosis result for the storage medium installed in each of the plurality of first information processing devices based on the item diagnosis result and the comprehensive diagnosis result; a diagnostic result transmission control means for controlling transmission of the failure prediction diagnosis result to the corresponding first information processing device via the network; Equipped with Each of the plurality of first information processing devices includes a diagnostic result presentation means for receiving the failure prediction diagnostic result transmitted from the second information processing device and presenting the failure prediction diagnostic result to the user.

2. The second information processing device is provided with a management means for collectively managing the failure prediction diagnosis results for each of the plurality of first information processing devices. The information processing device according to claim 1 .

3. Each of the plurality of first information processing devices periodically transmits the smart information, etc. to the second information processing device by the smart information transmission control means. The information processing device according to claim 2 .

4. Each of the plurality of first information processing devices is provided with an alert output control means that executes control to output an alert when the failure prediction diagnosis result is at a predetermined level.

10. The information processing device according to claim 1.

5. N is an integer value of 3 or more, The N items are allocated at equal intervals on the circumference of a predetermined circle, and an axis from the center of the predetermined circle to the circumference is allocated to each of the N items, and the overall state calculation means calculates, as the overall state value, an area of ​​a closed curve formed by connecting each of the points of the N items plotted on the coordinates of the axis in a radar chart. The information processing device according to claim 1 .

Images