Method for monitoring active storage apparatus

Abstract

Disclosed in the present invention is a method for monitoring an active storage apparatus. The method for monitoring an active storage apparatus in the present invention comprises the following steps: using a cloud server to train a prediction model; a storage apparatus actively sending health state data of the storage apparatus to the cloud server; and using the cloud server and the prediction model to predict a future state of the storage apparatus on the basis of the health state data of the storage apparatus.

Description

Monitoring method for active storage device Technical Field

[0001] The present invention relates to a storage device, and in particular to a monitoring method for an active storage device. Background Art

[0002] With the expansion of information and the advancement of technology, storage devices of various specifications, such as solid-state drives, are widely used in various fields to store different data. However, when a storage device fails, the data stored in the storage device may be lost.

[0003] The existing storage device health monitoring method is that the storage device passively provides its health status data. When the storage device is detected to be abnormal in health, the storage device may be close to being unusable. This places a heavy burden on maintenance costs.

[0004] Summary of the Invention

[0005] The present invention addresses the shortcomings of the prior art by providing a method for monitoring an active storage device. The method comprises the following steps: training a prediction model using a cloud server; actively transmitting health status data of the storage device to the cloud server from a storage device; and using the prediction model, the cloud server, to predict the future state of the storage device based on the health status data of the storage device.

[0006] As described above, the present invention provides a monitoring method for an active storage device, in which the storage device actively sends its own health status data to a cloud server, enabling the cloud server to monitor changes in the health status data of the storage device in real time, thereby dynamically predicting the future status of the storage device in real time. When a future failure state is predicted, the cloud server can issue an alert and perform processing operations such as data backup of the storage device.

[0007] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only used to provide standards and explanations and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG1 is a block diagram of a monitoring system for an active storage device according to an embodiment of the present invention.

[0009] FIG. 2 is a flowchart of the steps of actively notifying update information by the storage device included in the monitoring method of the active storage device according to an embodiment of the present invention.

[0010] FIG. 3 is a flowchart of a method for monitoring an active storage device according to an embodiment of the present invention, including steps for a host to confirm whether a cloud server supports the storage device.

[0011] FIG. 4 is a flowchart of steps of performing storage device failure prediction by a cloud server included in a monitoring method for an active storage device according to an embodiment of the present invention.

[0012] FIG. 5 is a flowchart of steps of backing up data of a storage device by a cloud server included in a monitoring method for an active storage device according to an embodiment of the present invention.

[0013] FIG. 6 is a schematic diagram showing a curve of a count value of erase times of a storage device as it changes over time, established by a monitoring method for an active storage device according to an embodiment of the present invention.

[0014] FIG. 7 is a schematic diagram showing a curve of a write count value of a storage device changing over time, established by a monitoring method for an active storage device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed based on different viewpoints and applications without departing from the concept of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual dimensions. It is stated in advance. The following embodiments will further explain the relevant technical content of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention. In addition, the term "or" used herein may include any one or more combinations of the associated listed items depending on the actual situation.

[0016] Please refer to Figures 1 and 2, where Figure 1 is a block diagram of a monitoring system for an active storage device according to an embodiment of the present invention, and Figure 2 is a flow chart of the steps of actively notifying update information by the storage device included in a monitoring method for an active storage device according to an embodiment of the present invention.

[0017] The monitoring system for active storage devices of the present invention may include a host HS and a cloud server CLU as shown in FIG1 , and may be used to manage one or more storage devices, such as the plurality of storage devices SRD1 to SRDn shown in FIG1 .

[0018] It should be understood that each storage device includes at least a storage component and a controller. For example, storage device SRD1 shown in FIG1 includes a storage component ST1 and a controller CR1, storage device SRD2 includes a storage component ST2 and a controller CR2, and storage device SRDn includes a storage component STn and a controller CRn. In practice, storage component ST1 may include multiple memory cells, but the present invention is not limited thereto. Therefore, the internal architecture of storage component ST1 will not be described in detail herein.

[0019] For ease of explanation, the cloud server CLU of the active storage device monitoring system of the present invention may include a database DB, a training model building component PR, and a state prediction component FA. In practice, these components may be replaced with other components or include more components.

[0020] The database DB of the cloud server CLU stores a large amount of collected reference data. The training model building component PR of the cloud server CLU performs big data analysis on the reference data obtained from the database DB and executes artificial intelligence machine learning algorithms to train a prediction model.

[0021] The respective controllers CR1 to CRn of the storage devices SRD1 to SRDn actively send the health status data of the storage devices SRD1 to SRDn (including the health status data of the respective controllers CR1 to CRn of the storage devices SRD1 to SRDn, the health status data of the respective storage components ST1 to STn of the storage devices SRD1 to SRDn, or a combination thereof), which are sequentially transmitted to the status prediction component FA through the database DB of the cloud server CLU and the training model establishment component PR.

[0022] The status prediction component FA of the cloud server CLU uses a prediction model established by the training model establishment component PR to predict the future status of the storage devices SRD1 to SRDn based on the health status data of the storage devices SRD1 to SRDn, including predicting whether the storage devices SRD1 to SRDn will fail in the future and predicting the future failure status of the storage devices SRD1 to SRDn.

[0023] More specifically, the active storage device monitoring system of the present invention can execute the active storage device monitoring method of the present invention. The order and content of the steps of the active storage device monitoring method described herein can be appropriately adjusted or omitted according to actual needs, and the present invention is not limited thereto.

[0024] The active storage device monitoring method of the present invention can be applied synchronously or asynchronously to multiple storage devices SRD1 to SRDn, or to any one of the multiple storage devices SRD1 to SRDn. For ease of explanation, the following description only describes the active storage device monitoring method of the present invention as applied to storage device SRD1. It should be understood that the same or similar description can also be applied to the other storage devices SRD2 to SRDn.

[0025] First, as shown in FIG. 2 , the monitoring method of an active storage device according to an embodiment of the present invention may include steps S101 to S106 .

[0026] In step S101 , the host HS sends data to be written into the storage element ST1 of the storage device SRD1 to the controller CR1 of the storage device SRD1 .

[0027] In step S102 , the controller CR1 of the storage device SRD1 stores a predefined total capacity C of the storage device SRD1 and sets a capacity threshold ratio P. The total capacities C of the plurality of storage devices SRD1 to SRDn may be different from each other.

[0028] In step S103 , the controller CR1 of the storage device SRD1 calculates the amount of data HWD that the host HS wants to write into the storage element ST1 of the storage device SRD1 .

[0029] In step S104, the controller CR1 of the storage device SRD1 multiplies the total capacity C of the storage device SRD1 by a critical capacity ratio P to calculate a critical capacity value. The controller CR1 of the storage device SRD1 determines whether the amount of data HWD that the host HS intends to write to the storage element ST1 of the storage device SRD1 is greater than or equal to the critical capacity value. The critical capacity value may be the current remaining capacity / usable capacity of the storage element ST1 of the storage device SRD1.

[0030] If the amount of data HWD that the host HS intends to write to the storage element ST1 of the storage device SRD1 is less than a capacity threshold, step S103 is executed. Specifically, the data currently sent by the host HS is written to the storage element ST1 of the storage device SRD1. The controller CR of the storage device SRD1 continues to receive new data subsequently sent by the host HS. The controller CR1 of the storage device SRD1 then calculates the amount of new data HWD that the host HS intends to write to the storage element ST1 of the storage device SRD1.

[0031] On the contrary, if the amount of data HWD that the host HS intends to write into the storage element ST1 of the storage device SRD1 is greater than or equal to a capacity threshold, step S105 is executed.

[0032] In step S105, the controller CR1 of the storage device SRD1 allows part of the data sent by the host HS to be written into the storage component ST1 of the storage device SRD1, and calculates the amount of data written into the storage component ST1 of the storage device SRD1 (and its proportion to the total capacity C of the storage device SRD1).

[0033] In step S106, the controller CR1 of the storage device SRD1 actively sends the data write / update status information of the storage component ST1 of the storage device SRD1 (including the data content, data volume and its proportion of the total capacity C of the storage device SRD1 that has been written to the storage component ST1 of the storage device SRD1) to the host HS in real time without waiting for the host HS to query, so as to actively notify the host HS and let the host HS know the usage / update status of the storage device SRD1 in real time.

[0034] Please refer to Figures 1 and 3, where Figure 1 is a block diagram of a monitoring system for an active storage device according to an embodiment of the present invention, and Figure 3 is a flow chart of the steps of confirming by a host whether a cloud server supports a storage device included in a monitoring method for an active storage device according to an embodiment of the present invention.

[0035] The monitoring method of the active storage device according to the embodiment of the present invention may include steps S201 to S206 as shown in FIG. 2 .

[0036] In step S201 , the host HS requests the controller CR1 of the storage device SRD1 to report parameters including the cumulative erase times and write times executed on the storage device SRD1 .

[0037] In step S202 , the controller CR1 of the memory device SRD1 counts the number of erase times of the memory device SRD1 to generate an erase number count value EC, and counts the number of write times of the memory device SRD1 to generate a write number count value FWC.

[0038] The controller CR1 of the storage device SRD1 sends an erase count EC and a write count FWC of the storage device SRD1 to the host HS. In practice, the controller CR1 of the storage device SRD1 can actively send an erase count EC and a write count FWC of the storage device SRD1 to the host HS.

[0039] In step S203 , the host HS sends a support confirmation command to the controller CR1 of the storage device SRD1 , which is then transmitted to the cloud server CLU via the controller CR1 of the storage device SRD1 .

[0040] In step S204, the training model building component PR of the cloud server CLU receives a support confirmation instruction through the database DB, determines whether it supports (manages or subscribes to) the storage device SRD1, and sends a support confirmation report message, which is transmitted to the controller CR1 of the storage device SRD1 through the database DB.

[0041] If the controller CR1 of the storage device SRD1 receives a support confirmation report message from the cloud server CLU indicating that the cloud server CLU supports (manages or subscribes to) the storage device SRD1 , step S205 is executed.

[0042] On the contrary, if the controller CR1 of the storage device SRD1 receives a support confirmation report message from the cloud server CLU indicating that the cloud server CLU does not support (manage or subscribe to) the storage device SRD1 , step S206 is executed.

[0043] In step S205 , the controller CR1 of the storage device SRD1 transmits a support confirmation report message received from the cloud server CLU to the host HS, so as to report to the host HS that the cloud server CLU supports (manages or subscribes to) the storage device SRD1 .

[0044] When the host HS confirms that the cloud server CLU supports (or manages or subscribes to) storage device SRD1, the host HS outputs a data provision command to the controller CR1 of storage device SRD1, instructing the controller CR1 of storage device SRD1 to transmit the health status data of storage device SRD1 (including the aforementioned parameter values ​​such as the erase count value EC and the write count value FWC) to the database DB of the cloud server CLU for storage. In this way, the training model building component of the cloud server CLU can monitor the health status data of storage device SRD1.

[0045] In step S206, the controller CR1 of the storage device SRD1 transmits a support confirmation report received from the cloud server CLU to the host HS, reporting to the host HS that the cloud server CLU does not support (manage, or subscribe to) the storage device SRD1. In this case, the host HS does not instruct the controller CR1 of the storage device SRD1 to transmit the health status data of the storage device SRD1 (including the aforementioned parameters such as the erase count value EC and the write count value FWC) to the cloud server CLU.

[0046] Please refer to Figures 1, 4, 6 and 7, wherein Figure 1 is a block diagram of a monitoring system for an active storage device according to an embodiment of the present invention, Figure 4 is a flow chart of the steps of performing storage device fault prediction by a cloud server included in a monitoring method for an active storage device according to an embodiment of the present invention, Figure 6 is a schematic diagram of a curve showing changes in a storage device erase count value over time, established by the monitoring method for an active storage device according to an embodiment of the present invention, and Figure 7 is a schematic diagram of a curve showing changes in a storage device write count value over time, established by the monitoring method for an active storage device according to an embodiment of the present invention.

[0047] The monitoring method of the active storage device according to the embodiment of the present invention may include steps S301 to S307 as shown in FIG. 3 .

[0048] In step S301 , the training model building component PR of the cloud server CLU trains a prediction model.

[0049] In step S302, the database DB of the cloud server CLU receives the number of writes to the storage device SRD1 counted at different time points from the controller CR1 of the storage device SRD1, thereby obtaining a write count value FWC at each time point. The training model building component PR of the cloud server CLU creates a write count variation curve, in a write count variation curve graph, showing how the write count value FWC of the storage device SRD1 changes over time. The training model building component PR of the cloud server CLU calculates the slope of this write count variation curve.

[0050] Furthermore, the database DB of the cloud server CLU receives the number of erase counts of the storage device SRD1 at different time points from the controller CR1 of the storage device SRD1, thereby obtaining an erase count value EC at each time point. The training model building component PR of the cloud server CLU creates an erase count variation curve, in an erase count variation curve graph, showing the change of the erase count value C of the storage device SRD1 over time. The training model building component PR of the cloud server CLU calculates the slope of this erase count variation curve.

[0051] In step S303, the training model building component PR of the cloud server CLU sets a slope of a write count standard curve and a slope tolerance error of the write count slope. In addition, the training model building component PR of the cloud server CLU sets a slope of a erase count standard curve and a slope tolerance error of the erase count slope.

[0052] For ease of explanation, the entire article is described uniformly using curves, but any curve described in this article can be replaced by a straight line in practice, depending on how the data generated in the actual application changes over time.

[0053] The slopes of the multiple write count standard curves of the multiple storage devices SRD1 to SRDn may be different from each other. For example, the slopes CU71 and CU72 of the two write count variation curves of two of the multiple storage devices SRD1 to SRDn shown in FIG. 7 are different from each other.

[0054] The slopes of the erase count standard curves of the plurality of storage devices SRD1 to SRDn may also be different from each other. For example, the slopes of the erase count variation curves CU61 and CU62 of two of the plurality of storage devices SRD1 to SRDn shown in FIG6 are different from each other.

[0055] That is, different thresholds may be set for different memory devices SRD1 to SRDn.

[0056] In step S304 , the training model building component PR of the cloud server CLU determines whether a prediction model has been trained.

[0057] The controller CR1 of the storage device SRD1 can calculate the difference between the slope of a write count variation curve of the storage device SRD1 and the slope of a standard write count curve as a write count error. The controller CR1 of the storage device SRD1 can also calculate the difference between the slope of a erase count variation curve of the storage device SRD1 and the slope of a standard erase count curve as an erase count error.

[0058] When a write count error of the storage device SRD1 is greater than a write count slope tolerance error or an erase count error is greater than an erase count slope tolerance error, the training model building component PR of the cloud server CLU may retrain a prediction model.

[0059] If the training model building component PR of the cloud server CLU has not yet trained a prediction model, the process returns to step S302. On the contrary, if the training model building component PR of the cloud server CLU has trained a prediction model, the process proceeds to step S305.

[0060] In step S305 , the training model building component PR of the cloud server CLU maintains or updates the slope of a programming count variation curve and the slope of an erasing count variation curve calculated in step S302 .

[0061] In step S306, the training model establishment component PR of the cloud server CLU determines whether the slope of a write count change curve of the storage device SRD1 is the same as the slope of a write count standard curve and whether the slope of an erase count change curve of the storage device SRD1 is the same as the slope of a erase count standard curve, so as to determine whether the usage status of the storage device SRD1 violates the standard.

[0062] If the slope of a write count change curve of the storage device SRD1 is the same as the slope of a write count standard curve (i.e., a standard value) and it is determined that the slope of a erase count change curve of the storage device SRD1 (i.e., a standard value) is the same as the slope of a erase count standard curve, the training model establishment component PR of the cloud server CLU determines that the usage status of the storage device SRD1 meets the standard, and then executes step S307.

[0063] Conversely, if the slope of a write count variation curve for storage device SRD1 differs from the slope of a write count standard curve, or the slope of an erase count variation curve for storage device SRD1 differs from the slope of a erase count standard curve, the training model building component PR of the cloud server CLU determines that the usage status of storage device SRD1 violates the standard and returns to step S305. At this point, in step S305, the training model building component PR of the cloud server CLU recalculates / updates the slope of a write count standard curve and the slope of an erase count variation curve based on data obtained from the database DB showing more data writes or erases performed on storage device SRD1. Next, step S306 is executed to determine whether the updated slopes of a write count standard curve and an erase count variation curve for storage device SRD1 meet the aforementioned standard values.

[0064] In step S307, the training model establishment component PR of the cloud server CLU predicts whether the storage device SRD1 may fail in the future based on the health status data of the storage device SRD1 (for example, a write count error and an erase count error of the storage device SRD1 mentioned above), and predicts the future failure status data of the storage device SRD1 (that is, the health status data of the storage device SRD1 when it fails in the future).

[0065] Please refer to Figures 1 and 5, where Figure 1 is a block diagram of a monitoring system for an active storage device according to an embodiment of the present invention, and Figure 5 is a flow chart of the steps of backing up data of the storage device by a cloud server included in a monitoring method for an active storage device according to an embodiment of the present invention.

[0066] The monitoring method of an active storage device according to an embodiment of the present invention may include steps S401 to S404 as shown in FIG. 4 , which may be performed after predicting that the storage device SRD1 may fail in the future.

[0067] In step S401, the database DB of the cloud server CLU obtains data of the storage device SRD1 from the controller CR1 of the storage device SRD1. In addition to the health status data and multiple setting parameters of the storage device SRD1 (including the above-mentioned erase count value EC and write count value FWC), other data stored in the storage component ST1 of the storage device SRD1 can also be obtained, especially when it is predicted that the storage device SRD1 may fail as mentioned above.

[0068] In step S402, the database DB of the cloud server CLU receives data (including data stored in the storage device ST1 of the storage device SRD1, operational data of the controller CR1 of the storage device SRD1, or a combination thereof) from the controller CR1 of the storage device SRD1 and backs up the received data from the storage device SRD1. If necessary, the training model building component PR of the cloud server CLU may encrypt the data backed up in the database DB of the storage device SRD1.

[0069] In step S403 , the training model building component PR of the cloud server CLU determines whether the database DB of the cloud server CLU has completed the data backup of the storage device SRD1 .

[0070] If the database DB of the cloud server CLU has not completed the data backup of the storage device SRD1 , the step S402 is continued to be executed. On the contrary, if the database DB of the cloud server CLU has completed the data backup of the storage device SRD1 , the step S404 is continued to be executed.

[0071] In step S404 , the cloud server CLU issues a potential failure notification message to the storage device SRD1 that may potentially fail.

[0072] In summary, the present invention provides a method for monitoring an active storage device, wherein the storage device actively sends its own health status data to a cloud server, enabling the cloud server to monitor changes in the storage device's health status data in real time, thereby dynamically predicting the future status of the storage device in real time. When a future failure is predicted, the cloud server can issue a warning and perform operations such as backing up the storage device's data.

[0073] The contents disclosed above are only preferred feasible embodiments of the present invention and are not intended to limit the claims of the present invention. Therefore, any equivalent technical changes made using the contents of the present invention's description and drawings are included in the claims of the present invention.

Claims

1. A monitoring method for an active storage device, characterized in that, The monitoring method of the active storage device described above includes the following steps: Using a cloud server to train a prediction model; A storage device actively sends a health status data of the storage device to the cloud server; And Using the cloud server, using the prediction model, and based on the health status data of the storage device, to predict the future state of the storage device.

2. The monitoring method of the active storage device according to claim 1, wherein, The monitoring method of the active storage device described above further includes the following steps: Using the storage device to determine whether the amount of data sent from a host to the storage device is greater than or equal to a capacity threshold value. If not, allow the host to write data to the storage device. If so, calculate the amount of data allowed to be written by the host to the storage device, calculate the amount of data that the host has written to the storage device, and send a data update notification message to the host.

3. The monitoring method of the active storage device according to claim 2, characterized in that, The monitoring method of the active storage device described above further includes the following steps: Using the storage device to multiply the total capacity of the storage device by a capacity critical ratio to calculate the capacity threshold value.

4. The monitoring method of the active storage device according to claim 1, characterized in that The monitoring method of the active storage device described above further includes the following steps: A host sends a support confirmation instruction to the storage device; The storage device transmits the support confirmation instruction received from the host to the cloud server; The cloud server determines whether it supports the storage device based on the received support confirmation instruction, and sends a support confirmation return message to the storage device; Through the storage device, the support confirmation return message of the cloud server is transmitted to the host; And The host determines whether the cloud server supports the storage device based on the support confirmation return message. If so, outputs a data providing instruction to the storage device to instruct the storage device to send the health status data to the cloud server. If not, does not instruct the storage device to send the health status data to the cloud server.

5. The monitoring method of the active storage device according to claim 1, wherein The monitoring method of the active storage device described above further includes the following steps: The storage device counts the number of erasure operations performed by the storage device to generate an erasure count value; and The storage device sends the erasure count value to a host.

6. The monitoring method of the active storage device according to claim 5, wherein The monitoring method of the active storage device described above further includes the following steps: Using the cloud server to establish an erasure count change curve of the storage device that changes with time in an erasure count change curve graph.

7. The monitoring method of the active storage device according to claim 6, characterized in that The monitoring method of the active storage device described above further includes the following steps: Using the cloud server to calculate the slope of the erasure count change curve of the storage device that changes with time; Using the cloud server to set a standard curve slope of the erasure count; Using the cloud server to determine whether the slope of the erasure count change curve of the storage device is the same as the standard curve slope of the erasure count. If so, perform the next step. If not, do not perform the next step; And Using the cloud server, predict the future failure state of the storage device according to the difference between the slope of the erasure count change curve of the storage device and the slope of the erasure count standard curve.

8. The monitoring method of the active storage device according to claim 7, characterized in that, The monitoring method of the active storage device further includes the following steps: Using the cloud server, set an erasure count slope tolerance error; Using the cloud server, calculate the difference between the slope of the erasure count change curve of the storage device and the slope of the erasure count standard curve as an erasure count error; and Using the cloud server, compare the erasure count error with the erasure count slope tolerance error to determine whether to predict the future failure state of the storage device.

9. The monitoring method of the active storage device according to claim 8, characterized in that, The monitoring method of the active storage device further includes the following steps: Using the cloud server, use the prediction model, according to the difference between the slope of the erasure count change curve of the storage device and the slope of the erasure count standard curve, to predict whether there is a potential possibility of future failure of the storage device. If so, perform the next step. If not, do not perform the next step; And Using the cloud server, obtain and back up the data of the storage device. The monitoring method of the active storage device further includes the following steps:

10. The monitoring method of the active storage device according to claim 9, characterized in that, Using the cloud server, encrypt the data of the storage device with a potential possibility of failure. The monitoring method of the active storage device further includes the following steps:

11. The monitoring method of the active storage device according to claim 9, wherein Using the cloud server, issue a potential failure notification message for the storage device with a potential possibility of failure. The monitoring method of the active storage device further includes the following steps:

12. The monitoring method of the active storage device according to claim 1, characterized in that, By the storage device, count the write times of the storage device to generate a write count value; and By the storage device, send the write count value to a host. The monitoring method of the active storage device further includes the following steps:

13. The monitoring method of the active storage device according to claim 12, wherein Using the cloud server, establish a write count change curve showing how the write count value of the storage device changes over time in a write count change curve graph. The monitoring method of the active storage device further includes the following steps:

14. The monitoring method of the active storage device according to claim 13, characterized in that, Using the cloud server, calculate the slope of the write count change curve showing how the write count value of the storage device changes over time; Using the cloud server, set a write count standard curve slope; Using the cloud server, determine whether the slope of the write count change curve of the storage device is the same as the write count standard curve slope. If so, perform the next step. If not, do not perform the next step; And Using the cloud server, predict the future failure state of the storage device according to the difference between the slope of the write count change curve of the storage device and the slope of the write count standard curve. The monitoring method of the active storage device further includes the following steps:

15. The monitoring method of the active storage device according to claim 14, wherein, Using the cloud server, set a write count slope tolerance error; ​ Using the cloud server, calculate the difference between the slope of the write count change curve of the storage device and the slope of the write count standard curve as a write count error; And Using the cloud server, compare the write count error with the write count slope tolerance error to determine whether to predict the future failure state of the storage device.

16. The monitoring method of the active storage device according to claim 15, characterized in that, The monitoring method of the active storage device further comprises the following steps: Using the cloud server, using the prediction model, based on the difference between the slope of the write count change curve of the storage device and the slope of the write count standard curve, to predict whether there is a potential possibility of future failure of the storage device. If so, execute the next step. If not, do not execute the next step; And Using the cloud server, obtain and back up the data of the storage device.

17. The monitoring method of the active storage device according to claim 16, wherein The monitoring method of the active storage device further comprises the following steps: Using the cloud server, encrypt the data of the storage device with a potential possibility of failure.

18. The monitoring method of the active storage device according to claim 16, wherein The monitoring method of the active storage device further comprises the following steps: Using the cloud server, issue a potential failure notification message for the storage device with a potential possibility of failure.

Images