Method, electronic device and computer program product for managing storage devices

Abstract

Embodiments according to the present disclosure relate to methods, electronic devices and computer program products for managing storage devices. A method for managing storage devices includes obtaining evaluation information for at least one candidate fitting model in a set of candidate fitting models. The evaluation information indicates at least a degree of historical fitting between the at least one candidate fitting model and a set of historical data. The method further includes determining a model selection strategy based on the evaluation information. The model selection strategy indicates at least a probability that the set of candidate fitting models is selected for fitting target data. The target data indicates a storage state of a storage device over a first time period. Furthermore, the method further includes determining a target fitting model from the set of candidate fitting models based on the model selection strategy for determining a storage state of the storage device over a second time period. In this way, a fitting model that is most suitable for actual business data can be determined more efficiently, and additional system overhead can be avoided.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of storage technology, and more specifically, to methods, electronic devices, and computer program products for managing storage devices. Background Technology

[0002] With the continuous development of internet technology, the demand for data processing and applications is constantly growing, and the amount of data needed and generated by people has increased significantly. People now recognize the importance of storage management. One known storage management method is to predict the storage status of storage devices in the future based on historical data, so as to manage the storage devices based on the predicted storage status. Summary of the Invention

[0003] According to an example embodiment of this disclosure, a scheme for managing storage devices is provided. This scheme determines a target fitting model for predicting the storage state of the storage device from a set of candidate fitting models by considering previously determined evaluation information associated with at least one candidate fitting model. Compared to conventional solutions, the scheme according to this disclosure can more efficiently determine the best-fitting model for actual business data and avoids additional system overhead.

[0004] In a first aspect of this disclosure, a method for managing a storage device is provided. The method includes acquiring evaluation information for at least one candidate fitting model from a set of candidate fitting models. The evaluation information at least indicates the historical fit between the at least one candidate fitting model and a set of historical data. The set of historical data indicates the storage state of the storage device within at least one historical time period. The method further includes determining a model selection strategy based on the evaluation information. The model selection strategy at least indicates the probability that a set of candidate fitting models will be selected to fit target data. The target data indicates the storage state of the storage device within a first time period. Furthermore, the method includes determining a target fitting model from the set of candidate fitting models based on the model selection strategy to determine the storage state of the storage device within a second time period.

[0005] In a second aspect of this disclosure, an electronic device is provided. The electronic device includes a processor and a memory coupled to the processor, the memory having instructions stored therein, the instructions causing the device to perform actions when executed by the processor. The actions include acquiring evaluation information for at least one candidate fitting model from a set of candidate fitting models. The evaluation information at least indicates the historical fit between the at least one candidate fitting model and a set of historical data. The set of historical data indicates the storage state of the storage device within at least one historical time period. The actions also include determining a model selection strategy based on the evaluation information. The model selection strategy at least indicates the probability that a set of candidate fitting models will be selected to fit target data. The target data indicates the storage state of the storage device within a first time period. Furthermore, the actions also include determining a target fitting model from a set of candidate fitting models based on the model selection strategy to determine the storage state of the storage device within a second time period.

[0006] In a third aspect of this disclosure, a computer program product is provided, which is tangibly stored on a computer-readable medium and includes machine-executable instructions that, when executed, cause a machine to perform the method according to the first aspect.

[0007] The summary section is provided to present the chosen concepts in a simplified form, which will be further described in the detailed description below. The summary section is not intended to identify key or principal features of this disclosure, nor is it intended to limit the scope of this disclosure. Attached Figure Description

[0008] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0009] Figure 1 A block diagram of an example environment according to some embodiments of the present disclosure is shown;

[0010] Figure 2 A flowchart of a method for managing a storage device according to some embodiments of the present disclosure is shown;

[0011] Figure 3 A flowchart illustrating a method for determining a model selection strategy according to some embodiments of the present disclosure is shown; and

[0012] Figure 4 A schematic block diagram of an example device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0013] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0014] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0015] As mentioned above, conventional storage management methods manage storage devices by predicting their storage state over future time periods. Known methods typically process historical data by trying different fitting models one by one in a predetermined order and determining the fitting model that best suits the actual business data, so that this fitting model can be used to predict the storage state of the storage device. However, since different fitting models fit different types of actual business data to varying degrees, this conventional method usually requires numerous trials to determine the suitable fitting model, thus introducing unwanted system overhead. Furthermore, because this conventional method often stops trying other fitting models after finding a fitting model that meets the minimum requirements, the final fitting model used for prediction may not be the most suitable among the candidate fitting models for the actual business data, thereby affecting the prediction results.

[0016] Embodiments of this disclosure provide a scheme for managing storage devices to address one or more of the problems described above and other potential problems. In this disclosure, a target fitting model is determined from a set of candidate fitting models by considering previously determined evaluation information associated with at least one candidate fitting model, for predicting the storage state of the storage device over a future time period. Compared to the conventional schemes described above, because the previously determined evaluation information is considered during the model selection process, the scheme according to this disclosure can more efficiently determine the most suitable fitting model for actual business data from a set of candidate fitting models, and can reduce the number of attempts at candidate fitting models in subsequent prediction processes, thereby avoiding unwanted system overhead.

[0017] The following description will continue with reference to the accompanying drawings, which will provide some exemplary embodiments of this disclosure.

[0018] Figure 1A block diagram of an example environment 100 according to some embodiments of the present disclosure is shown. Figure 1 As shown, the example environment 100 may generally include storage device 110 and electronic device 120. In some embodiments, storage device 110 may include one or more physical storage devices. Examples of physical storage devices include, but are not limited to, solid-state drives (SSDs), Fibre Channel (FC) drives, Serial Advanced Technology Accessories (SATA) drives, Serial Connected Small Computer System Interface (SAS) drives, and so on. In some embodiments, storage device 110 may be configured as a Redundant Array of Independent Disks (RAID). It should be understood that storage device 110 may also be implemented in any other suitable form, and the scope of this disclosure is not limited in this respect.

[0019] like Figure 1 As shown, electronic device 120 may be communicatively coupled to storage device 110, for example, to receive data from and / or send data to storage device 110. According to this disclosure, electronic device 120 determines a target fitting model 140 from a set of candidate fitting models 130 for predicting the storage state of storage device 110 by considering previously determined evaluation information associated with at least one candidate fitting model 130. This will be described in conjunction with... Figures 2 to 3 Further detailed description. In some embodiments, electronic device 120 may be a device with computing capabilities, such as an electronic control module, a personal computer, a workstation, or a server. It should be understood that, although in Figure 1 In this disclosure, electronic device 120 is shown as being located outside of storage device 110, but electronic device 120 may also be implemented as being located inside storage device 110 or as part of storage device 110, and the scope of this disclosure is not limited in this respect.

[0020] Figure 2 A flowchart of a method 200 for managing a storage device 110 according to some embodiments of the present disclosure is shown. For example, method 200 may be performed by, for example, Figure 1 The illustrated electronic device 120 performs this action. It should be understood that method 200 may also include additional boxes not shown, and / or the boxes shown may be omitted. The scope of this disclosure is not limited in this respect.

[0021] At box 202, electronic device 120 acquires evaluation information for at least one candidate fitting model 130 in a set of candidate fitting models 130, the evaluation information indicating at least the historical fitting degree between at least one candidate fitting model 130 and a set of historical data, the set of historical data indicating the storage status of storage device 110 in at least one historical time period.

[0022] In some embodiments, the electronic device 120 may, for example, be communicatively coupled to a recording device for storing evaluation information associated with the candidate fitted model 130. Figure 1 (Not shown in the image). For example, after a candidate fitting model 130 is evaluated, the evaluation information obtained can be stored by a recording device for use in subsequent model selection processes. Electronic device 120 can obtain evaluation information for at least one candidate fitting model 130 from a set of candidate fitting models 130 by sending a request to the recording device. In some embodiments, electronic device 120 itself can also be configured to store evaluation information associated with a candidate fitting model 130. It should be understood that any other suitable device can also store and provide evaluation information associated with a candidate fitting model 130, and electronic device 120 can obtain evaluation information from that device; the scope of this disclosure is not limited in this respect.

[0023] In some embodiments, candidate fitting model 130 may include a regression model for performing regression analysis on the storage status of storage device 110 over a historical time period. Examples of regression models include, but are not limited to, linear regression models, autoregressive moving average (ARMA) models, autoregressive integrated moving average (ARIMA) models, etc. It should be understood that candidate fitting model 130 may also include any other suitable models for analyzing the storage status of storage device 110 over a historical time period, and the scope of this disclosure is not limited in this respect.

[0024] In some embodiments, the historical time period is a past period of time and may have a predetermined length. The historical time period may be, for example, a past day, a past week, or a past month, etc. In some embodiments, the length of the historical time period can be selected and set by the user. The scope of this disclosure is not limited in this respect.

[0025] In some embodiments, historical data may include storage status data collected by storage device 110 at predetermined time intervals within a historical period, such as storage device utilization. Depending on the length of the historical period, the predetermined time interval may be, for example, one hour, one day, etc., and the scope of this disclosure is not limited in this respect.

[0026] In some embodiments, evaluation information may indicate the degree of historical fit between the candidate fitting model 130 and corresponding historical data in a set of historical data. In some embodiments, the electronic device 120 measures the degree of historical fit between the candidate fitting model 130 and the corresponding historical data by calculating indicators such as root-mean squared value (RMS), root-mean square error (RMSE), and coefficient of determination (R²), and may store the values ​​of the above indicators as evaluation information. It should be understood that the degree of historical fit between the candidate fitting model 130 and the corresponding historical data may also be measured in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0027] In some embodiments, the evaluation information may also indicate the time cost of at least one candidate fitting model 130 for fitting a set of historical data. For example, the electronic device 120 may measure the time cost of the candidate fitting model 130 by recording time information such as the time required to construct the candidate fitting model 130 and / or the time spent making predictions using the constructed candidate fitting model 130. It should be understood that the time cost of the candidate fitting model 130 may also be measured in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0028] In some embodiments, the evaluation information may also indicate the storage overhead of at least one candidate fitting model 130 for fitting a set of historical data. For example, the electronic device 120 may measure the storage overhead of the candidate fitting model 130 by recording the storage capacity required when fitting the historical data using the candidate fitting model 130. It should be understood that the storage overhead of the candidate fitting model 130 may also be measured in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0029] In some embodiments, the evaluation information may also indicate the moment when at least one candidate fitting model 130 is selected to determine the storage state of the storage device 110, which may be referred to as the "recent usage moment". As will be described in detail below, the electronic device 120 will determine a candidate fitting model 130 from a set of candidate fitting models 130 based on the evaluation information to determine the storage state of the storage device 110 in a future time period. Therefore, the electronic device 120 may record the moment when the candidate fitting model 130 is selected to determine the storage state of the storage device 110.

[0030] The evaluation metrics that may be included in the evaluation information have been described above by way of example. In some embodiments, the evaluation information may include at least one of the evaluation metrics described above. It should be understood that the evaluation information may also include evaluation metrics not listed, and / or the listed evaluation metrics may be omitted. The scope of this disclosure is not limited in this respect.

[0031] At box 204, electronic device 120 determines a model selection strategy based on evaluation information. The model selection strategy indicates at least one set of candidate fitting models 130 that are selected to fit the target data, which indicates the storage status of storage device 110 in the first time period.

[0032] In some embodiments, the first time period is a past period of time and may have a predetermined length. The first time period may be, for example, the most recent day, the most recent week, the most recent 30 days, or the most recent calendar month, etc. In some embodiments, the first time period may be selected and set by the user. The scope of this disclosure is not limited in this respect.

[0033] In some embodiments, the target data may include storage status data collected by storage device 110 at predetermined time intervals during a first time period, such as the utilization rate of the storage device. Depending on the length of the first time period, the predetermined time interval may be, for example, one hour, one day, etc., and the scope of this disclosure is not limited in this respect.

[0034] In some embodiments, the electronic device 120 may determine, based on evaluation information, the probability that each candidate fitting model 130 will be selected for fitting the target data as a model selection strategy. This will be discussed in conjunction with the following. Figure 3 Further detailed description.

[0035] At box 206, electronic device 120 determines target fitting model 140 from a set of candidate fitting models 130 based on a model selection strategy, in order to determine the storage state of storage device 110 in the second time period.

[0036] In some embodiments, the second time period is a future period of time, and may have a predetermined length of time, for example. The second time period may be, for example, a future day, a future week, a future 30 days, etc. In some embodiments, the second time period may be selected and set by the user. The scope of this disclosure is not limited in this respect.

[0037] In some embodiments, electronic device 120 may determine a first fitting model from a set of candidate fitting models 130 based on a model selection strategy. Electronic device 120 may use the determined first fitting model to fit the target data to determine the degree of fit between the first fitting model and the target data. Electronic device 120 may use the degree of fit between the determined first fitting model and the target data to update the historical fit associated with the first fitting model in the evaluation information. When the evaluation information also indicates the time cost mentioned above, electronic device 120 may, for example, use the time required to construct the first fitting model to update the time cost associated with the first fitting model in the evaluation information. Depending on the evaluation metrics included in the evaluation information, electronic device 120 may also update the evaluation information in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0038] In some embodiments, the electronic device 120 may compare the degree of fit between the determined first fitting model and the target data with a predetermined threshold. This predetermined threshold may be pre-determined by a user, may be a default value, or may be adjusted during method execution; the scope of this disclosure is not limited in this respect. If the degree of fit is determined to be greater than or equal to the predetermined threshold, the electronic device 120 may determine the first fitting model as the target fitting model 140. If the degree of fit is determined to be less than the predetermined threshold, the electronic device 120 may update the model selection strategy based on updated evaluation information. Exemplarily, the electronic device 120 may determine a new model selection strategy based on updated evaluation information in a manner similar to that described in reference block 204 to update the model selection strategy. The electronic device 120 may also update the model selection strategy in any other suitable manner; the scope of this disclosure is not limited in this respect. The electronic device 120 may determine the target fitting model 140 from a set of candidate fitting models 130 based on the updated model selection strategy in a manner similar to that described above. In some embodiments, the electronic device 120 can perform the operations described above iteratively to determine a target fitting model 140 from a set of candidate fitting models 130 that has a fitting degree greater than or equal to a predetermined threshold for the target data. In this way, the electronic device 120 can update the evaluation information associated with the candidate fitting models 130 in real time during the determination of the target fitting model 140, and adaptively adjust the model selection strategy based on the latest evaluation information. This allows for more efficient determination of the most suitable fitting model for the actual business data from a set of candidate fitting models 130, and reduces the number of attempts to re-apply to the candidate fitting models 130 in subsequent prediction processes, thereby avoiding unwanted system overhead.

[0039] In other embodiments, the electronic device 120 may compare the degree of fit between the determined first fitting model and the target data with a predetermined threshold. This predetermined threshold may be pre-determined by the user, may be a default value, or may be adjusted during method execution; the scope of this disclosure is not limited in this respect. If the degree of fit is determined to be greater than or equal to the predetermined threshold, the electronic device 120 may determine the first fitting model as the target fitting model 140. If the degree of fit is determined to be less than the predetermined threshold, the electronic device 120 may continue to select a candidate fitting model 130 from a set of candidate fitting models 130 as the first fitting model based on the model selection strategy determined at block 204, and repeat the above operation until a target fitting model 140 with a degree of fit to the target data greater than or equal to the predetermined threshold is determined from the set of candidate fitting models 130. In some embodiments, the electronic device 120 may also use the degree of fit between the first fitting model and the target data to update evaluation information. The electronic device 120 may determine the target fitting model 140 from a set of candidate fitting models 130 based on the updated evaluation information to determine the storage state of the storage device 110 in a third time period following the second time period. In this way, the electronic device 120 can consider the most recent evaluation information of the candidate fitting models 130 that have been tried before in the process of determining the target fitting model 140, so as to more efficiently determine the fitting model that is most suitable for the actual business data from a set of candidate fitting models 130, thereby reducing the number of attempts to the candidate fitting models 130 in the subsequent prediction process and avoiding unwanted system overhead.

[0040] Figure 3 A flowchart of a method 300 for determining a model selection strategy according to some embodiments of the present disclosure is shown. For example, method 300 may be as follows: Figure 2 An example implementation of box 204 is shown. In some embodiments, method 300 may be provided by, for example... Figure 1 The illustrated electronic device 120 performs this action. It should be understood that method 300 may also include additional boxes not shown and / or the boxes shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0041] At block 302, electronic device 120 determines the best candidate fitting model from at least one candidate fitting model 130 based on evaluation information of at least one candidate fitting model 130. In some embodiments, where the evaluation information only indicates the historical fit between at least one candidate fitting model 130 and a set of historical data, electronic device 120 may determine the candidate fitting model 130 with the best historical fit among the at least one candidate fitting model 130 as the best candidate fitting model. In other embodiments, where the evaluation information indicates multiple different evaluation metrics, electronic device 120 may, for example, determine an evaluation value associated with a corresponding candidate fitting model 130 based on the evaluation information, and determine the best candidate fitting model based on the determined evaluation value.

[0042] For example, given the evaluation information indicating the historical fit, time cost, storage cost, and most recent usage time mentioned above, electronic device 120 can calculate the evaluation value associated with the corresponding candidate fitting model 130 based on the following equation (1):

[0043]

[0044] Where A represents the candidate fitting model to be evaluated, E(A) represents the evaluation value associated with the candidate fitting model A, D(A) represents the historical fit between the candidate fitting model A and the corresponding historical data, TO(A) represents the time cost of the candidate fitting model A for fitting the corresponding historical data, SO(A) represents the storage cost of the candidate fitting model A for fitting the corresponding historical data, current_time represents the current time, and LSPT(A) represents the most recent usage time associated with the candidate fitting model A.

[0045] In this example, electronic device 120 can identify the candidate fitting model 130 with the highest evaluation value as the best candidate fitting model. In other words, electronic device 120 tends to identify the candidate fitting model 130, which has a good historical fit, low time cost, low storage cost, and has been used recently, as the best candidate fitting model. In this way, the performance of candidate fitting models can be comprehensively measured from multiple dimensions in order to select the most suitable fitting model. It should be noted that equation (1) is only intended to provide an exemplary way of determining the evaluation value, and the evaluation value can also be determined based on the evaluation information in any other suitable way, such as calculating the weighted sum of all indicators included in the evaluation information. The scope of this disclosure is not limited in this respect.

[0046] It should be understood that, in addition to determining the evaluation value based on the evaluation information, the best candidate fitting model can be determined from at least one candidate fitting model 130 in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0047] At box 304, the electronic device 120 can set the probability of the best candidate fitted model being selected as a first probability value. In some embodiments, the first probability value can be a predetermined value, such as 40%. In some embodiments, the first probability value can be a variable value.

[0048] In some embodiments, the electronic device 120 can count model selection strategies to determine the cumulative number of model selection strategies, and the electronic device 120 can set a first probability value to increase as the cumulative number increases. In a specific example, the first probability value can have an initial value of 50%, and the first probability value increases by 10% with each increment of the cumulative number by 1 until it reaches 100%, and then the first probability value always remains at 100%. This means that after the cumulative number of model selection strategies is greater than 5, the first probability value will remain at 100%, that is, subsequent model selection strategies will always indicate the selection of the determined best candidate fitting model, without trying other candidate fitting models 130. In this way, the electronic device 120 can directly select the best candidate fitting model determined according to the stored evaluation information as the target fitting model 140 after accumulating a certain amount of model selection experience, in order to determine the storage state of the storage device 110 in the second time period. By reasonably setting the first probability value and its variation rules, the method according to this embodiment can achieve an appropriate balance between utilizing the best candidate fitting model and exploring candidate fitting models 130, thereby achieving the determination of a better-performing target fitting model 140 with higher efficiency. Therefore, compared to conventional solutions that always try candidate fitting models 130 in a predetermined order, the method according to this embodiment can further reduce the number of attempts required to determine the target fitting model 140, thereby saving system overhead and improving prediction efficiency and prediction quality. It should be understood that the first probability value can also be set in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0049] In some embodiments, a threshold N for the number of attempts can be preset, and the first probability value is set to 100% after the number of model selection strategies exceeds the threshold N. In this case, after trying a certain number of candidate fitting models 130, the electronic device 120 always determines the best candidate fitting model as the target fitting model 140. In this way, trying the remaining candidate fitting models 130 can be avoided in subsequent prediction processes, thereby further saving system overhead and improving prediction efficiency.

[0050] At box 306, electronic device 120 determines a model selection strategy based on a first probability value. In some embodiments, electronic device 120 may determine a second probability value as the probability value obtained by subtracting the first probability value from 100%, and distribute the second probability value equally among the remaining candidate fitting models 130 in a set of candidate fitting models 130, excluding the best candidate fitting model. In other words, electronic device 120 may randomly select a candidate fitting model 130 from the remaining candidate fitting models 130 based on the second probability value. For example, when the first probability value is 50% and the set of candidate fitting models 130 has a total of 6 candidate fitting models 130, electronic device 120 may determine a model selection strategy as follows: the probability of the best candidate fitting model being selected is 50%, and the probability of the remaining 5 candidate fitting models 130 being selected is 10%.

[0051] In some embodiments, the electronic device 120 may determine a second probability value as the probability value obtained by subtracting the first probability value from 100%, and distribute the second probability value equally among a set of candidate fitting models 130 that have not yet been tried. In this case, during subsequent model selection, the electronic device 120 may disregard candidate fitting models 130 that have been previously tried and whose performance is lower than the best candidate fitting model. In this way, the efficiency of model selection can be further improved, thereby saving system overhead.

[0052] It should be understood that the model selection strategy can also be determined based on the first probability value in any other suitable manner, and the scope of this disclosure is not limited in this respect.

[0053] Through the above combination Figures 1 to 3 As described, the method for managing storage devices according to this disclosure can determine a target fitting model from a set of candidate fitting models by considering previously determined evaluation information associated with at least one candidate fitting model, in order to predict the storage state of the storage device over a future time period. Compared to the conventional approach of trying candidate fitting models in a predetermined order, the method according to this disclosure can more efficiently determine the most suitable fitting model for actual business data from a set of candidate fitting models because the previously determined evaluation information is considered during the model selection process. This reduces the number of attempts to try candidate fitting models in subsequent prediction processes, thereby avoiding unwanted system overhead.

[0054] Figure 4 A schematic block diagram of an example device 400 that can be used to implement embodiments of this disclosure is shown. Figure 4As shown, device 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) 402 or loaded from storage unit 408 into random access memory (RAM) 403. The RAM 403 may also store various programs and data required for the operation of device 400. The CPU 401, ROM 402, and RAM 403 are interconnected via bus 404. Input / output (I / O) interface 405 is also connected to bus 404.

[0055] Multiple components in device 400 are connected to I / O interface 405, including: input unit 406, such as keyboard, mouse, etc.; output unit 407, such as various types of monitors, speakers, etc.; storage unit 408, such as disk, optical disk, etc.; and communication unit 409, such as network card, modem, wireless transceiver, etc. Communication unit 409 allows device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0056] The various processes and procedures described above, such as methods 200 and 300, can be executed by processing unit 401. For example, in some embodiments, methods 200 and 300 can be implemented as computer software programs tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by CPU 401, one or more actions of methods 200 and 300 described above can be performed.

[0057] This disclosure can be a method, apparatus, system, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of this disclosure.

[0058] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0059] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0060] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0061] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0062] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0063] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0064] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0065] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for managing a storage device, comprising: Obtain evaluation information for at least one candidate fitting model from a set of candidate fitting models, the evaluation information indicating at least the historical fit between the at least one candidate fitting model and a set of historical data, the set of historical data indicating the storage status of the storage device within at least one historical time period; Based on the evaluation information, a model selection strategy is determined. The model selection strategy at least indicates the probability that the set of candidate fitting models is selected to fit the target data, and the target data indicates the storage status of the storage device in the first time period. as well as Based on the model selection strategy, a target fitting model is determined from the set of candidate fitting models to determine the storage state of the storage device in the second time period.

2. The method according to claim 1, wherein determining the model selection strategy comprises: Based on the evaluation information of the at least one candidate fitting model, the best candidate fitting model is determined from the at least one candidate fitting model; The probability of the best candidate fitting model being selected is set as the first probability value; as well as Based on the first probability value, the model selection strategy is determined.

3. The method according to claim 2, further comprising: The model selection strategies are counted to determine the cumulative number of model selection strategies, and the first probability value increases as the cumulative number increases.

4. The method according to claim 1, wherein determining the target fitting model comprises: Based on the model selection strategy, a first fitting model is determined from the set of candidate fitting models; The evaluation information is updated using the degree of fit between the first fitting model and the target data; In response to determining that the fit is less than a predetermined threshold, the model selection strategy is updated based on the updated evaluation information; as well as The target fitting model is determined from the set of candidate fitting models based on the updated model selection strategy.

5. The method of claim 1, wherein determining the target fitting model comprises: Based on the model selection strategy, a first fitting model is determined from the set of candidate fitting models; as well as In response to determining that the degree of fit between the first fitting model and the target data is less than a predetermined threshold, the target fitting model is determined from the set of candidate fitting models based on the model selection strategy.

6. The method according to claim 5, further comprising: The evaluation information is updated using the degree of fit between the first fitting model and the target data to determine the storage status of the storage device in a third time period following the second time period.

7. The method according to any one of claims 1 to 6, wherein the evaluation information further indicates at least one of the following: the time or storage overhead of the at least one candidate fitting model for fitting the set of historical data, or the moment when the at least one candidate fitting model is selected to determine the storage state of the storage device.

8. An electronic device, comprising: processor; as well as A memory coupled to the processor, the memory having instructions stored therein, the instructions causing the device to perform actions when executed by the processor, the actions including: Obtain evaluation information for at least one candidate fitting model from a set of candidate fitting models, the evaluation information indicating at least the historical fit between the at least one candidate fitting model and a set of historical data, the set of historical data indicating the storage status of the storage device within at least one historical time period; A model selection strategy is determined based on the evaluation information. This model selection strategy at least indicates the probability that the set of candidate fitting models will be selected to fit the target data, where the target data indicates the storage status of the storage device during a first time period. Based on the model selection strategy, a target fitting model is determined from the set of candidate fitting models to determine the storage state of the storage device in the second time period.

9. The electronic device of claim 8, wherein determining the model selection strategy comprises: Based on the evaluation information of the at least one candidate fitting model, the best candidate fitting model is determined from the at least one candidate fitting model; The probability of the best candidate fitting model being selected is set as the first probability value; as well as Based on the first probability value, the model selection strategy is determined.

10. The electronic device of claim 9, wherein the action further comprises: The model selection strategies are counted to determine the cumulative number of model selection strategies, and the first probability value increases as the cumulative number increases.

11. The electronic device of claim 8, wherein determining the target fitting model comprises: Based on the model selection strategy, a first fitting model is determined from the set of candidate fitting models; The evaluation information is updated using the degree of fit between the first fitting model and the target data; In response to determining that the fit is less than a predetermined threshold, the model selection strategy is updated based on the updated evaluation information; as well as The target fitting model is determined from the set of candidate fitting models based on the updated model selection strategy.

12. The electronic device of claim 8, wherein determining the target fitting model comprises: Based on the model selection strategy, a first fitting model is determined from the set of candidate fitting models; as well as In response to determining that the degree of fit between the first fitting model and the target data is less than a predetermined threshold, the target fitting model is determined from the set of candidate fitting models based on the model selection strategy.

13. The electronic device of claim 12, wherein the action further comprises: The evaluation information is updated using the degree of fit between the first fitting model and the target data to determine the storage status of the storage device in a third time period following the second time period.

14. The electronic device according to any one of claims 8 to 13, wherein the evaluation information further indicates at least one of the following: the time or storage overhead of the at least one candidate fitting model for fitting the set of historical data, or the moment when the at least one candidate fitting model is selected for determining the storage state of the storage device.

15. A computer program product tangibly stored on a computer-readable medium and comprising machine-executable instructions that, when executed, cause a machine to perform the method according to any one of claims 1 to 7.

Images