Data cache prefetching method and device, storage medium and program product

Abstract

The invention relates to the technical field of cloud computing, and particularly provides a data cache prefetching method and device, a storage medium and a program product. According to the method, file access characteristics representing file access preferences of a user are obtained based on historical access data of the user, a target file meeting a cache prefetching strategy is determined based on the file access characteristics, and the target file is cached and prefetched. By dynamically analyzing the historical access data of the user, the system can adjust the cache content in real time, actively load the data possibly required by the user and ensure that the high-frequency access data is always available, and the mechanism not only improves the data access speed, but also remarkably reduces the influence caused by network delay.

Description

Technical Field

[0001] This disclosure relates to the field of cloud computing technology, and in particular to a data caching prefetching method, device, storage medium, and program product. Background Technology

[0002] Existing cloud desktop technologies primarily rely on transmission protocols and networks for data transfer. Users access data resources located on cloud servers through their terminals. However, network latency and instability can negatively impact the user experience. Current technologies suffer from over-reliance on the network environment. While some caching and prefetching mechanisms exist, most solutions are still based on static rules or simple historical records. Existing technologies lack intelligent dynamic analysis and predictive capabilities. User habits and needs are often diverse and changing, making caching mechanisms reliant on fixed rules ill-suited to adapt to these changes. Summary of the Invention

[0003] This disclosure is made in view of the above-mentioned problems. This disclosure provides a data caching prefetching method, apparatus, storage medium, and program product.

[0004] According to a first aspect of this disclosure, a data caching prefetching method is provided, comprising: Obtain the user's historical access data; Based on the historical access data, the user's file access features are extracted, and the file access features are used to characterize the user's file access preferences. Based on the file access characteristics, target files that meet the preset cache prefetching strategy in the historical access data are identified. The target file is cached and prefetched.

[0005] Furthermore, according to the data caching prefetching method of the first aspect of this disclosure, the historical access data includes first access data and second access data, wherein the first access data is used to reflect the user's historical file access preferences, and the second access data is used to reflect the user's current file access preferences; Based on the historical access data, the user's file access characteristics are extracted, including: Based on the first access data, extract the user's first file access feature; and based on the second access data, extract the user's second file access feature; The first file access feature and the second file access feature are used as the file access feature; Based on the file access characteristics, target files in the historical access data that satisfy a preset cache prefetching strategy are identified, including: Based on the first file access features, a first target file that satisfies a preset cache prefetching strategy in the first access data is determined; and based on the second file access features, a second target file that satisfies a preset cache prefetching strategy in the second access data is determined; the first file access features are features extracted from the first access data, and the second file access features are features extracted from the second access data. The union of the first target file and the second target file is taken as the target file.

[0006] Furthermore, according to the data caching prefetching method of the first aspect of this disclosure, based on the first access data, extracting the user's first file access characteristics includes: Construct a time-series dataset of the first accessed data; The time-series dataset is encoded using a pre-trained first long short-term memory network to capture the temporal features of user behavior; The timing characteristics are used as the first file access characteristics; Based on the first file access characteristics, a first target file in the first access data that satisfies a preset cache prefetching strategy is determined, including: The file access features are decoded using a pre-trained second long short-term memory network to obtain the predicted accessed file; The prefetch probability of the predicted access file is calculated using a normalization function; The N predicted access files with the highest prefetch probability are used as the first target files; The target file is cached and prefetched, including: The first target file and the second target file are prefetched to the cloud and stored in the designated location.

[0007] Furthermore, according to the data cache prefetching method of the first aspect of this disclosure, based on the file access characteristics, determining the target file in the historical access data that satisfies a preset cache prefetching strategy includes: Based on the file access characteristics, calculate the user's access frequency for each file involved in the historical access data; Files whose access frequency is greater than the access frequency threshold are selected as target files that satisfy the cache prefetching strategy. The target file is cached and prefetched, including: The target file is cached locally.

[0008] Furthermore, according to the data caching prefetching method of the first aspect of this disclosure, after performing caching prefetching on the target file, the method further includes: Receive a user's file access request, the file access request carrying identifiers of multiple files to be accessed; Based on the identifier, obtain the cache status and prefetch probability of each of the files to be accessed; The priority of each file to be accessed is calculated based on the cache status and the prefetch probability. The multiple files to be accessed are obtained in descending order of priority.

[0009] Furthermore, according to the data caching prefetching method of the first aspect of this disclosure, after performing caching prefetching on the target file, the method further includes: Obtain the source information of the file to be accessed returned to the user in response to the user's file access request within a set time period. The source information includes local, cloud-defined storage location, or cloud-remaining storage location. The cloud-remaining storage location is a location other than the cloud-defined storage location. Based on the source information, calculate the cache prefetch hit rate; If the hit rate does not meet the preset hit rate requirement, the cache prefetching strategy is optimized.

[0010] According to a second aspect of this disclosure, a data cache prefetching apparatus is provided, comprising: The acquisition module is used to acquire users' historical access data; The extraction module is used to extract the user's file access features based on the historical access data, and the file access features are used to characterize the user's file access preferences; The determination module is used to determine, based on the file access characteristics, the target file in the historical access data that meets the preset cache prefetching strategy; The cache prefetch module is used to perform cache prefetching on the target file.

[0011] According to a third aspect of this disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, the processor executing the computer program to implement the steps of the method described in the first aspect. According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program / instructions stored thereon that, when executed by a processor, implements the steps of the method described in the first aspect. According to a fifth aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0012] As will be described in detail below, the data caching prefetching method according to embodiments of this disclosure obtains file access features characterizing a user's file access preferences based on the user's historical access data, determines target files that meet the caching prefetching strategy based on these file access features, and performs caching prefetching on the target files. By dynamically analyzing the user's historical access data, the system can adjust the cached content in real time, proactively loading data that the user may need, ensuring that frequently accessed data is always available. This mechanism not only improves the speed of data access but also significantly reduces the impact of network latency.

[0013] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further illustration of the claimed technology. Attached Figure Description

[0014] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0015] Figure 1 This is a flowchart illustrating a data caching prefetching method according to an embodiment of the present disclosure.

[0016] Figure 2 This is a flowchart illustrating yet another data caching prefetching method according to an embodiment of the present disclosure.

[0017] Figure 3 This is a structural diagram illustrating a data cache prefetching apparatus according to an embodiment of the present disclosure.

[0018] Figure 4 This is a hardware block diagram illustrating an electronic device according to an embodiment of the present disclosure.

[0019] Figure 5 This is a schematic diagram illustrating a computer program product according to an embodiment of the present disclosure. Detailed Implementation

[0020] The technical methods of the embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings.

[0021] To facilitate understanding of this embodiment, a data cache prefetching method disclosed in this disclosure will first be described in detail. The execution subject of the data cache prefetching method provided in this disclosure is generally an electronic device with a certain computing power, such as a terminal device, a server, or other processing device. In some possible implementations, the data cache prefetching method can be implemented by the processor calling computer-readable instructions stored in memory.

[0022] See Figure 1 The diagram shows a flowchart of a data caching prefetching method provided in this embodiment of the present disclosure. The method includes the following steps: Step 101: Obtain the user's historical access data.

[0023] In one embodiment, historical access data includes first access data for a first historical period. The first historical period is a period of time earlier than the present. For example, historical access data could be a user's access records over the past 7 days. This historical access data can reflect the user's file access preferences over the past 7 days.

[0024] In another embodiment, the historical access data includes second access data for a second historical duration. The second historical duration is a period of time earlier than the present. For example, the historical access data could be a user's access records over the past hour. This historical access data can reflect the user's current file access preferences. Typically, the second historical duration is shorter than the first historical duration.

[0025] In another embodiment, the historical access data can also be a combination of the two mentioned above. That is, the historical access data includes first access data for a first historical duration and second access data for a second historical duration.

[0026] In this embodiment, the historical access data records user access to files, including file type, size, access time, and frequency. For example, if a user frequently accesses "Project Report.docx", the system will mark it as a high-frequency access file.

[0027] Operation type: Analyze user actions on files, such as viewing, editing, and sharing, to form multi-dimensional access characteristics. For example, if a user frequently edits a table, the latest version of that table should be cached first.

[0028] Step 102: Based on the historical access data, extract the user's file access features, which are used to characterize the user's file access preferences.

[0029] In one embodiment, when the historical access data includes first access data and second access data, a first file access feature of the user is extracted based on the first access data; and a second file access feature of the user is extracted based on the second access data; and the first file access feature and the second file access feature are used as file access features.

[0030] Similarly, if the historical access data only includes the first access data or the second access data, the extracted first file access feature or the second file access feature will be used as the file access feature.

[0031] In one embodiment, where historical access data includes first access data, a user's first file access characteristics may include temporal characteristics, spatial characteristics, and / or behavioral characteristics. Temporal characteristics include the user's active time periods, during which the user typically accesses files more frequently. Spatial characteristics include the user's geographical location when accessing the file. Different geographical locations affect file reading speed. For example, file access speed within a company using an intranet is faster than accessing files outside the company using an external network. Behavioral characteristics characterize the files and applications the user frequently accesses. For example, a user might frequently open relevant files during peak project periods.

[0032] In one embodiment, where historical access data includes second access data, the user's second file access characteristics may include time characteristics, location characteristics, and / or task characteristics. The time characteristics characterize the user's access habits within a specific time period. The location characteristics reflect the impact of the user's geographical location on data access speed. The task characteristics reflect the file requirements of the task the user is currently performing.

[0033] In one embodiment, a neural network can be used to extract file access features. Taking the extraction of the first file access feature as an example, extracting the user's first file access feature may include the following steps: Construct a time-series dataset of the first accessed data; A pre-trained first long short-term memory network is used to encode a time-series dataset to capture the temporal features of user behavior; The timing characteristics are used as the first file access characteristics.

[0034] Time-series datasets are used to represent a user's first access data in vector form. A time-series dataset includes each file opened by the user, the time the file was opened, and the type of operation performed on the file.

[0035] In prefetching scenarios, extracting file access features through neural networks helps predict the next file a user might access. After determining the next file a user might access, the file is stored in a designated cloud storage location. This way, when a user requests access to this file, the file can be quickly retrieved from the designated cloud storage location, improving access efficiency.

[0036] Step 103: Based on the file access characteristics, determine the target files in the historical access data that meet the preset cache prefetching strategy.

[0037] In one embodiment, in a prefetching scenario, taking the example of file access features including a first file access feature and the target file including a first target file, step 103 may include the following steps: The file access features are decoded using a pre-trained second long short-term memory network to obtain the predicted access file; The prefetch probability of the predicted access file is calculated using a normalization function; The N predicted access files with the highest prefetch probability are selected as the first target files.

[0038] In another embodiment, in a caching scenario, step 103 may include the following steps: Based on file access characteristics, calculate the frequency of user access to each file involved in historical access data; Files accessed more frequently than the access frequency threshold are selected as target files to satisfy the cache prefetching strategy.

[0039] The formula for calculating access frequency is: F(u, f ) = N(u, f )} / T(u); Where F(u, f) represents the frequency of user u accessing file f, N(u, f) is the number of accesses, and T(u) is the observation period.

[0040] For example, if a user accesses "Project Plan.xlsx" ten times in the past week, then the access frequency of this file is 10 times / 7 days.

[0041] Based on access frequency, the system uses a threshold mechanism to decide whether to cache files, for example: If F(u, f) > θ, cache file f; Otherwise, do not cache file f.

[0042] In this formula, if a file's access frequency exceeds a set threshold θ, the system will preferentially cache it in local storage. When network latency occurs, access operations can be performed directly by accessing local data, without being affected by network delays. For example, if a user accesses the file more than 5 times per day, the file will be cached. Using this embodiment to determine cached files in caching scenarios can alleviate the problem of low access efficiency caused by network latency.

[0043] In one example, suppose a user opens "Financial Statements.xlsx" every day at work and edits and views the file multiple times throughout the week. The system recognizes this high access frequency and automatically caches the file locally when the user enters the work environment, ensuring quick access without having to download it from the cloud each time, greatly improving the user experience.

[0044] In another embodiment, in the prefetching scenario, the file access features may include a first file access feature and a second file access feature. When determining the target file, the first target file obtained based on the first file access feature and the second target file obtained based on the second file access feature can be combined. Both the first and second target files can be implemented using the second long short-term memory network described in the previous embodiment. That is, the pre-trained second long short-term memory network decodes the first and second file features respectively to obtain the first predicted access file corresponding to the first file feature and the second predicted access file corresponding to the second file feature. The prefetch probability of the first and second predicted access files is calculated using a normalization function. The N files with the highest prefetch probability in the first predicted access file are selected as the first target files, and the M files with the highest prefetch probability in the second predicted access file are selected as the second target files.

[0045] Step 104: Perform cache prefetching on the target file.

[0046] In caching scenarios, the target file is cached locally. In prefetching scenarios, the target file is prefetched to a specified storage location in the cloud.

[0047] In this embodiment, both locally cached files and files pre-fetched to a designated cloud storage location can be scheduled for periodic cleanup, which helps reduce the occupation of local storage resources and lower response latency.

[0048] In one embodiment, the method may further include the following steps: Receive file access requests from users, which carry identifiers of multiple files to be accessed. The cache status and prefetch probability of each file to be accessed are obtained based on the identifier. The priority of each file to be accessed is calculated based on cache status and prefetch probability. Retrieve multiple files to be accessed in descending order of priority.

[0049] It should be understood that cloud storage contains identifiers for cached and prefetched files. Therefore, upon receiving a user's file access request, querying using the identifier of the file to be accessed carried in the file access request can obtain the cache status and prefetch probability of each file to be accessed.

[0050] The cache status includes cached, cached, and not cached. Cached means the file has been cached locally, cached means the file is being stored locally, and not cached means the file has not been cached. These three statuses can be represented by different values; for example, cached is represented by 1, cached by 0.5, and not cached by 0.

[0051] In this embodiment, a weighted summation method can be used to calculate the priority based on the cache state and prefetch probability. The constant parameters in the weighted summation can be pre-configured. It should be understood that a larger weighted summation result indicates a higher priority.

[0052] In this embodiment, when retrieving each file to be accessed, the system prioritizes searching for the file locally. If the file is not stored locally, it searches for the file in the cloud storage location. This ensures that the user's request can be met in a timely manner, thereby reducing latency.

[0053] In one embodiment, a feedback mechanism can also be set to evaluate the effectiveness of caching and prefetching, and continuously optimize the strategy. Specifically, the method may further include the following steps: Get the source information of the file to be accessed returned to the user in response to the user's file access request within a set time. The source information includes local, cloud-set storage location or cloud remaining storage location. The cloud remaining storage location is the location other than the cloud-set storage location. Calculate the cache prefetch hit rate based on the source information; If the hit rate does not meet the preset hit rate requirements, the cache prefetching strategy is optimized.

[0054] In this embodiment, the cache hit rate and the prefetch hit rate can be calculated based on the source information.

[0055] The cache hit rate can be calculated using the following formula: HitRate 1= Hits1 / TotalRequests; Where HitRate 1 is the cache hit rate, Hits1 is the number of times the file to be accessed is returned from the local machine to the user in response to the user's file access request within a set time, and TotalRequests is the total number of times the file access request is responded to to the user within a set time.

[0056] The prefetch hit rate can be calculated using the following formula: HitRate2= Hits2 / TotalRequests; Where HitRate2 is the prefetch hit rate, Hits2 is the number of times the file to be accessed is returned from the cloud storage location to the user in response to the user's file access request within a set time, and TotalRequests is the total number of times the file access request to the user is responded to within a set time.

[0057] By monitoring the hit rate, the system can determine the effectiveness of the current strategy. Based on the hit rate, the parameters of the caching and prefetching strategies are adjusted to ensure the system remains efficient under constantly changing user behavior. For example, if the cache hit rate is low, it indicates that there are not enough files cached locally; in this case, the access frequency threshold in the cache prefetching strategy can be appropriately reduced. Similarly, if the prefetch hit rate is low, it indicates that there are few files cached in the cloud storage location; in this case, the value of N can be reduced, where N is the number of predicted access files with the highest prefetch probability when selecting target files according to the prefetch probability.

[0058] In the solution provided in this embodiment, file access characteristics representing a user's file access preferences are obtained based on the user's historical access data. Based on these file access characteristics, target files that meet the cache prefetching strategy are determined, and these target files are prefetched into the cache. By dynamically analyzing the user's historical access data, the system can adjust the cached content in real time, proactively loading data that the user might need, ensuring that frequently accessed data is always available. This mechanism not only improves the speed of data access but also significantly reduces the impact of network latency.

[0059] In addition, advanced machine learning techniques are employed to build user behavior analysis and prediction models. By collecting users' historical access data, the system can identify users' usage habits and extract valuable feature information.

[0060] Furthermore, by monitoring user feedback, the system can evaluate the effectiveness of the current caching and prefetching strategies and adjust algorithm parameters in a timely manner. This mechanism ensures that the system can continuously learn and improve itself in actual use.

[0061] Finally, it can reduce unnecessary repeated data transmissions, reduce bandwidth pressure, and improve the overall system performance.

[0062] This disclosure also provides a data method, such as... Figure 2 As shown, it may include: Step 201, Intelligent caching mechanism.

[0063] The goal of the intelligent caching mechanism is to improve data access efficiency on cloud desktops by dynamically adjusting cached content. This mechanism includes a user behavior analysis module and a dynamic cache update strategy.

[0064] The user behavior analysis module collects users' historical access data to build user profiles, enabling a better understanding of user needs and habits. This module is used for data collection and feature extraction.

[0065] Data collection is the foundation of intelligent caching mechanisms, and includes the following aspects: Historical access data: Records user access to files, including file type, size, access time, operation type, and frequency. For example, if a user frequently accesses "Project Report.docx", the system will mark it as a high-frequency access file.

[0066] Operation type: Analyze user actions on files, such as viewing, editing, and sharing, to form multi-dimensional access characteristics. For example, if a user frequently edits a table, the latest version of that table should be cached first.

[0067] Feature extraction aims to extract temporal, spatial, and / or behavioral features that are important for caching decisions from historical access data.

[0068] Time characteristics: such as the user's active time period, there is usually a higher access frequency during certain time periods.

[0069] Spatial characteristics: The user's geographical location may affect the speed of data access. For example, accessing files is faster when the user is inside a company.

[0070] Behavioral characteristics: By analyzing user behavior patterns, we can identify frequently used files and applications. For example, a user might frequently open relevant files during peak project periods.

[0071] The dynamic cache update strategy is implemented based on the frequency access model.

[0072] The access frequency model is as follows: F(u, f ) = N(u, f )} / T(u); Where F(u, f) represents the frequency of user u accessing file f, N(u, f) is the number of accesses, and T(u) is the observation period.

[0073] For example, if a user accesses "Project Plan.xlsx" ten times in the past week, then the access frequency of this file is 10 times / 7 days. The system will prioritize caching it in the local storage space. When there is a network delay, the user can directly access the local data to perform the operation, and the operation data source will not be affected by the network delay.

[0074] Based on access frequency, the system uses a threshold mechanism to decide whether to cache a file.

[0075] If F(u, f) > θ, cache the file; Otherwise, do not cache the file.

[0076] In this formula, if a file is accessed more frequently than a set threshold θ, it will be cached. For example, if a user accesses the file more than 5 times per day, the file will be cached.

[0077] Example Application Suppose a user opens "Financial Statements.xlsx" every day during work hours and edits and views the file multiple times throughout the week. Through user behavior analysis, the system identifies this high access frequency and automatically caches the file locally when the user enters the work environment. This ensures quick access without requiring a download from the cloud each time, significantly improving the user experience.

[0078] Step 202: File prefetching.

[0079] First, the system needs to prepare users' historical access data and build it into a time-series dataset. For example, it can record users' behavior over the past week, including the files opened each time, the time, and the type of operation.

[0080] Long Short-Term Memory (LSTM) networks are used to process time-series data and capture the temporal characteristics of user behavior.

[0081] The softmax function is used to calculate the probability that a user will access a file.

[0082] The system can determine the probability of each file being accessed and select the file with the highest probability for prefetching.

[0083] Step 203: Coordinate working mechanisms.

[0084] By combining the results of caching and prefetching, the data access process can be optimized.

[0085] When processing user requests, the system calculates the priority of each requested file. Based on the priority, the system will prioritize serving files in the cache. If the cache is not hit, prefetching will be performed. This mechanism ensures that user requests are satisfied in a timely manner, thereby reducing latency.

[0086] Step 204, Feedback Mechanism.

[0087] By collecting cache and prefetch hit rates through user feedback, the system adjusts the parameters of caching and prefetching strategies based on the hit rates, ensuring that the system remains efficient amidst constantly changing user behavior. For example, if a file has a low hit rate, the system will re-evaluate the caching and prefetching strategy for that file.

[0088] This disclosure also provides a data cache prefetching apparatus for performing the data cache prefetching method provided in any of the above embodiments. Figure 3 As shown, the device includes: Module 31 is used to obtain the user's historical access data; Extraction module 32 is used to extract the user's file access features based on the historical access data, the file access features being used to characterize the user's file access preferences; The determination module 33 is used to determine the target file in the historical access data that meets the preset cache prefetching strategy based on the file access characteristics. The cache prefetch module 34 is used to perform cache prefetching on the target file.

[0089] In an optional embodiment, the historical access data includes first access data and second access data, wherein the first access data is used to reflect the user's historical file access preferences, and the second access data is used to reflect the user's current file access preferences; Extraction module 32 is used for: Based on the first access data, extract the user's first file access feature; and based on the second access data, extract the user's second file access feature; The first file access feature and the second file access feature are used as the file access feature; Module 33 is used for: Based on the first file access features, a first target file that satisfies a preset cache prefetching strategy in the first access data is determined; and based on the second file access features, a second target file that satisfies a preset cache prefetching strategy in the second access data is determined; the first file access features are features extracted from the first access data, and the second file access features are features extracted from the second access data. The union of the first target file and the second target file is taken as the target file.

[0090] In an optional embodiment, the extraction module 32 is used for: Construct a time-series dataset of the first accessed data; The time-series dataset is encoded using a pre-trained first long short-term memory network to capture the temporal features of user behavior; The timing characteristics are used as the first file access characteristics; Based on the first file access characteristics, a first target file in the first access data that satisfies a preset cache prefetching strategy is determined, including: The file access features are decoded using a pre-trained second long short-term memory network to obtain the predicted accessed file; The prefetch probability of the predicted access file is calculated using a normalization function; The N predicted access files with the highest prefetch probability are used as the first target files; The target file is cached and prefetched, including: The first target file and the second target file are prefetched to the cloud and stored in the designated location.

[0091] In an optional embodiment, the determining module 33 is used to: Based on the file access characteristics, calculate the user's access frequency for each file involved in the historical access data; Files whose access frequency is greater than the access frequency threshold are selected as target files that satisfy the cache prefetching strategy. The target file is cached and prefetched, including: The target file is cached locally.

[0092] In an optional embodiment, the device is also used for: After caching and prefetching the target file, the user's file access request is received, and the file access request carries the identifiers of multiple files to be accessed; Based on the identifier, obtain the cache status and prefetch probability of each of the files to be accessed; The priority of each file to be accessed is calculated based on the cache status and the prefetch probability. The multiple files to be accessed are obtained in descending order of priority.

[0093] In an optional embodiment, the device is also used for: After caching and prefetching the target file, obtain the source information of the file to be accessed returned to the user in response to the user's file access request within a set time. The source information includes local, cloud-defined storage location, or cloud-remaining storage location. The cloud-remaining storage location is a location other than the cloud-defined storage location. Based on the source information, calculate the cache prefetch hit rate; If the hit rate does not meet the preset hit rate requirement, the cache prefetching strategy is optimized.

[0094] The data cache prefetching device and the data cache prefetching method provided in this disclosure are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.

[0095] This disclosure also provides an electronic device for performing the above-described data cache prefetching method. Please refer to... Figure 4 It illustrates a schematic diagram of an electronic device provided by some embodiments of this disclosure. For example... Figure 4 As shown, the electronic device 4 includes: a processor 400, a memory 401, a bus 402, and a communication interface 403. The processor 400, the communication interface 403, and the memory 401 are connected via the bus 402. The memory 401 stores a computer program that can run on the processor 400. When the processor 400 runs the computer program, it executes the data cache prefetching method provided in any of the foregoing embodiments of this disclosure.

[0096] The memory 401 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this device network element and at least one other network element is achieved through at least one communication interface 403 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0097] Bus 402 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 401 is used to store programs. After receiving an execution instruction, the processor 400 executes the program. The data cache prefetching method disclosed in any of the foregoing embodiments of this disclosure can be applied to the processor 400, or implemented by the processor 400.

[0098] The processor 400 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 400 or by instructions in software form. The processor 400 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this disclosure can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 401. The processor 400 reads the information in memory 401 and, in conjunction with its hardware, completes the steps of the above method.

[0099] The electronic device provided in this disclosure and the data cache prefetching method provided in this disclosure are based on the same inventive concept and have the same beneficial effects as the methods they employ, operate, or implement.

[0100] This disclosure also provides a computer-readable storage medium corresponding to the data cache prefetching method provided in the foregoing embodiments. The computer-readable storage medium is an optical disc, on which a computer program (i.e., a computer program product) is stored. When the computer program is run by a processor, it executes the data cache prefetching method provided in any of the foregoing embodiments.

[0101] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0102] The computer-readable storage medium provided in the above embodiments of this disclosure and the data caching prefetching method provided in the embodiments of this disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the applications stored therein.

[0103] This disclosure also provides a computer program product; please refer to [reference needed]. Figure 5 The computer program product 500 carries program code, namely computer program 501. The instructions included in the computer program 501 can be used to execute the steps of the data cache prefetching method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here.

[0104] The aforementioned computer program product can be implemented through hardware, software, or a combination thereof. In one optional embodiment, the computer program product is specifically embodied in a computer storage medium; in another optional embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0105] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0106] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0107] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0108] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0109] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0110] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0111] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A data cache prefetching method, characterized by, The method comprises: obtaining historical access data of a user; extracting file access features of the user based on the historical access data, the file access features being used to represent file access preferences of the user; determining target files in the historical access data that meet a preset cache pre-fetching strategy based on the file access features; and pre-fetching the target files.

2. The method of claim 1, wherein, The historical access data comprises first access data and second access data, the first access data being used to reflect historical file access preferences of the user, and the second access data being used to reflect current file access preferences of the user; extracting file access features of the user based on the historical access data comprises: extracting first file access features of the user based on the first access data, and extracting second file access features of the user based on the second access data; the first file access features and the second file access features are used as the file access features; determining target files in the historical access data that meet a preset cache pre-fetching strategy based on the file access features comprises: determining first target files in the first access data that meet a preset cache pre-fetching strategy based on the first file access features, and determining second target files in the second access data that meet a preset cache pre-fetching strategy based on the second file access features; the first file access features are features extracted from the first access data, and the second file access features are features extracted from the second access data; a union of the first target files and the second target files is used as the target files.

3. The method of claim 2, wherein, extracting first file access features of the user based on the first access data comprises: constructing a time series data set of the first access data; encoding the time series data set by using a pre-trained first long short-term memory network to capture time sequence features of user behaviors; the time sequence features are used as the first file access features; determining first target files in the first access data that meet a preset cache pre-fetching strategy based on the first file access features comprises: decoding the file access features by using a pre-trained second long short-term memory network to obtain predicted access files; calculating pre-fetching probabilities of the predicted access files by using a normalization function; the N predicted access files with the highest pre-fetching probabilities are used as the first target files; pre-fetching the target files comprises: pre-fetching the first target files and the second target files to a cloud storage location.

4. The method of claim 1, wherein, determining target files in the historical access data that meet a preset cache pre-fetching strategy based on the file access features comprises: calculating access frequencies of the user for each file involved in the historical access data based on the file access features; files with access frequencies greater than an access frequency threshold are used as target files that meet the cache pre-fetching strategy; pre-fetching the target files comprises: caching the target files to a local device.

5. The method according to any one of claims 1 to 4, characterized in that, after pre-fetching the target files, the method further comprises: Receiving a file access request of a user, the file access request carrying identifications of a plurality of files to be accessed; Obtaining cache states and pre-fetch probabilities of the files to be accessed based on the identifications; Calculating priorities of the files to be accessed based on the cache states and the pre-fetch probabilities; Obtaining the plurality of files to be accessed in an order from high to low of the priorities.

6. The method according to any one of claims 1 to 4, characterized in that, After the cache pre-fetching of the target file, the method further comprises: Obtaining source information of the files to be accessed returned to the user within a set time in response to the file access request of the user, the source information comprising a local, a cloud set storage location or a cloud remaining storage location, the cloud remaining storage location being a location other than the cloud set storage location; Calculating a hit rate of the cache pre-fetching based on the source information; Optimizing the cache pre-fetching strategy in a case where the hit rate does not meet a preset hit rate requirement.

7. A data cache prefetch apparatus, comprising: The method comprises: Obtaining a historical access data of a user; Extracting a file access feature of the user based on the historical access data, the file access feature being used to represent a file access preference of the user; Determining a target file in the historical access data that meets a preset cache pre-fetching strategy based on the file access feature; Cache pre-fetching the target file.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program, when executed by the processor, is arranged to perform the method of any one of claims 1 to 7. The processor executes the computer program to implement the steps of the method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the method of any one of claims 1-6.

10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the method of any one of claims 1-6.

Images