Cloud storage data access method and device

Abstract

The invention discloses a cloud storage data access method and device, and is applied to the technical field of cloud services, the method comprises the following steps: obtaining historical behavior information of a target terminal, the historical behavior information being a parameter associated with a cloud storage data access behavior of the target terminal in a historical time period; target behavior information and a plurality of pieces of target access data of the target terminal are obtained according to the historical behavior information, and the target behavior information is a parameter associated with a cloud storage data access behavior of the target terminal in a future time period; determining multiple pieces of target preloading data according to the target behavior information and the multiple pieces of target access data; carrying out preloading processing on each piece of target preloading data by utilizing the target behavior information to obtain each piece of preloaded target preloading data; and in response to an access instruction of the target terminal, performing access processing on each piece of pre-loaded target pre-loaded data. According to the invention, the data delay of the cloud storage service can be effectively reduced, and the user experience of the cloud storage service is improved.

Description

Technical Field

[0001] The present application relates to the field of cloud service technology, and in particular to a method and device for accessing cloud storage data. Background Art

[0002] Cloud storage service is a technology that stores data on remote servers via the Internet. By storing data on remote servers, it provides users with a cost-effective, flexible and scalable storage solution. At present, although cloud storage services provide users with convenient data storage and access methods, the data latency problem of cloud storage services is still relatively serious, which reduces the user experience of cloud storage services. Summary of the invention

[0003] The embodiments of the present application provide a cloud storage data access method and device for reducing the data latency of the cloud storage service and improving the user experience of the cloud storage service.

[0004] On the one hand, an embodiment of the present application provides a cloud storage data access method, comprising the following steps: Acquire historical behavior information of the target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; According to the historical behavior information, target behavior information and a plurality of target access data of the target terminal are obtained; wherein the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; Determining a plurality of target preload data according to the target behavior information and a plurality of the target access data; Performing preloading processing on each of the target preloaded data using the target behavior information to obtain each of the preloaded target preloaded data; In response to the access instruction of the target terminal, access processing is performed on the preloaded target preload data.

[0005] On the other hand, an embodiment of the present application provides a cloud storage data access device, including: An acquisition module, used to acquire historical behavior information of a target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; A first processing module, configured to obtain target behavior information and a plurality of target access data of the target terminal according to the historical behavior information; wherein the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; A second processing module, configured to determine a plurality of target preload data according to the target behavior information and a plurality of the target access data; A third processing module, configured to perform preloading processing on each of the target preloaded data using the target behavior information to obtain each of the preloaded target preloaded data; The fourth processing module is used to perform access processing on each preloaded target preloaded data in response to the access instruction of the target terminal.

[0006] According to the cloud storage data access method and device of the embodiment of the present application, first, historical behavior information of the target terminal is obtained, and the historical behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the historical time period; secondly, based on the historical behavior information, target behavior information and a number of target access data of the target terminal are obtained, and the target behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the future time period; then, based on the target behavior information and the number of target access data, a plurality of target preloaded data are determined; thereafter, each target preloaded data is preloaded using the target behavior information to obtain each preloaded target preloaded data; finally, in response to the access instruction of the target terminal, each preloaded target preloaded data is accessed, which can effectively reduce the data delay of the cloud storage service, speed up the response to the data access request of the cloud storage service, and thus improve the user experience of the cloud storage service.

[0007] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a flow chart of the cloud storage data access method provided by this application; Figure 2 is a flowchart provided by the present application for predicting the future access needs of a target terminal; Figure 3 is a flow chart of the preloading process provided by this application; Figure 4 is a schematic diagram of determining preloading parameters provided by the present application; Figure 5 is another flow chart of the preloading process provided by the present application; Figure 6 It is a schematic diagram of the relationship between data, queues, media and terminals provided by this application; Figure 7 is a flowchart of preloading optimization provided by this application; Figure 8is a structural diagram of a cloud storage data access device provided by the present application; Fig. 9 It is a schematic diagram of the cloud storage data access method provided by this application; Fig.10 This is another flow chart of the cloud storage data access method provided by this application. DETAILED DESCRIPTION

[0009] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0010] The present application is further described below in conjunction with the accompanying drawings and specific embodiments. The described embodiments should not be regarded as limiting the present application, and all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0011] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0012] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0013] With the development of Internet Technology (IT) and the significant decline in storage and computing costs, cloud computing has gradually become popular and has become a mainstream Internet technology service model. The core concept of cloud computing is to provide computing resources such as servers, storage, networks, and applications as a service to users through the Internet. Users can dynamically adjust computing resources according to actual needs without pre-configuring hardware devices. This model not only reduces users' Internet service costs, but also improves the utilization and flexibility of computing resources.

[0014] Among the many services of cloud computing, cloud storage services play a vital role. With the acceleration of digital transformation, the amount of data is growing explosively, and users' demand for storage is increasing. Cloud storage services provide users with a cost-effective, flexible and scalable storage solution by storing data on remote servers managed by cloud service providers. Cloud storage services can provide reliable security, data redundancy and disaster recovery capabilities to ensure the continuous availability and security of data.

[0015] The user groups of cloud storage services are quite broad, including individual users, small and medium-sized enterprises, large enterprises and government agencies, etc. The needs and data access behaviors of different users vary greatly. Data access behavior refers to the behavioral patterns of users when accessing resources such as computer systems, network services, and databases. It usually includes data such as the time, frequency, target and operation of data access, which can reveal the data access habits of users. In actual applications, individual users tend to pay more attention to the size, price and data synchronization capabilities of storage space, while enterprise users tend to have higher requirements for data security, access speed and application programming interface (API) integration. In addition, even within the same user group, there are obvious differences in the access behaviors of different objects. For example, for the enterprise user group, some departments may need to frequently access financial statements, while other departments may pay more attention to product design documents.

[0016] Affected by the complexity of data access behavior, data access is non-uniform and volatile. Specifically, first, data access is not evenly distributed, that is, data access is non-uniform, and there are often obvious hot data and access peaks. For example, in the morning and afternoon of weekdays, the access needs of corporate users will increase significantly, while on holidays and weekends, individual users will increase their access to multimedia content such as photos and videos. Secondly, data access behavior usually changes dynamically with time or specific events, that is, data access is volatile. For example, specific events such as corporate restructuring, new business launches, and holiday promotions are prone to changes in data access behavior.

[0017] As users' expectations for data access speed increase, data latency has become a key factor affecting the user experience of cloud storage services, and the non-uniformity and volatility of data access will have a certain negative impact on the data latency of cloud storage services. Currently, cloud storage services still have a serious data latency problem, which reduces the user experience of cloud storage services, especially in scenarios with high concurrent access or large data transmission.

[0018] In view of this, the embodiments of the present application provide a cloud storage data access method and device, which aims to predict the data access behavior of the target terminal in the future time period through intelligent prediction technology, and then load the data that the target terminal needs to access in the future time period in advance into storage media such as cache servers and edge nodes based on the prediction results, thereby reducing the data latency of the cloud storage service, improving the response speed of data access, and providing users with a smoother and faster data access experience.

[0019] First, the implementation steps of the cloud storage data access method provided by the embodiment of the present application will be described in detail below with reference to the accompanying drawings.

[0020] The cloud storage data access method provided in the embodiment of the present application can be applied to a terminal, a server, or software running in a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. In addition, the server can also be a node server in a blockchain network, but is not limited to this. Among them, blockchain is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.

[0021] Reference Figure 1 , Figure 1 is a flow chart of a cloud storage data access method provided by the present application, and the cloud storage data access method may include the following steps S101-S105: S101, obtaining historical behavior information of a target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; S102, obtaining target behavior information and a plurality of target access data of a target terminal according to the historical behavior information; wherein the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; S103, determining a plurality of target preload data according to the target behavior information and a plurality of target access data; S104, performing preloading processing on each target preloading data using the target behavior information to obtain each preloaded target preloading data; S105 , in response to the access instruction of the target terminal, performing access processing on each preloaded target preload data.

[0022] In an embodiment of the present application, first, historical behavior information of the target terminal is obtained, and the historical behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the historical time period; secondly, based on the historical behavior information, target behavior information and a number of target access data of the target terminal are obtained, and the target behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the future time period, and the target access data refers to data objects that the target terminal expects to access in the future time period; then, based on the target behavior information and the number of target access data, a plurality of target preloaded data are determined, and the target preloaded data refers to preloaded data objects; thereafter, each target preloaded data is preloaded using the target behavior information to obtain each preloaded target preloaded data; finally, in response to an access instruction of the target terminal, each preloaded target preloaded data is accessed.

[0023] It can be seen that the embodiment of the present application determines the target behavior information and several target access data of the target terminal through the historical behavior information of the target terminal, and then uses the target behavior information and several target access data to determine multiple target preloaded data, and before the data access request of the target terminal arrives, uses the target behavior information to implement preloading processing of each target preloaded data, and finally when the data access request of the target terminal arrives, each preloaded target preloaded data is accessed and processed, thereby realizing data access of the cloud storage service. In this way, the data delay of the cloud storage service can be effectively reduced, especially the data delay in scenarios with high concurrent access or large data volume transmission, and the response to the data access request of the cloud storage service can be accelerated, thereby improving the user experience of the cloud storage service.

[0024] In the above step S101, parameters associated with the cloud storage data access behavior of the target terminal in the historical time period are obtained as historical behavior information, so that in subsequent steps, the access behavior and access requirements of the target terminal in the future time period can be accurately captured through the historical behavior information.

[0025] The above-mentioned target terminal refers to a terminal suitable for the cloud storage data access method in the embodiment of the present application. The type of the terminal can be flexibly set according to actual conditions. For example, the terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this.

[0026] The above-mentioned historical behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the historical time period, wherein the cloud storage data access behavior refers to the behavioral patterns of users when accessing cloud storage service resources.

[0027] The above historical behavior information can be set according to actual conditions, and the embodiments of the present application do not specifically limit this.

[0028] Exemplarily, in some embodiments, the historical behavior information may include the behavior characteristic data, time characteristic data, user characteristic data and context characteristic data of the target terminal in the historical time period. Specifically, the behavior characteristic data of the target terminal may include the target terminal's access record, access data type, access data parameter, access operation type, click stream data and search query times, etc., wherein the access operation type may include but is not limited to read operation, download operation and upload operation, etc.; click stream data is used to indicate the access track of the target terminal when performing data access, which may be a Pageviews table or a Visits table, etc., but is not limited thereto. The time characteristic data of the target terminal may include the data access time value, data access date and data access holiday of the target terminal, etc., and these parameters may be used to capture the periodic changes of the data access behavior of the target terminal. The user characteristic data of the target terminal may include the user information using the target terminal, such as the user's occupation, gender and preference information, etc. The context characteristic data of the target terminal may include the device type of the target terminal, the browser version number, etc., and these parameters can help understand the software and hardware environment of the target terminal when performing data access.

[0029] The above-mentioned method of obtaining the historical behavior information can be set according to the actual situation, and the embodiments of the present application do not specifically limit this. For example, the historical behavior information of the target terminal is obtained by analyzing the log file of the target terminal in the historical time period; or the historical behavior information of the target terminal is directly obtained through a database, and the database is used to store the historical behavior information of several terminals, but is not limited to this.

[0030] In the above step S102, after obtaining the historical behavior information, the historical behavior information and intelligent prediction technology are used to obtain target behavior information and a number of target access data, so as to accurately capture the access behavior and access needs of the target terminal in the future time period.

[0031] The above-mentioned target access data refers to data objects that the target terminal may expect to access in a future period of time.

[0032] The above-mentioned target behavior information refers to parameters associated with the cloud storage data access behavior of the target terminal in the future time period, wherein the cloud storage data access behavior refers to the behavioral patterns of users when accessing cloud storage service resources.

[0033] The above target behavior information can be set according to actual conditions, and the embodiments of the present application do not specifically limit this.

[0034] Exemplarily, in some embodiments, the target behavior information may include the access frequency of each target access data, the access mode of each target access data, the data type of each target access data, the data access peak time period of the target terminal, the potential demand data of the target terminal, and the loss risk data of the target terminal, etc. Among them, the access frequency of the target access data refers to the difference between the access volume of the target access data in the future time period and the duration of the future time period; the access mode of the target access data may include the access path of the target access data, etc.; the data type of the target access data can be flexibly set according to the actual situation, for example, the data type of the target access data can be an image type, a text type, a video type, etc., but not limited thereto; the data access peak time period of the target terminal refers to the sub-time period with the highest data access volume in the future time period, and the unit of the sub-time period is hours; the potential demand data of the target terminal refers to the product type or service type expected by the target terminal in the future time period, which can indicate the products or services that the target terminal may be interested in in the future time period; the loss risk data of the target terminal refers to the probability that the target terminal does not use the cloud storage service in the future time period, which can indicate the target terminal that may be lost.

[0035] The above-mentioned obtaining the target behavior information and several target access data of the target terminal based on the historical behavior information may include obtaining the target behavior information and several target access data of the target terminal based on the historical behavior information in combination with a machine learning method, but is not limited to this.

[0036] The above-mentioned machine learning method can be set according to actual conditions, and the embodiments of the present application do not make any specific limitations on this.

[0037] For example, the above-mentioned machine learning method may be logistic regression; or, the above-mentioned machine learning method may be a support vector machine, but is not limited thereto.

[0038] In the above step S103, the target behavior information and the plurality of target access data are processed in order to determine a plurality of target preloaded data from the plurality of target access data.

[0039] The above-mentioned target preloaded data refers to the data objects involved in the preload process, which can also be understood as the data objects that the target terminal expects to access in the future time period.

[0040] The above-mentioned determining multiple target preloading data based on the target behavior information and several target access data may include obtaining the access probability of several target access data based on the target behavior information and several target access data in combination with a machine learning method, where the access probability refers to the probability that the target access data will be accessed in a future time period, and then selecting the target access data with an access probability greater than a preset probability threshold as the target preloading data, but is not limited to this.

[0041] In the above step S104, after determining multiple target preloaded data, each target preloaded data is preloaded using the target behavior information, aiming to load each target preloaded data in the cloud storage service to a storage medium such as a cache server, edge node, etc. in advance before the data access request from the target terminal arrives.

[0042] The above-mentioned preloading process refers to loading data in advance to storage media such as cache servers, edge nodes, etc.

[0043] The preloaded target preloaded data refers to the target preloaded data that is preloaded to a storage medium such as a cache server, an edge node, etc.

[0044] The above-mentioned preloading processing of each target preload data using target behavior information can include determining the preloading time based on the target behavior information in combination with a machine learning method, and loading each target preload data to a preset edge node when the preloading time of the target terminal arrives, but is not limited to this.

[0045] In the above step S105, before the data access request of the target terminal arrives, the target preloaded data of the target terminal has been preloaded, so when the access instruction of the target terminal is received, it means that the data access request of the target terminal has arrived. At this time, in response to the access instruction of the target terminal, each preloaded target preloaded data is obtained, and each preloaded target preloaded data is sent to the target terminal. It can be understood that after obtaining each preloaded target preloaded data, the target terminal can perform operations such as decryption, decompression, etc. on each preloaded target preloaded data and render and display it.

[0046] The access instruction of the target terminal is used to indicate the data access request of the target terminal.

[0047] The above steps will be further described below.

[0048] In some embodiments, reference Figure 2 In the above step S102, the specific implementation process of obtaining the target behavior information of the target terminal and several target access data according to the historical behavior information may include the following steps S1021-S1023: S1021, preprocessing the historical behavior information to obtain preprocessed historical behavior information; S1022, extracting features from the preprocessed historical behavior information to obtain historical behavior features; S1023, predicting historical behavior characteristics to obtain target behavior information and a number of target access data.

[0049] In this embodiment, first, the historical behavior information is preprocessed to effectively ensure the accuracy of the historical behavior information and reduce the interference of redundant and useless data on intelligent prediction, thereby obtaining the preprocessed historical behavior information; then, the preprocessed historical behavior information is feature extracted to capture the feature information associated with the cloud storage data access behavior of the target terminal in the historical time period from various types of historical behavior information, thereby obtaining the historical behavior features; then, the historical behavior features are input into the pre-trained prediction model, and the access behavior and access requirements of the target terminal in the future time period are accurately captured through the prediction model, thereby obtaining the target behavior information and several target access data. In this way, this embodiment can effectively and accurately capture the access behavior and access requirements of the target terminal in the future time period, so that data preloading can be achieved in the subsequent steps through the access behavior and access requirements of the target terminal in the future time period.

[0050] The above historical behavior information may include behavior characteristic data, time characteristic data, user characteristic data and context characteristic data of the target terminal within a historical time period, but is not limited thereto.

[0051] The implementation of the above preprocessing can be set according to actual conditions, and this embodiment does not specifically limit this. For example, preprocessing can include data cleaning such as missing value processing, outlier processing and duplicate value processing, data conversion such as normalization, discretization and mathematical transformation, etc., but is not limited to this.

[0052] The implementation method of the above feature extraction can be set according to actual conditions, and this embodiment does not make any specific limitation to this.

[0053] Exemplarily, in some embodiments, the above historical behavior features include behavior features, time features, user features, and context features, and the feature extraction method is: feature extraction is performed on the preprocessed behavior feature data, preprocessed time feature data, preprocessed user feature data, and preprocessed context feature data, respectively, to obtain behavior features, time features, user features, and context features. Feature extraction can be achieved through natural language processing technologies such as bag of words model, word embedding, etc., or can be achieved through other methods such as convolutional neural network, one-hot encoding, etc., but is not limited thereto.

[0054] Exemplarily, in some embodiments, historical behavior features include behavior time series feature sequences, user features, and context features. The feature extraction method is: first, the preprocessed behavior feature data and the preprocessed time characteristic data are integrated into behavior time series data, and feature extraction is performed on the behavior time series data to obtain a behavior time series feature sequence, and then feature extraction is performed on the preprocessed user characteristic data to obtain user features, and then feature extraction is performed on the preprocessed context characteristic data to obtain context features. Among them, feature extraction can be achieved through natural language processing technologies such as bag of words model, word embedding, etc., and can also be achieved through other methods such as convolutional neural network, one-hot encoding, etc., but is not limited to this.

[0055] The target behavior information may include the access frequency of each target access data, the access mode of each target access data, the data type of each target access data and the data access peak period of the target terminal, but is not limited thereto.

[0056] The implementation method of the above prediction can be set according to actual conditions, and this embodiment does not make any specific limitation to this.

[0057] Exemplarily, in some embodiments, the historical behavior features may include behavior features, time features, user features, and context features. The prediction method is: integrating the behavior features, time features, user features, and context features into a feature sequence and inputting it into a first prediction model, performing prediction processing through the first prediction model, and obtaining a number of target access data, the access frequency of each target access data, the access mode of each target access data, the data type of each target access data, and the data access peak period of the target terminal.

[0058] Among them, the first prediction model can be a model trained by multiple preset first feature sequences and label information corresponding to each preset first feature sequence, the preset first feature sequence can be expressed as {preset behavior feature, preset time feature, preset user feature, preset context feature}, and the label information corresponding to the first feature sequence can be expressed as {multiple preset access data, access frequency of each preset access data, access mode of each preset access data, data type of each preset access data, preset data access peak time period}. For example, the preset access data includes access data 1 and access data 2, then the label information corresponding to the first feature sequence can be expressed as {access data 1, access frequency of access data 1, access mode of access data 1, data type of access data 1, access data 2, access frequency of access data 2, access mode of access data 2, data type of access data 2, data access peak time period of access data 1 and access data 2}. In addition, the first prediction model can be a neural network model such as a long short-term memory network (Long Short-Term Memory, LSTM), a recurrent neural network (Recurrent Neural Network, RNN), or other machine learning models such as random forests, support vector machines, etc., but is not limited to this.

[0059] Exemplarily, in some embodiments, intelligent prediction can be achieved by using model integration for the characteristics of different features, for example, using a prediction model that is good at processing time series type data to predict time series type data, and using a prediction model that is good at processing non-time series type data to predict non-time series type data, so as to give full play to the advantages of each prediction model and improve the overall prediction performance. In addition, a phased prediction strategy can also be adopted, first using some historical behavior features for preliminary prediction, obtaining some future data, and then using the remaining historical behavior features and some future data for more accurate subsequent prediction. Based on this, the model integration method is combined with the segmented prediction strategy, and the prediction method is: first use the second prediction model to predict the behavior time series feature sequence, obtain a number of target access data, the access frequency of each target access data, the data type of each target access data and the data access peak period of the target terminal, and then use the third prediction model to predict the access frequency of each target access data, the data access peak period of the target terminal, the user characteristics and the context characteristics to obtain the access mode of each target access data.

[0060] Among them, the second prediction model can be a model trained by multiple preset behavior time series feature sequences and label information corresponding to each preset behavior time series feature sequence, and the label information corresponding to the behavior time series feature sequence can be expressed as {multiple preset access data, access frequency of each preset access data, data type of each preset access data, preset data access peak time period}. For example, the preset access data is access data 1 and access data 2, then the label information corresponding to the behavior time series feature sequence can be expressed as {access data 1, access frequency of access data 1, data type of access data 1, access data 2, access frequency of access data 2, data type of access data 2, data access peak time period of access data 1 and access data 2}. In addition, the second prediction model can be a neural network model that is good at processing time series type data, such as a long short-term memory network, a recurrent neural network, etc., but is not limited to this.

[0061] Among them, the third prediction model can be a model trained by multiple preset second feature sequences and label information corresponding to each preset second feature sequence. The preset second feature sequence can be expressed as {access frequency of each preset access data, preset data access peak time period, user characteristics, contextual characteristics}, and the label information corresponding to the second feature sequence can be expressed as {access mode of each preset access data}. For example, if the preset access data is access data 1 and access data 2, then the second feature sequence can be expressed as {access frequency of access data 1, access frequency of access data 2, data access peak time period of access data 1 and access data 2, user characteristics, contextual characteristics}, and the label information corresponding to the second feature sequence can be expressed as {access mode of access data 1, access mode of access data 2}. In addition, the third prediction model can be a machine learning model that is good at processing non-time series type data, such as random forests, decision trees, support vector machines, etc., but not limited to this.

[0062] In some implementations, the target behavior information of the target terminal may include the access frequency of each target access data; in the above step S103, the specific implementation process of determining multiple target preloaded data according to the target behavior information and the plurality of target access data may include the following step S1031: S1031 , determining a plurality of target preloaded data from a plurality of target access data according to the access frequency of each target access data and in combination with the timeliness coefficient and the user tendency coefficient of each target access data.

[0063] In this embodiment, the access frequency of each target access data predicted in the above steps is used, combined with the timeliness coefficient and user tendency coefficient of each target access data, to select target access data that meets the preset conditions from a number of target access data as target preloaded data, thereby obtaining multiple target preloaded data. In this way, this embodiment, while fully considering the access behavior and access requirements of the target terminal in the future time period and multi-dimensional factors such as data timeliness and data user tendency, accurately selects the data objects that the target terminal expects to access in the future time period from a number of target access data, namely, target preloaded data, which can effectively improve the accuracy of the target preloaded data and make the target preloaded data more suitable for the data access requirements of the target terminal.

[0064] The timeliness coefficient of the target access data is positively correlated with the effectiveness of the target access data, that is, the higher the timeliness coefficient of the target access data is, the more effective the target access data is.

[0065] The user tendency coefficient of the target access data is used to indicate the similarity between the target access data and the data access demand of the target terminal. The higher the user tendency coefficient of the target access data is, the closer the target access data is to the data access demand of the target terminal.

[0066] The timeliness coefficient of the target access data and the user tendency coefficient of the target access data may be pre-calibrated according to actual conditions, and this embodiment does not specifically limit this.

[0067] The above-mentioned method of selecting the target preloaded data can be set according to actual conditions, and this embodiment does not limit this.

[0068] Exemplarily, in some embodiments, among several target access data, target access data that meets the preloading conditions are selected as target preloading data, thereby obtaining multiple target preloading data; wherein the preloading conditions include at least one of an access frequency greater than a preset frequency threshold, a timeliness coefficient greater than a preset first coefficient threshold, or a user tendency coefficient greater than a preset second coefficient threshold.

[0069] Exemplarily, in some embodiments, first, based on the access frequency, timeliness coefficient and user tendency coefficient of each target access data, combined with a machine learning method, an importance coefficient of each target access data is obtained, wherein the importance coefficient is used to indicate the importance of the target preloaded data to the target terminal; then, the target access data whose importance coefficient is greater than a preset importance threshold is selected as the target preloaded data, thereby obtaining multiple target preloaded data.

[0070] In some embodiments, reference Figure 3In the above step S104, the specific implementation process of preloading each target preloaded data by using the target behavior information to obtain each preloaded target preloaded data may include the following steps S1041-S1042: S1041, obtaining preloading parameters according to the target behavior information and the preset comprehensive parameters; wherein the preloading parameters include the target storage medium of each target preloading data, the target storage form of each target preloading data and the preloading time of the target terminal; S1042, performing preloading processing on each target preloading data using the preloading parameters to obtain each preloaded target preloading data.

[0071] In this embodiment, after determining the target preload data, the target behavior information and the preset comprehensive parameters are used as the parameter acquisition benchmark to determine the target storage medium of each target preload data, the target storage form of each target preload data and the preload time of the target terminal as the preload parameters. The preload parameters are used to preload the target preload data. In this way, the access behavior and access requirements of the target terminal in the future time period and other related parameters are fully considered to achieve the determination of the preload parameters, which can effectively improve the accuracy of the preload parameters, thereby improving the accuracy of the preload processing. After obtaining the preload parameters, the preload parameters are used to preload each target preload data, thereby obtaining each preloaded target preload data. In this way, before the data access request of the target terminal arrives, each target preload data in the cloud storage service can be preloaded to a storage medium such as a cache server, an edge node, etc., thereby effectively reducing the data delay of the cloud storage service and improving the user experience of the cloud storage service.

[0072] The above-mentioned comprehensive parameters refer to parameters associated with the target terminal, each target preloaded data and each preset storage medium. It can be pre-calibrated or acquired in real time according to actual conditions, and this embodiment does not make any specific limitation on this.

[0073] The above-mentioned preloading parameters may include a target storage medium for each target preloading data, a target storage format for each target preloading data, and a preloading time for a target terminal, but are not limited thereto.

[0074] The target storage medium of the above-mentioned target preloaded data refers to a storage medium used to store the target preloaded data. It is understandable that the type of the target storage medium can be set according to the actual situation, and this embodiment does not specifically limit this. For example, the target storage medium can be an edge node of a cloud storage service. It is understandable that an edge node refers to a business platform built on the edge of a network close to the user, which provides storage, computing, network and other resources, and sinks some key business applications to the edge of the access network to reduce the width and delay losses caused by network transmission and multi-level forwarding. For another example, the target storage medium can be a cache server of a cloud storage service. It is understandable that a cache server is a hardware device that stores and provides data, and is used to accelerate data access and improve system performance.

[0075] The target storage form of the target preloaded data refers to the data form of the target preloaded data when the preload process is performed. It is understandable that the target storage form can be set according to actual conditions, and this embodiment does not specifically limit this. For example, the target storage form can be an image form, a text form, a video form, etc., but is not limited thereto.

[0076] The preloading time of the target terminal refers to the time value for preloading the target preloading data.

[0077] The above-mentioned obtaining the preload parameters according to the target behavior information and the preset comprehensive parameters may include obtaining the preload parameters according to the target behavior information and the preset comprehensive parameters in combination with a machine learning method, but is not limited thereto.

[0078] In some embodiments, reference Figure 4 , the above-mentioned comprehensive parameters may include the importance coefficient of each target preloaded data, the usage frequency of each target preloaded data, the terminal geographical location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographical location of each preset storage medium; In the above step S1041, the specific implementation process of obtaining the preloading parameters according to the target behavior information and the preset comprehensive parameters may include the following steps: According to the importance coefficient and usage frequency of each target preloaded data, combined with the terminal geographic location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographic location of each preset storage medium, the target storage medium of each target preloaded data is determined from multiple preset storage media.

[0079] In this embodiment, for the selection of the target storage medium, first, the terminal geographic location and network characteristic data of the target terminal and the terminal geographic location of each preset storage medium are obtained, so as to dynamically select the storage medium closest to the target terminal to reduce the delay; and the load rate of each preset storage medium is obtained, so as to dynamically select the storage medium with lower load; wherein the load rate can indicate the load degree of the preset storage medium; and the importance coefficient and usage frequency of each target preloaded data are obtained, so as to allocate a storage medium with lower load and sufficient storage space to the high-priority target preloaded data. Then, based on these data, the corresponding target storage medium is allocated to each target preloaded data.

[0080] It can be seen that this implementation method determines the target storage medium for each target preloaded data while fully considering multi-dimensional factors such as data importance, data usage frequency, terminal location, terminal network performance, node load rate and node location. This not only improves the adaptability between the target preloaded data and the target storage medium, balances the network load of the target storage medium, reduces data delay, and optimizes the storage space utilization efficiency of the target storage medium, reduces resource waste caused by excessive resource preloading, but also improves preloading processing and accuracy, which is conducive to achieving better preloading effects, thereby reducing data delays in cloud storage services, meeting users' data access needs, and providing a solid foundation for cloud storage data access.

[0081] It should be noted that different target preloaded data are allowed to share the same target storage medium.

[0082] The above-mentioned comprehensive parameters may include the importance coefficient of each target preloaded data, the usage frequency of each target preloaded data, the terminal geographic location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographic location of each preset storage medium, but are not limited to these.

[0083] The importance coefficient of the above-mentioned target preloaded data is used to indicate the importance of the target preloaded data to the target terminal. The higher the importance coefficient of the target preloaded data, the more important the target preloaded data is to the target terminal. In addition, the method for obtaining the importance coefficient of the above-mentioned target preloaded data can be set according to actual conditions, and this embodiment does not specifically limit this. For example, the importance coefficient can be a pre-calibrated value. For another example, according to the access frequency, timeliness coefficient and user tendency coefficient of the target preloaded data, combined with the machine learning method, the importance coefficient of the target preloaded data is obtained, but it is not limited to this.

[0084] The usage frequency of the target preloaded data refers to the usage rate of the target preloaded data per unit time, which may be a pre-calibrated value, but is not limited thereto.

[0085] The terminal geographic location of the target terminal refers to the location of the target terminal. It is understandable that the terminal geographic location of the target terminal can be flexibly obtained according to actual conditions, and this embodiment does not specifically limit this. For example, the terminal geographic location of the target terminal is determined by technologies such as Geographic Information System (GIS) or Internet Protocol (IP) address, Global Positioning System (GPS) information, but is not limited to this.

[0086] The network characteristic data of the target terminal is used to indicate the network transmission performance of the target terminal. The higher the network characteristic data of the target terminal, the better the network transmission performance of the target terminal. It can be understood that the network characteristic data of the target terminal can be the network characteristic data of the target terminal acquired in real time, or the average value of the network characteristic data of the target terminal in a historical time period, but is not limited to this. In addition, the specific type of the network characteristic data of the target terminal can be set according to actual conditions, and this embodiment does not specifically limit this. For example, the network characteristic data of the target terminal can be the uplink speed and downlink speed of the target terminal, etc., but is not limited to this.

[0087] The above-mentioned preset storage medium refers to a storage medium applicable to the cloud storage data access method in the embodiment of the present application. It can be a pre-set storage medium, but is not limited to this. In addition, the type of the above-mentioned preset storage medium can be flexibly set according to actual conditions. For example, the preset storage medium can be a cache server, or other storage media such as edge nodes, which is not specifically limited in this embodiment.

[0088] The terminal geographical location of the preset storage medium is the location of the preset storage medium. It is understandable that the terminal geographical location of the preset storage medium can be flexibly obtained or preset according to actual conditions, and this embodiment does not specifically limit this. For example, the terminal geographical location of the preset storage medium is determined by technologies such as geographic information systems and global positioning systems, but is not limited thereto.

[0089] The load rate of the preset storage medium is used to indicate the load degree of the storage medium. The higher the load rate of the preset storage medium, the more serious the load of the preset storage medium. In addition, the load rate of the preset storage medium can be the load rate of the preset storage medium obtained in real time, or it can be the average value of the load rate of the preset storage medium in a historical time period, but is not limited thereto.

[0090] The above-mentioned method for determining the target storage medium can be set according to actual conditions, and this embodiment does not specifically limit this.

[0091] Exemplarily, in some embodiments, first, according to the importance coefficient and usage frequency of each target preloaded data, combined with the machine learning method, the priority of each target preloaded data is obtained; wherein, the priority of the target preloaded data is positively correlated with the terminal need degree and timeliness coefficient of the target preloaded data, and the terminal need degree of the target preloaded data refers to the degree to which the target terminal needs the target preloaded data, that is, the higher the priority of the target preloaded data, the more the target terminal needs the target preloaded data and the higher the timeliness coefficient of the target preloaded data; then, through an optimization algorithm such as an ant colony algorithm and a genetic algorithm, the priority of each target preloaded data, the terminal geographical location of the target terminal, the network characteristic data of the target terminal and the load rate of each preset storage medium are optimized to obtain the target storage medium of each target preloaded data. It can be understood that the optimization object of the optimization algorithm is the optimal preset storage medium of each target preloaded data, that is, the target storage medium of each target preloaded data. In addition, the objective function of the optimization algorithm and the constraint conditions of the objective function can be flexibly set according to actual conditions.

[0092] For example, the objective function may include three sub-objective functions, the first sub-objective function is to minimize the distance between the target storage medium and the target terminal for each target preloaded data, the distance between the target storage medium and the target terminal can be obtained by the terminal geographical location of the preset storage medium and the terminal geographical location of the target terminal, the second sub-objective function is to maximize the network characteristic data of the target terminal, and the third sub-objective function is to minimize the load rate of the target storage medium for each target preloaded data. Among them, the first sub-objective function and the second sub-objective function can reduce data delay, while the third sub-objective function can achieve node load balancing, reducing the delay and performance degradation caused by overload of data access requests in high-load nodes.

[0093] The constraint conditions of the objective function may include that the priority of the target preloaded data is positively correlated with the storage space of the preset storage medium, that is, the higher the priority of the target preloaded data, the larger the storage space of the preset storage medium, and the priority of the target preloaded data is negatively correlated with the load rate of the preset storage medium, that is, the higher the priority of the target preloaded data, the lower the load rate of the preset storage medium. This constraint condition can be understood as that high-priority data should be preferentially stored in a storage medium with a lower load and sufficient storage space.

[0094] Exemplarily, in some embodiments, for each target preloaded data, based on the importance coefficient of the target preloaded data, the usage frequency of the target preloaded data, the terminal geographic location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographic location of each preset storage medium, combined with a machine learning method, the storage coefficient of each preset storage medium relative to the target preloaded data is obtained, and the storage coefficient can map the performance of the preset storage medium when the preset storage medium stores the target preloaded data, and then the preset storage medium with the highest storage coefficient is selected as the target storage medium for the target preloaded data.

[0095] In some embodiments, reference Figure 4 The target behavior information of the target terminal may include the data type, access mode and access frequency of each target preloaded data; the comprehensive parameter may also include the sensitivity coefficient and medium load rate of each target preloaded data; in the above step S1041, the specific implementation process of obtaining the preloaded parameters according to the target behavior information and the preset comprehensive parameters may also include the following steps: According to the data type, access mode and access frequency of each target preloaded data, combined with the sensitivity coefficient and medium load rate of each target preloaded data, the target storage form of each target preloaded data is determined.

[0096] In this embodiment, for determining the target storage form, first, the data type of each target preloaded data, the access mode of each target preloaded data and the access frequency of each target preloaded data predicted in the previous steps are obtained, and at the same time, the sensitivity coefficient of each target preloaded data and the medium load rate of each target preloaded data are obtained, wherein the sensitivity coefficient refers to the severity of the consequences caused by the leakage of the target access data, and the higher the sensitivity coefficient of the target access data, the more serious the consequences caused by the leakage of the target access data, and the medium load rate refers to the load rate of the target storage medium; then, based on these data, the target storage form of each target preloaded data is determined.

[0097] Specifically, the target storage form is mainly divided into two aspects, one is the encryption compression method, and the other is the storage type.

[0098] For the encryption compression method, the data type and access frequency of the target preloaded data can be used to dynamically select a suitable compression algorithm to compress the target preloaded data, thereby saving the storage space of the target storage medium and improving the preload speed and subsequent data access speed. Among them, the compression algorithm can be flexibly set according to actual conditions. For example, for target preloaded data with an access frequency higher than a preset frequency threshold, a form with a smaller compression rate can be selected as the target storage form of the target preloaded data, so as to improve the decompression speed during subsequent data access, thereby ensuring the data access speed.

[0099] At the same time, the sensitivity coefficient of the target preloaded data can be used to dynamically select a suitable encryption algorithm to encrypt the target preloaded data, thereby ensuring the security of the target preloaded data. The encryption algorithm can be flexibly set according to actual conditions. For example, for target preloaded data with a sensitivity coefficient higher than a preset sensitivity threshold, a strong encryption algorithm such as a symmetric encryption algorithm (Advanced Encryption Standard, AES) is used for encryption processing, while for target preloaded data with a sensitivity coefficient not higher than the sensitivity threshold, a lightweight encryption algorithm can be used for encryption processing to ensure good performance.

[0100] For the storage type, the data type of the target preloaded data, the access mode of the target preloaded data, and the medium load rate of the target preloaded data can be used to dynamically select a suitable storage type to optimize the preloading speed and storage efficiency. For example, the target preloaded data can be preloaded and stored in a lower resolution image type to increase the preloading speed and subsequent data access speed. For another example, the original type of target preloaded data can be used for storage in a storage medium with a higher bandwidth, while the compressed target preloaded data can be used for storage in a storage medium with a limited bandwidth.

[0101] It can be seen that this embodiment determines the target storage form of each target preloaded data while fully considering multi-dimensional factors such as data type, data access mode, data access frequency, data sensitivity coefficient and node load rate. This not only improves the adaptability between the target preloaded data and the target storage medium and reduces the waste of resources caused by excessive preloading of resources, but also improves the efficiency and accuracy of the preloading process, which is conducive to achieving better preloading effects, thereby reducing the data delay of the cloud storage service, meeting the data access needs of users, and providing a solid foundation for cloud storage data access.

[0102] The above-mentioned comprehensive parameters may also include a sensitivity coefficient of each target preloaded data and a medium load rate, but are not limited thereto.

[0103] The above sensitivity coefficient refers to the severity of the consequences caused by the leakage of target access data. The higher the sensitivity coefficient of the target access data, the more serious the consequences caused by the leakage of the target access data. It can be a pre-calibrated value, but is not limited to this.

[0104] The above medium load rate refers to the load rate of the target storage medium, which may be data obtained in real time, but is not limited thereto.

[0105] The method for determining the above target storage format can be set according to actual conditions, and this embodiment does not specifically limit this.

[0106] Exemplarily, in some embodiments, for each target preloaded data, the target storage form of the target preloaded data is obtained based on the data type of the target preloaded data, the access mode of the target preloaded data, the sensitivity coefficient of the target preloaded data and the medium load rate of the target preloaded data, combined with a machine learning method.

[0107] The target storage form includes, but is not limited to, an encrypted image form, a compressed image form, an encrypted text form, a compressed text form, an encrypted video form, a compressed video form, and an original form. It is understood that the encrypted image form refers to an image type processed by an encryption algorithm, and the encrypted text form and the encrypted video form are the same. The compressed image form refers to an image type processed by a compression algorithm, and the compressed text form and the compressed video form are the same. The ordinary image form refers to the original type that has not been encrypted or compressed.

[0108] Exemplarily, in some embodiments, for each target preloaded data, the encryption compression method of the target preloaded data is determined based on the data type, access frequency and sensitivity coefficient of the target preloaded data, combined with a machine learning method, wherein the encryption compression method includes any one of an encryption method, a compression method or a non-processing method, and the non-processing method can be understood as a method without any encryption and compression processing. Then, based on the encryption compression method of the target preloaded data, the data type of the target preloaded data, the access mode of the target preloaded data and the medium load rate of the target preloaded data, combined with a machine learning method, the storage type of the target preloaded data is obtained, and the storage type includes any one of an image type, a text type or a video type. Finally, based on the storage type and the encryption compression method of the target preloaded data, the target storage form of the target preloaded data is obtained.

[0109] The target storage form includes encrypted image form, ordinary image form, compressed image form, encrypted text form, ordinary text form, compressed text form, encrypted video form, ordinary video form and compressed video form, but is not limited thereto. It can be understood that the encrypted image form means that the storage type is image type and the encryption compression method is encryption method, the encrypted text form and the encrypted video form are the same, the compressed image form means that the storage type is image type and the encryption compression method is compression method, the compressed text form and the compressed video form are the same, the ordinary image form means that the storage type is image type and the encryption compression method is non-processing method, the ordinary text form and the compressed video form are the same.

[0110] In some embodiments, reference Figure 4The target behavior information of the target terminal may also include the peak data access period of the target terminal; the comprehensive parameter may also include the network characteristic data of the target terminal; in the above step S1041, the specific implementation process of obtaining the preload parameter according to the target behavior information and the preset comprehensive parameter may also include the following steps: According to the data access peak period of the target terminal and the network characteristic data of the target terminal, the preloading time of the target terminal is obtained.

[0111] In this embodiment, for the determination of the preloading time, first, the data access peak period of the target terminal predicted in the above steps is obtained, and the network characteristic data of the target terminal is obtained at the same time; then, the preloading time of the target terminal is dynamically selected based on the data access peak period and the network characteristic data of the target terminal, for example, large-scale data preloading is performed during the time period when the network characteristic data is optimal and not at the data access peak, so as to reduce the network occupancy rate and the impact on the user experience. It can be seen that this embodiment determines the preloading time while fully considering multi-dimensional factors such as data access peak and terminal network performance, which can improve the degree of adaptation between the preloading process and the cloud storage data access timing, reduce the waste of resources caused by excessive resource preloading, and improve the efficiency and accuracy of the preloading process, which is conducive to achieving a better preloading effect, thereby reducing the data delay of the cloud storage service, meeting the data access needs of users, and providing a solid foundation for cloud storage data access.

[0112] The method for determining the preloading time can be set according to actual conditions, and this embodiment does not specifically limit this.

[0113] Exemplarily, in some embodiments, the preloading time of the target terminal is obtained based on the data access peak period of the target terminal and network characteristic data, combined with a machine learning method.

[0114] Exemplarily, in some embodiments, a preset time period with the highest network characteristic data is selected from several preset time periods as a candidate time period, and then the candidate time period closest to the data access peak period is selected as the preloading time of the target terminal.

[0115] In some embodiments, reference Figure 5 and Figure 6 In the above step S1042, the specific implementation process of preloading each target preloaded data using the preloaded parameters to obtain each preloaded target preloaded data may include the following steps S01-S03: S01, transforming the storage form of the target preloaded data into the target storage form of the target preloaded data; S02, loading the target preload data into a preload queue corresponding to the target preload data; S03, when the preloading time of the target terminal is reached, the target preloading data is mapped from the preloading queue corresponding to the target preloading data to the target storage medium of the target preloading data to obtain the preloaded target preloading data; wherein the preloaded target preloading data is stored in the target storage medium of the target preloading data in the target storage form of the target preloading data.

[0116] In this implementation, when performing the preloading process, for each target preloading data, there are: First, the data storage form of the target preloaded data is transformed according to the target storage form of the target preloaded data, so as to transform the data storage form of the target preloaded data into the target storage form of the target preloaded data. For example, the current storage form of the target preloaded data is an image type that is not encrypted or compressed, while the target storage form of the target preloaded data is a text type that has been processed by an encryption algorithm. When the preload process is performed, the target preloaded data is transformed from an image type that is not encrypted or compressed to a text type that has been processed by an encryption algorithm. For another example, the current storage form of the target preloaded data is an image type that is not encrypted or compressed, while the target storage form of the target preloaded data is an image type that has been processed by a compression algorithm. When the preload process is performed, the target preloaded data is transformed from an image type that is not encrypted or compressed to an image type that has been processed by a compression algorithm.

[0117] After completing the transformation of the data storage format, several preloading queues are obtained, each of which is used to preload the corresponding target preloading data, and the preloading queue corresponding to the target preloading data is determined from the several preloading queues, and the target preloading data is loaded into the corresponding preloading queue. It can be understood that the determination of the preloading queue can be flexibly set according to the actual situation, for example, an idle preloading queue is randomly assigned to the target preloading data, and the fully loaded preloading queue does not participate in the allocation. In addition, the preloading queues corresponding to the target preloading data run in parallel to ensure that different target preloading data can be processed in parallel, shortening the overall time of the preloading process and improving the user experience.

[0118] Afterwards, wait for the arrival of the preloading time of the target terminal. When the preloading time of the target terminal arrives, map the target preloading data from the preloading queue corresponding to the target preloading data to the target storage medium of the target preloading data, so that the target preloading data is stored in the target storage medium in the target storage form, thereby obtaining the preloaded target preloading data. By traversing all the target preloading data, multiple preloaded target preloading data can be obtained.

[0119] It can be seen that this embodiment transforms the storage form of the target preloaded data into the target storage form of the target preloaded data before the preloading time is arrived, and uses the preloading queue as an intermediate carrier to load the target preloaded data into the preloading queue corresponding to the target preloaded data. When the preloading time of the target terminal is arrived, the target preloaded data is mapped from the preloading queue corresponding to the target preloaded data to the target storage medium of the target preloaded data, thereby realizing the preloading processing of the target preloaded data. This can effectively improve the efficiency of the preloading processing, ensure that the preloading processing of each target preloaded data is completed before the target terminal accesses the data, and at the same time effectively improve the accuracy of the preloading processing, ensure that each target preloaded data can be preloaded into the corresponding target storage medium, which is beneficial to reduce the data delay of the cloud storage service, speed up the response to data access requests of the cloud storage service, and improve the user experience of the cloud storage service.

[0120] It should be noted that, when there is only one target terminal and the target terminal corresponds to at least one target preloaded data, the relationship between the target preloaded data, the preloaded queue, the target storage medium and the target terminal is as follows: one target preloaded data corresponds to one preloaded queue, and at the same time, one preloaded queue is allowed to preload one or more target preloaded data, which means that different target preloaded data can correspond to the same preloaded queue. A preloaded queue can correspond to a target storage medium of one or more target preloaded data, that is, the same preloaded queue is allowed to have a mapping relationship with one or more target storage media. The target terminal can correspond to the target storage medium of one or more target preloaded data, that is, the target terminal is allowed to be configured with one or more target storage media. Furthermore, when facing multiple target terminals, one target storage medium can correspond to one or more target terminals, that is, different target terminals are allowed to share the same target storage medium.

[0121] For example, in some embodiments, referring to Figure 7 , terminal 1 corresponds to target preloaded data 1 and target preloaded data 2, target preloaded data 1 and target preloaded data 2 can share preloaded queue 1, the target storage medium of target preloaded data 1 is edge node 1, and the target storage medium of target preloaded data 2 is edge node 2; when preloading, preload target preloaded data 1 and target preloaded data 2 are preloaded into preloaded queue 1, waiting for the arrival of preload time; when the preload time arrives, store target preloaded data 1 in preloaded queue 1 into edge node 1, and store target preloaded data 2 in preloaded queue 1 into edge node 2; when accessing data, terminal 1 can access target preloaded data 1 through edge node 1, and access target preloaded data 2 through edge node 2.

[0122] Terminal 2 corresponds to target preloaded data 3 and target preloaded data 4, target preloaded data 3 is configured with preloaded queue 2, target preloaded data 4 is configured with preloaded queue 3, target storage medium of target preloaded data 3 is edge node 2, target storage medium of target preloaded data 4 is cache server 1; during preloading, target preloaded data 3 is preloaded into preloaded queue 2, target preloaded data 4 is preloaded into preloaded queue 3, waiting for the arrival of preloading time; when the preloading time arrives, target preloaded data 3 in preloaded queue 2 is stored into edge node 2, and target preloaded data 4 in preloaded queue 3 is stored into cache server 1; during data access, terminal 2 can access target preloaded data 3 through edge node 2, and access target preloaded data 4 through cache server 1.

[0123] The mapping method between the preload queue and the target storage medium can be set according to actual conditions, and this embodiment does not specifically limit this. For example, the mapping method can be a hash mapping method; or the mapping method can be a tree mapping method, a linear mapping method, or other mapping methods, but is not limited thereto.

[0124] In some embodiments, reference Figure 8 In the above step S104, after preloading each target preloaded data using the target behavior information to obtain each preloaded target preloaded data, the above method may further include at least one of the following steps S11-S12: S11, performing data migration processing on a target storage medium of each target preloaded data according to the medium usage rate of each target preloaded data.

[0125] In this step, after the preloading process is completed, the target storage medium of each target preloaded data is monitored in real time to migrate data for the overloaded target storage medium, thereby releasing the storage space of part of the target storage medium. Specifically, for each target storage medium of the preloaded data, first, the medium usage rate is obtained; then, based on the medium usage rate, it is determined whether the target storage medium is overloaded; if so, data migration processing is performed on the target storage medium to release the storage space of the overloaded target storage medium, otherwise no processing is performed. In this way, this step can improve the resource utilization of the target storage medium with sufficient storage space, and reduce the load of the target storage medium with insufficient storage space, thereby balancing the storage space between the target storage media and ensuring the efficiency of subsequent cloud storage data access.

[0126] The above-mentioned medium usage rate refers to the ratio of the used storage space parameter of the target storage medium to the total storage space parameter of the target storage medium, and it can be data obtained in real time.

[0127] The specific implementation of the above data migration process can be set according to actual conditions, and this embodiment does not make any specific limitation to this.

[0128] Exemplarily, in some embodiments, for the target storage medium of each target preloaded data, if the medium usage rate is greater than a preset usage rate threshold, it means that the target storage medium is overloaded. At this time, the target storage medium whose usage rate is less than the preset usage rate threshold is first determined as the storage medium to be migrated; then, from the target preloaded data stored in the target storage medium, some target preloaded data are randomly selected as the preloaded data to be migrated; thereafter, the preloaded data to be migrated is migrated to the storage medium to be migrated, so that the medium usage rate drops to the usage rate threshold, and it is ensured that the usage rate of the storage medium to be migrated after the data migration continues to be less than the usage rate threshold; finally, the target storage medium of the preloaded data to be migrated is updated to the storage medium to be migrated.

[0129] Exemplarily, in some embodiments, for the target storage medium of each target preloaded data, if the medium usage rate is greater than a preset usage rate threshold, it means that the target storage medium is overloaded. At this time, the target storage medium whose usage rate is less than the preset usage rate threshold is first determined as the storage medium to be migrated; then, from the target preloaded data stored in the target storage medium, the target preloaded data whose usage rate is lower than the preset usage rate threshold and whose timeliness coefficient is lower than the preset timeliness threshold is selected as the preloaded data to be migrated; thereafter, the preloaded data to be migrated is migrated to the storage medium to be migrated, so that the medium usage rate drops to the usage rate threshold, and it is ensured that the usage rate of the storage medium to be migrated after the data migration continues to be less than the usage rate threshold; finally, the target storage medium of the preloaded data to be migrated is updated to the storage medium to be migrated.

[0130] The above usage rate threshold can be set according to actual conditions, and this embodiment does not specifically limit this. For example, the usage rate threshold is 70%, but it is not limited thereto.

[0131] S12, performing data clearing processing on the target storage medium of each target preloaded data according to the priority of each target preloaded data.

[0132] In this step, after the preloading process is completed, the target storage medium of each target preloaded data is monitored in real time to delete expired or unnecessary data, thereby optimizing the use of storage resources of the target storage medium. Specifically, for each target preloaded data, first, the priority of the target preloaded data is obtained; then, the target preloaded data with a lower priority is deleted from the corresponding target storage medium to release the storage space of the target storage medium. In this way, this step can effectively reduce the overload probability of the target storage medium and improve the resource utilization of the target storage medium, thereby balancing the storage space between each target storage medium and ensuring the efficiency of subsequent cloud storage data access.

[0133] The priority of the above-mentioned target preloading data is positively correlated with the terminal need degree and timeliness coefficient of the target preloading data. The terminal need degree of the target preloading data refers to the degree to which the target terminal needs the target preloading data, that is, the higher the priority of the target preloading data, the more the target terminal needs the target preloading data and the higher the timeliness coefficient of the target preloading data, which means that the target preloading data with low priority is expired or not needed.

[0134] The method for obtaining the priority of the above-mentioned target preloaded data can be preset according to actual conditions, and this embodiment does not specifically limit this.

[0135] Exemplarily, in some embodiments, for each target preloaded data, the priority of the target preloaded data is obtained according to the importance coefficient and usage frequency of the target preloaded data in combination with a machine learning method.

[0136] Exemplarily, in some embodiments, for each target preloading data, the priority corresponding to the importance coefficient and usage frequency of the target preloading data is searched from the fourth mapping data as the priority of the target preloading data, wherein the fourth mapping data may include multiple preset comprehensive parameters and the priority corresponding to each preset comprehensive parameter, and the preset comprehensive parameter may include a preset importance coefficient and a preset usage frequency.

[0137] The fourth mapping data may be chart data or table data, but is not limited thereto.

[0138] The implementation method of the above data cleaning process can be preset according to actual conditions, and this embodiment does not make any specific limitation to this.

[0139] Exemplarily, in some embodiments, for each target preloaded data, if the priority of the target preloaded data is lower than a preset priority threshold, the target preloaded data is treated as preloaded data to be deleted; and all preloaded data to be deleted are deleted from the corresponding target storage medium.

[0140] Exemplarily, in some embodiments, each target preloaded data is sorted according to its priority to obtain sorted target preloaded data, wherein the higher the priority of the target preloaded data, the higher the sorting of the target preloaded data, and then from all the sorted target preloaded data, the last N target preloaded data are selected as the preloaded data to be deleted, and all the preloaded data to be deleted are deleted from the target storage medium corresponding to them.

[0141] Secondly, refer to Fig. 9 , an embodiment of the present application provides a cloud storage data access device, which may include: The acquisition module 201 is used to acquire historical behavior information of the target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; The first processing module 202 is used to obtain target behavior information and a number of target access data of the target terminal according to the historical behavior information; the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; The second processing module 203 is used to determine the preloading parameters and the multiple target preloading data according to the target behavior information and the multiple target access data; The third processing module 204 is used to perform preloading processing on each target preloading data according to the preloading parameters to obtain each preloaded target preloading data; The fourth processing module 205 is used to perform access processing on each preloaded target preloaded data in response to the access instruction of the target terminal.

[0142] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0143] To facilitate the understanding of the cloud storage data access method and device in the embodiments of the present application, the actual application scenario of the cloud storage data access method and device in the present application is used as an example. Fig.10 For a single terminal, the specific process of implementing cloud storage data access in the embodiment of the present application is shown in the following steps S301-S306.

[0144] S301, historical data collection and analysis: obtaining parameters associated with the cloud storage data access behavior of the target terminal in the historical time period through the application programming interface, that is, historical behavior information, which includes the behavior characteristic data, time characteristic data, user characteristic data and context characteristic data of the target terminal in the historical time period.

[0145] S302, data processing and feature engineering: first, preprocess the historical behavior information to obtain preprocessed historical behavior information; then, perform feature extraction on the preprocessed historical behavior information to obtain historical behavior features.

[0146] S303: Intelligent prediction: Predictive processing is performed on historical behavior features to capture the access behavior and access needs of the target terminal in the future time period, thereby obtaining target behavior information and a number of target access data. The target behavior information includes the access frequency of each target access data, the access mode of each target access data, the data type of each target access data and the data access peak period of the target terminal.

[0147] S304, analysis of preloading parameters: Based on the prediction results, a dynamic preloading strategy is designed, which includes the selection of target preloading data, the determination of preloading time, the determination of the storage medium of the target preloading data, and the determination of the storage form of the target preloading data. In this way, multiple target preloading data, the target storage medium of each target preloading data, the target storage form of each target preloading data, and the preloading time of the target terminal can be obtained.

[0148] S305, implementation of data preloading processing: First, the data storage form of the target preloading data is transformed according to the target storage form of the target preloading data, so as to transform the data storage form of the target preloading data into the target storage form of the target preloading data. After completing the data storage form transformation, several preloading queues are obtained, each of which is used to preload the corresponding target preloading data, and the preloading queue corresponding to the target preloading data is determined from the several preloading queues, and then the target preloading data is loaded into the corresponding preloading queue. After that, the arrival of the preloading time of the target terminal is waited. When the preloading time of the target terminal arrives, the target preloading data is mapped from the preloading queue corresponding to the target preloading data to the target storage medium of the target preloading data, so that the target preloading data is stored in the target storage medium in the target storage form, thereby obtaining the preloaded target preloading data. By traversing all the target preloading data, multiple preloaded target preloading data can be obtained. In addition, in order to optimize the storage space and resource utilization of each target storage medium, after the data is preloaded, each target storage medium can also be subjected to optimization processing such as data migration and data cleaning.

[0149] S306, access to cloud storage data: receiving an access instruction from the target terminal, and in response to the access instruction from the target terminal, obtaining each preloaded target preloaded data from the target storage medium of each preloaded target preloaded data, and sending each preloaded target preloaded data to the target terminal, so that the target terminal performs operations such as decryption, decompression, etc. on each preloaded target preloaded data and renders and displays it.

[0150] In some application scenarios, the user of the cloud storage service is an enterprise user, who needs to access the financial report and project report of the previous month in large quantities at the beginning of each month. Historical data shows that each access peak occurs in the first week of each month. In this application scenario, the historical behavior information of the enterprise user is predicted and analyzed through intelligent prediction technology, and the preloading strategy of the enterprise user is designed based on the prediction results. In the preloading strategy of the enterprise user, the first target preloading data is the financial report, the target storage medium of the target preloading data is the edge node A, and the target storage form of the target preloading data is the encrypted text form; the second target preloading data is the project report, the target storage medium of the target preloading data is the cache server B, and the target storage form of the target preloading data is the compressed text form; the preloading time is the week before the peak period of data access, and it is predicted that the peak period of data access is the first week of the next month. Based on this, in the last week of this month, the financial report is preloaded and stored in the encrypted text form to the edge node A through the preloading queue, and the project report is preloaded and stored in the compressed text form to the cache server B through the preloading queue. When the first week of the next month arrives, enterprise users access financial reports preloaded on edge node A and project reports preloaded on cache server B. The average access delay of financial reports and project reports can be reduced from 2.5 seconds to 0.5 seconds, significantly improving the access speed of cloud storage data and user experience.

[0151] In other application scenarios, the users of the cloud storage service are entertainment platform users, who usually watch popular movies and TV series on weekends and holidays. Historical data shows that the users of the entertainment platform prefer to watch specific types of programs in the evening. In this application scenario, the historical behavior information of the entertainment platform user is predicted and analyzed through intelligent prediction technology, and the preloading strategy of the entertainment platform user is designed based on the prediction results. In the preloading strategy of the entertainment platform user, the target preloading data includes program 1, program 2, ..., program N, the target storage medium of all target preloading data is edge node C, the target storage form of all target preloading data is ordinary video form, and the preloading time is the first two hours of the peak data access period. It is predicted that the peak data access period is from 17:00 to 20:00 today. Based on this, at 15:00 today, program 1, program 2, ..., program N are preloaded and stored in the edge node C in the form of ordinary video through the preloading queue. At 5 p.m. today, users of the entertainment platform accessed and watched various programs preloaded on edge node C. The video loading speed of various programs was increased by 50%, and the video freeze problems reported by users were reduced by 70%, improving the video viewing experience.

[0152] In some other application scenarios, the user of the cloud storage service is an online education platform user, who usually provides students with access to a large amount of course materials during the peak class period. Historical data shows that students start to access a large amount of course materials 10 minutes before class. In this application scenario, the historical behavior information of the online education platform user is predicted and analyzed by intelligent prediction technology, and the preloading strategy of the online education platform user is designed based on the prediction results. In the preloading strategy of the online education platform user, the target preloading data is course material 1, course material 2, course material 3, ..., course material N, the target storage medium of all target preloaded data is cache server D, the target storage form of all target preloaded data is ordinary text form, and the preloading time is thirty minutes during the peak data access period. It is predicted that the peak data access period is from 15:00 to 17:00 today. Based on this, at 14:30 today, course material 1, course material 2, course material 3, ..., course material N are preloaded and stored in cache server D through the preloading queue. At 3:00 p.m. today, users of the online education platform accessed various course materials preloaded on cache server D. The average loading time of various course materials could be shortened from the original 5 seconds to 1 second. The problem of slow loading of course materials reported by students was basically solved, and the class experience was improved.

[0153] In summary, it can be seen that the embodiments of the present application can effectively improve the data access efficiency of the cloud storage service and optimize the user experience. It not only alleviates the data latency problem, but also improves the resource utilization of the cloud storage service, demonstrating its great potential and value in the optimization of cloud storage services.

[0154] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

[0155] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the described embodiments. Technical personnel familiar with the field may make various equivalent modifications or substitutions without violating the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A cloud storage data access method, characterized in that: The following steps are involved: Acquire historical behavior information of the target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; According to the historical behavior information, target behavior information and a plurality of target access data of the target terminal are obtained; wherein the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; Determining a plurality of target preload data according to the target behavior information and a plurality of the target access data; Performing preloading processing on each of the target preloaded data using the target behavior information to obtain each of the preloaded target preloaded data; In response to the access instruction of the target terminal, access processing is performed on the preloaded target preload data.

2. The cloud storage data access method according to claim 1, characterized in that: The step of obtaining target behavior information and a plurality of target access data of the target terminal according to the historical behavior information includes: Preprocessing the historical behavior information to obtain the preprocessed historical behavior information; Performing feature extraction on the preprocessed historical behavior information to obtain historical behavior features; The historical behavior characteristics are predicted to obtain the target behavior information and a plurality of target access data.

3. The cloud storage data access method according to claim 1, characterized in that: The target behavior information of the target terminal includes the access frequency of each target access data; and determining a plurality of target preloaded data according to the target behavior information and a plurality of the target access data includes: According to the access frequency of each target access data, combined with the timeliness coefficient and the user tendency coefficient of each target access data, a plurality of target preloaded data are determined from a plurality of target access data.

4. The cloud storage data access method according to claim 1, characterized in that: The preloading process is performed on each of the target preloaded data by using the target behavior information to obtain each of the preloaded target preloaded data, including: According to the target behavior information and the preset comprehensive parameters, a preloading parameter is obtained; wherein the preloading parameter includes a target storage medium of each of the target preloading data, a target storage form of each of the target preloading data and a preloading time of the target terminal; The preloading parameters are used to perform preloading processing on each of the target preloading data to obtain each of the preloaded target preloading data.

5. The cloud storage data access method according to claim 4, characterized in that: The comprehensive parameters include the importance coefficient of each of the target preloaded data, the usage frequency of each of the target preloaded data, the terminal geographical location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographical location of each of the preset storage medium; The step of obtaining the preloading parameters according to the target behavior information and the preset comprehensive parameters includes: According to the importance coefficient and usage frequency of each target preloaded data, combined with the terminal geographic location of the target terminal, the network characteristic data of the target terminal, the load rate of each preset storage medium and the terminal geographic location of each preset storage medium, the target storage medium of each target preloaded data is determined among the multiple preset storage media.

6. The cloud storage data access method according to claim 4, characterized in that: The target behavior information of the target terminal includes the data type, access mode and access frequency of each of the target preloaded data; the comprehensive parameters include the sensitivity coefficient and medium load rate of each of the target preloaded data; The step of obtaining the preloading parameters according to the target behavior information and the preset comprehensive parameters includes: According to the data type, access mode and access frequency of each target preloaded data, combined with the sensitivity coefficient and medium load rate of each target preloaded data, the target storage form of each target preloaded data is determined.

7. The cloud storage data access method according to claim 4, characterized in that: The target behavior information of the target terminal includes the data access peak period of the target terminal; the comprehensive parameter includes the network characteristic data of the target terminal; The step of obtaining the preloading parameters according to the target behavior information and the preset comprehensive parameters includes: The preloading time of the target terminal is obtained according to the data access peak period of the target terminal in combination with the network characteristic data of the target terminal.

8. The cloud storage data access method according to claim 4, characterized in that: The preloading process is performed on each of the target preloading data by using the preloading parameters to obtain each of the preloaded target preloading data, including: transforming the storage format of the target preloaded data into the target storage format of the target preloaded data; Loading the target preload data into a preload queue corresponding to the target preload data; When the preloading time of the target terminal is reached, mapping the target preloading data from the preloading queue corresponding to the target preloading data to the target storage medium of the target preloading data to obtain the preloaded target preloading data; The preloaded target preloaded data is stored in the target storage medium of the target preloaded data in the target storage format of the target preloaded data.

9. The cloud storage data access method according to claim 1, characterized in that: After performing preloading processing on each of the target preloaded data using the target behavior information to obtain each of the preloaded target preloaded data, the method further includes at least one of the following: performing data migration processing on a target storage medium of each target preloaded data according to a medium usage rate of each target preloaded data; According to the priority of each target preloaded data, data clearing processing is performed on the target storage medium of each target preloaded data.

10. A data access device, characterized in that: include: An acquisition module, used to acquire historical behavior information of a target terminal; wherein the historical behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a historical time period; A first processing module, configured to obtain target behavior information and a plurality of target access data of the target terminal according to the historical behavior information; wherein the target behavior information is a parameter associated with the cloud storage data access behavior of the target terminal in a future time period; A second processing module, configured to determine a plurality of target preload data according to the target behavior information and a plurality of the target access data; A third processing module, configured to perform preloading processing on each of the target preloaded data using the target behavior information to obtain each of the preloaded target preloaded data; The fourth processing module is used to perform access processing on each preloaded target preloaded data in response to the access instruction of the target terminal.

Images