A data storage management method, system and medium of a cloud computer

Abstract

The present application relates to the technical field of data storage, and particularly relates to a data storage management method and system of a cloud computer and a medium, access data of different terminal devices is acquired, access data of each logical unit and path distance between each two logical units are obtained based on teaching data logic, then access habits of a single user are analyzed, and access heat indexes of each user for each logical unit are determined; for each two users with the same identity tag and interactive behaviors, access habits of a group are analyzed according to the interactive times of each two users, the access heat indexes and edge storage nodes of the two users, and storage priority indexes of each logical unit at the current moment are determined; and through the storage priority indexes of each logical unit at the current moment and the access heat indexes, a storage scheme is adjusted. The present application enhances the data storage efficiency in a high-concurrency access scene by performing hierarchical storage on group high-heat and individual high-heat data.

Description

Technical Field

[0001] This invention relates to the field of data storage technology, and specifically to a data storage management method, system, and medium for cloud computers. Background Technology

[0002] Currently, the teaching platform needs to store data including user learning data and resource cache. Learning data includes things like word lookup history, spoken practice recordings, custom word libraries, and AI learning plans; resource cache includes things like offline dictionary packages and electronic textbooks. When existing cloud computers manage the storage of this data for the teaching platform, due to the limited storage space of terminal devices, they typically store frequently accessed user data on the terminal devices and less frequently accessed data in the cloud, updating it periodically. Frequently accessed data refers to data that has been accessed recently, while less frequently accessed data refers to data that has not been accessed recently, such as knowledge points from the beginning of the semester.

[0003] However, existing data management methods lack a dynamic adjustment mechanism for cached data in actual teaching scenarios, leading to reduced data storage and access efficiency. For example, the same user may log in on multiple devices. Current methods only adjust cached data based on the local access information of a single device, which is not conducive to analyzing individual access habits across multiple devices to adjust cached data. Furthermore, during final week, a large number of users may simultaneously access cold data stored in the cloud, significantly reducing access efficiency. Summary of the Invention

[0004] To address the technical problem of low access efficiency in the aforementioned data management methods applied in teaching scenarios, the present invention aims to provide a data storage management method, system, and medium for cloud computers. The specific technical solution adopted is as follows:

[0005] One embodiment of the present invention provides a data storage management method for a cloud computer, the method comprising the following steps:

[0006] To obtain access data of several users on different terminal devices for several consecutive monitoring periods;

[0007] Based on the integration of teaching data logic, the access data of users in each logical unit for each monitoring period and the path distance between every two logical units are obtained.

[0008] Based on the access data of each user in each logical unit and the path distance between each two logical units, analyze the access habits of individual users and determine the access popularity index of each user for each logical unit.

[0009] For each pair of users with the same identity tag and interactive behavior, the access habits of the group are analyzed based on the number of interactions between the two users, the access popularity index, and the edge storage nodes of the two users to obtain each shared area.

[0010] Based on the frequency of each teacher's access to the shared area in each monitoring period and the access popularity index of each user for each logical unit, the storage priority index of each logical unit at the current moment is determined.

[0011] The storage scheme is adjusted based on the storage priority index and access popularity index of each logical unit at the current moment.

[0012] Furthermore, the step of analyzing individual users' access habits based on each user's access data in each logical unit and the path distance between every two logical units, and determining each user's access popularity index for each logical unit, includes:

[0013] For each user, the access frequency of the user for each logical unit in each monitoring period is obtained based on the access data. The logical units that are frequently accessed are filtered out by the access frequency of each logical unit and recorded as the logical units to be analyzed.

[0014] Based on the path distance between any two logical units to be analyzed and the access frequency corresponding to each logical unit to be analyzed, the correlation degree of high-frequency access data of the user on different terminal devices is determined; the correlation degree is inversely proportional to the path distance and directly proportional to the access frequency.

[0015] Based on the access data, the frequency and duration of user access to each logical unit in the current monitoring period are obtained. Combined with the correlation degree, the access popularity index of the user for each logical unit is determined.

[0016] Furthermore, based on the user's access frequency and duration for each logical unit during the current monitoring period, and in conjunction with the aforementioned correlation degree, a user access popularity index for each logical unit is determined, including:

[0017] Calculate the average access duration of users for all logical units in the current monitoring period, and record it as the average access duration.

[0018] The user's access popularity index for each logical unit is determined based on the first ratio of the user's access duration for each logical unit in the current monitoring period to the average access duration, the access frequency, and the correlation degree.

[0019] The first ratio, the access frequency, and the degree of correlation are all directly proportional to the access popularity index.

[0020] Furthermore, the process of analyzing the access habits of the group based on the number of interactions between each pair of users, the access popularity index, and the edge storage nodes of the two users, to obtain each shared area, includes:

[0021] The evaluation index for each pair of users belonging to the same shared area is determined based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index of the identical high-heat logical units, and the total number of logical units; the high-heat logical units are logical units whose access popularity index is greater than the set access popularity threshold.

[0022] By setting an evaluation threshold, two users whose evaluation indicators are greater than the evaluation threshold and who have the same edge storage node are merged into a shared area to obtain various shared areas.

[0023] Furthermore, the evaluation index for determining whether each pair of users belongs to the same shared area based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index corresponding to the identical high-heat logical units, and the total number of logical units includes:

[0024] Determine a second ratio between the number of identical high-heat logic units corresponding to every two users and the total number of logic units;

[0025] The evaluation index for each pair of users belonging to the same shared area is determined based on the number of interactions between each pair of users, the second ratio, and the average of the access popularity index.

[0026] The average of the number of interactions, the second ratio, and the access popularity index are all directly proportional to the evaluation indicators.

[0027] Furthermore, the determination of the storage priority index of each logical unit at the current moment, based on the access frequency of each teacher interacting with the shared area in each monitoring period and the access popularity index of each user for each logical unit, includes:

[0028] For each shared area, determine the third ratio between the frequency of each teacher's access to the shared area in the current monitoring period and the average frequency of access to the shared area in all monitoring periods for each teacher who has interactive behavior with the shared area.

[0029] Based on the third ratio between the access frequency and the average access frequency, and combined with the average path distance between every two high-traffic logic units accessed by the teacher in the current monitoring period, the pre-exam behavioral characteristic indicators of each teacher who has interactive behavior with the shared area are determined; the third ratio and the average path distance are both proportional to the pre-exam behavioral characteristic indicators.

[0030] Obtain the number of students in the shared area who have marked the target logic unit as a high-heat logic unit at the current time, and the total number of students in the shared area; the target logic unit can be any logic unit.

[0031] Based on the fourth ratio of the number of students to the total number of students, the average access popularity index of each teacher who interacts with the shared area to the target logical unit in the current monitoring period, and the average value of the pre-exam behavioral characteristic indicators, the storage priority index of the target logical unit at the current moment is determined.

[0032] Further, determining the storage priority index of the target logical unit at the current moment based on the fourth ratio of the number of students to the total number of students, the average access popularity index of each teacher interacting with the shared area to the target logical unit during the current monitoring period, and the average of the pre-exam behavioral characteristic indicators includes:

[0033] Obtain the time characteristic value of the shared area at the current moment, wherein the time characteristic value is determined by the time elapsed between the current moment and the next exam time;

[0034] Based on the time feature value, the fourth ratio, the average of the access popularity index, and the average of the pre-exam behavior feature indicators, the storage priority index of the target logic unit at the current moment is determined.

[0035] The time characteristic value, the fourth ratio, the average access popularity index, and the average pre-exam behavioral characteristic index are all proportional to the storage priority index.

[0036] Furthermore, adjusting the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment includes:

[0037] For any user's terminal device, the first target cache data is selected based on the access popularity index, the first target cache data is moved from the device cache to cloud storage, and the current remaining capacity of the terminal device is obtained; the first target cache data refers to the relevant data that has not been marked as a high-heat logic unit from the beginning of the previous monitoring period to the current moment, and the relevant data is all types of teaching data corresponding to the construction of the logic unit;

[0038] According to the access popularity index of each logical unit in descending order, the relevant data of the logical unit that conforms to the storage data type of the terminal device are stored in the terminal device until the storage amount reaches the preset percentage of the current remaining capacity. The remaining data is then stored in the edge storage node corresponding to the user.

[0039] For a single shared area, the second target cache data is selected based on the storage priority index, the second target cache data is moved from the edge storage node to the cloud storage, and the third target cache data is stored in the corresponding edge storage node based on differential storage;

[0040] The second target cache data refers to the relevant data of logical units whose storage priority index is less than a preset priority threshold from the beginning of the previous monitoring period to the present moment, while the third target cache data refers to the relevant data of logical units whose storage priority index is not less than the preset priority threshold.

[0041] Another embodiment of the present invention provides a data storage management system for a cloud computer, comprising:

[0042] Data acquisition module: used to acquire access data of several users on different terminal devices for several consecutive monitoring periods, and integrate it based on teaching data logic to obtain the user's access data in each logical unit and the path distance between every two logical units;

[0043] Data processing module: This module analyzes individual user access habits based on each user's access data in each logical unit and the path distance between any two logical units, determining the access popularity index for each user in each logical unit. For any two users with the same identity tag and interactive behavior, it analyzes the group's access habits based on the number of interactions, access popularity index, and edge storage nodes of the two users, obtaining each shared area. Based on the access frequency of each teacher interacting with the shared area in each monitoring period and the access popularity index of each user in each logical unit, it determines the storage priority index of each logical unit at the current moment.

[0044] The data storage module is used to adjust the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment.

[0045] Another embodiment of the present invention provides a data storage medium for a cloud computer, including a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a data storage management method for a cloud computer.

[0046] The present invention has the following beneficial effects:

[0047] This invention provides a data storage management method, system, and medium for cloud computers. The method determines the access popularity index and storage priority index of each logical unit, and performs layered storage for high-frequency group data and high-frequency individual data. This reduces the possibility of untimely cache updates for less frequently accessed data in high-concurrency scenarios, thus improving data storage efficiency. To adapt to the complex characteristics of data access in teaching scenarios, adjustments are made at the logical unit level, and cache analysis is performed, which saves some efficiency in data storage analysis. The access popularity index of each user for each logical unit is determined by combining access data from multiple devices. This allows for a thorough analysis of individual user access habits, reflecting true and comprehensive user access habits and avoiding waste of storage resources. Users with similar access characteristics are grouped into a shared area. The access frequency of each teacher interacting with the shared area in each monitoring period, as well as the access popularity index of each user for each logical unit, are analyzed to determine the storage priority index of each logical unit at the current moment. This avoids a large number of users making access requests for the same logical unit's data within a similar time period, preventing the possibility of many users caching duplicate data. Attached Figure Description

[0048] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 A flowchart illustrating the implementation of a data storage management method for a cloud computer, as provided in one embodiment of the present invention;

[0050] Figure 2 This is a tree diagram of the logic units in an embodiment of the present invention;

[0051] Figure 3 This is a flowchart illustrating the implementation of step S3 in an embodiment of the present invention;

[0052] Figure 4 This is a flowchart illustrating the implementation of step S4 in an embodiment of the present invention;

[0053] Figure 5 This is a flowchart illustrating the implementation of step S5 in an embodiment of the present invention. Detailed Implementation

[0054] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the technical solution proposed according to the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0056] The application scenarios of this invention can be:

[0057] Existing data management methods used in actual teaching scenarios cannot adjust cached data according to actual conditions, resulting in reduced efficiency in data storage and access. This invention analyzes user access habits and group access habits based on access data from multiple terminal devices, enabling the hierarchical storage of individual high-frequency data and group high-frequency data, thereby effectively improving data storage efficiency in high-concurrency access scenarios.

[0058] One embodiment of the present invention provides a data storage management method for a cloud computer, such as... Figure 1 As shown, it includes the following steps:

[0059] S1, acquire access data of several users on different terminal devices for several consecutive monitoring periods.

[0060] Here, the monitoring period includes the current monitoring period and the historical monitoring period. The current monitoring period is the monitoring period that includes the current moment, and the monitoring period can be set to 24 hours. The terminal device can be a teaching device, and one user corresponds to one account in different teaching devices.

[0061] In this embodiment, in order to facilitate subsequent analysis of individual and group high-temperature data, access data of each user on their corresponding teaching device for each monitoring period is obtained, such as timestamps and access durations for chapter clicks, audio playback, and material downloads.

[0062] The number of monitoring periods can be set by the implementer according to the specific circumstances. An empirical value of 7 can be used, and no specific limit is set here. The teaching equipment accessed by users and the data accessed in the teaching equipment may be different under different monitoring periods. Different users may use different types and numbers of teaching equipment, so only the access data of teaching equipment that have been logged into the user account are obtained.

[0063] It should be noted that devices with weak network conditions, such as word generators, transmit user access records to cloud storage at a fixed time each day via a differential synchronization protocol, while other devices update in real time.

[0064] Thus, this embodiment has obtained access data for each user on different terminal devices for each monitoring period.

[0065] S2 integrates teaching data logic to obtain user access data in each logical unit for each monitoring period, as well as the path distance between every two logical units.

[0066] The teaching platform needs to store rather disorganized data, including heterogeneous data such as text, images, and exercises. Different types of heterogeneous data are typically stored in different locations within the database, and different teaching devices can store different data types. Therefore, to facilitate subsequent cache analysis, the access data of users on different teaching devices for each monitoring period is integrated into logical unit access data based on the teaching data logic.

[0067] As an exemplary implementation, the steps for obtaining access data for each logical unit include:

[0068] To achieve chapter-based data storage, knowledge points, images, exercises, etc., of a specific unit can be aggregated into a logical unit. Each user has multiple teaching devices, and each device has corresponding access data. By dividing this access data into different logical units, we can obtain the access data of each teaching device in a subset of logical units, and thus obtain the user's access data in each logical unit for each monitoring period.

[0069] It is worth noting that the accessed data is only a part of all the data corresponding to the logical unit. When performing cache fractionation later, adjustments will be made on a unit-by-unit basis. The storage granularity of the specific logical unit can be adjusted according to the specific teaching scenario requirements.

[0070] To facilitate subsequent analysis of the correlation between two logical units, it is necessary to determine the path distance between each pair of logical units. As an exemplary implementation, the steps for obtaining the path distance between each pair of logical units include:

[0071] Specifically, a tree diagram of logical units is drawn according to the hierarchical relationship of teaching data, and the path distance in the tree diagram is assigned a value through the hierarchical relationship to obtain the path distance between every two logical units.

[0072] In this diagram, adjacent nodes at the same level are connected by dashed lines, with a path distance value of 1. The path distance between adjacent levels is assigned as follows: the path distance from the lowest level node to its parent node is 2, the path distance from the parent node to its grandparent node is 3, and so on. The tree diagram of the logic unit is shown below. Figure 2 As shown.

[0073] It should be noted that the path distance between two logical units can be the minimum of the sum of the lengths of all line segments between the two logical units. The smaller the path distance, the stronger the correlation between the two logical units.

[0074] Thus, this embodiment has obtained the user access data for each logical unit in each monitoring period and the path distance between every two logical units.

[0075] S3 analyzes the access habits of individual users based on the access data of each user in each logical unit and the path distance between each two logical units, and determines the access popularity index of each user for each logical unit.

[0076] Here, the access popularity index is an indicator determined by analyzing the access data characteristics of individual users. It can be used to represent the access status of learning materials in each logical unit in different monitoring periods. The higher the access popularity index, the more important the learning materials in the logical unit are to the user, and the more necessary it is to cache them in the terminal device so that the user can access them.

[0077] It is common for the same user to log in to the same account on multiple teaching devices. Current data management methods mostly update the cached data of corresponding devices based on the user's data access patterns on their local devices. This results in the access data of a single teaching device failing to reflect the user's true and comprehensive access habits, and also easily leads to duplicate caching of learning materials across different teaching devices, causing a waste of storage resources. Therefore, it is necessary to analyze user access habits by combining access data from different teaching devices, that is, to determine the access frequency index of each user for each logical unit.

[0078] In this embodiment, taking a single user as an example, there may be situations where a user's access habits on different teaching devices are quite fragmented, such as only looking up English words on a single machine or only watching teaching videos on a computer. Therefore, based on the access data of each monitoring period, the correlation between the user's access habits on multiple teaching devices is analyzed, and the correlation between teaching devices is used as a weight to analyze the access popularity index of a single user for a certain logical unit.

[0079] As an exemplary implementation, step S3 described above can be achieved through... Figure 3 Steps S301 to S303 shown are implemented as follows:

[0080] S301: Based on access data, obtain the user's access frequency for each logical unit in each monitoring period, and filter out the frequently accessed logical units by access frequency of each logical unit, and record them as the logical units to be analyzed.

[0081] Specifically, based on all access data corresponding to the user, the access frequency of each logical unit on all teaching devices in all monitoring periods is obtained, and then the average access frequency of all logical units on all teaching devices is determined as the frequency threshold. Logical units with access frequencies greater than the frequency threshold are regarded as frequently accessed logical units and are recorded as logical units to be analyzed.

[0082] Here, access frequency refers to the number of times a user accesses learning materials belonging to the same logical unit multiple times through different terminal devices.

[0083] S302, based on the path distance between every two logical units to be analyzed and the access frequency corresponding to each logical unit to be analyzed, determine the degree of correlation of high-frequency access data of users on different terminal devices.

[0084] Here, the degree of correlation is inversely proportional to the path distance and directly proportional to the access frequency.

[0085] In this embodiment, the shorter the path distance between two logical units to be analyzed, the smaller the associated path length between frequently accessed logical units, the stronger the correlation of data accessed frequently by users on multiple teaching devices, and the higher the correlation of access habits of different terminal devices. By analyzing the access frequency of the logical units to be analyzed, the access characteristics of logical units frequently accessed by users can be identified, and then the user's access habits can be inferred. For example, the access frequency can characterize the stability of access over a period of time. If the access frequency of certain logical units remains at a high level for a long period of time, it indicates that the user's access habits for these logical units are relatively fixed. Therefore, these frequently accessed logical units can reflect the user's regular needs or preferences.

[0086] As an example, the formula for calculating the correlation of frequently accessed data by the i-th user on different terminal devices can be:

[0087] In the formula, This indicates the degree of correlation between frequently accessed data by the i-th user on different terminal devices. This represents a normalization function used to constrain data to the range of 0 to 1. This represents the average access frequency of all logical units to be analyzed for the i-th user. This represents the average path distance between any two logical units to be analyzed corresponding to the i-th user.

[0088] In the formula for calculating the degree of correlation, the average path distance is... The smaller the value, the higher the degree of correlation between frequently accessed logical units. The larger; the average access frequency The larger the value, the more frequently the user accesses the analysis logic unit. This reflects the user's access habits, reveals potential patterns in these habits, and highlights the average path distance. Data characteristics, making by Numerical values ​​indicating a definite degree of correlation are more reliable.

[0089] S303: Based on access data, obtain the user's access frequency and duration for each logical unit in the current monitoring period, and determine the user's access popularity index for each logical unit by combining the degree of correlation.

[0090] In this embodiment, within the current monitoring period including the current moment, the higher the frequency of a user's access to a certain logical unit across all teaching devices up to the current moment, the longer the duration of each access, and the greater the correlation between the user's access habits across multiple teaching devices, the higher the user's access popularity index for the corresponding logical unit at the current moment.

[0091] Specifically, the average access duration of users for all logical units in the current monitoring period is calculated and denoted as the average access duration. Based on the first ratio of the user's access duration for each logical unit to the average access duration, the access frequency, and the degree of correlation, the access popularity index of the user for each logical unit is determined. Among them, the first ratio, the access frequency, and the degree of correlation are all proportional to the access popularity index.

[0092] As an example, the formula for calculating the access popularity index of the i-th user to the k-th logical unit at the current moment can be:

[0093] In the formula, This represents the access popularity index of the i-th user to the k-th logical unit at the current moment. This indicates the degree of correlation between frequently accessed data by the i-th user on different terminal devices. This represents the access frequency of the i-th user to the k-th logical unit during the current monitoring period. This represents the access duration of the i-th user to the k-th logical unit during the current monitoring period. This represents the average access duration of the i-th user across all logical units during the current monitoring period. This represents the first ratio of the access duration of the i-th user to the k-th logical unit to the average access duration, where norm represents the normalization function used to limit the data to between 0 and 1.

[0094] By referring to the access popularity index of the i-th user for the k-th logical unit, we can obtain the access popularity index of each user for each logical unit.

[0095] Thus, this embodiment obtains the access popularity index of each user for each logical unit.

[0096] S4. For each pair of users with the same identity tag and interactive behavior, the access habits of the group are analyzed based on the number of interactions between the two users, the access popularity index, and the edge storage nodes of the two users to obtain each shared area.

[0097] Here, for each user's edge storage node, the user is matched with the edge storage node that has the largest remaining capacity and is spatially closest at the current moment, thus obtaining each user's edge storage node. A shared region groups users with similar access characteristics together; a shared region consists of multiple users and can be used to analyze group access characteristics.

[0098] Typically, users who share information may have similar access habits. For example, classmates may share knowledge points or teachers may distribute exercises to students. If the cache is updated based on the access habits of a single user, a large number of users may send access requests for the same logical unit of data at the same time within a similar time period, which may result in a large number of users caching duplicate data. Therefore, it is necessary to obtain a shared area containing multiple users based on the access correlation between users for analyzing the characteristics of group access.

[0099] In this embodiment, taking two users with the same identity tag and interactive behavior as an example, if the two users have a high frequency of interaction in all monitoring periods and their access to logical units is similar, the two users are more likely to belong to the same shared area. If the remaining users also meet the above requirements with any user in the shared area, the users who meet the requirements are added to the current shared area. The above operation is repeated until there are no users who meet the conditions, thereby obtaining each shared area.

[0100] Interactive behaviors are counted by recording information in the background. Interactive behaviors can include: resource sharing, collaborative editing, handwriting interaction, group learning, and mutual assistance with incorrect questions. Each interactive behavior is counted as one interaction, such as sharing once equals one interaction.

[0101] As an exemplary implementation, step S4 described above can be achieved through... Figure 4 Steps S401 to S402 shown are implemented as follows:

[0102] S401. Based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index corresponding to the identical high-heat logical units, and the total number of logical units, determine the evaluation index for each pair of users belonging to the same shared area.

[0103] Here, a high-heat logic unit is a logic unit whose access heat index is greater than the set access heat threshold. The access heat threshold can be set to 0.5. Implementers can set it according to the specific actual situation, and there is no specific limitation here.

[0104] For every pair of users, the i-th user and the i-th user... Taking a user as an example, since there may be situations where interactions are frequent but the user's location is far away or unstable, for the first user... When iterating through users, priority is given to users whose most recent access time is close and whose interaction frequency is higher. Users who have no interaction with the i-th user are not included in the iteration. If, as of the current time, the i-th user and the i-th user... If two users interact frequently and share many identical high-frequency logic units, then it is highly likely that the two users belong to the same shared group, meaning that the evaluation index of the two users belonging to the same shared area is relatively high.

[0105] Specifically, determine the i-th user and the... The second ratio of the number of identical high-heat logic units corresponding to each user to the total number of logic units; based on the i-th user and the... The average of the number of interactions of each user, the second ratio, and the access popularity index are used to determine the relationship between the i-th user and the i-th user. Evaluation metrics for users belonging to the same shared area; the average of interaction count, second ratio, and access popularity index are all directly proportional to the evaluation metrics.

[0106] As an example, the i-th user and the i-th The formula for calculating the evaluation index for users belonging to the same shared area can be:

[0107] In the formula, Represents the i-th user and the i-th... The evaluation metric for users belonging to the same shared area, where t represents the current time. This indicates that, up to the current moment, the i-th user and the... Number of interactions per user Represents the i-th user and the i-th... The average access popularity index of each user for the same high-popularity logical unit. Represents the i-th user and the i-th... The number of identical high-heat logic units corresponding to each user, where n represents the total number of logic units, is used to analyze the proportion of identical high-heat logic units. Represents the i-th user and the i-th... The second ratio of the number of identical high-temperature logic units corresponding to each user to the total number of logic units, where norm represents the normalization function used to limit the data to between 0 and 1.

[0108] Referring to the i-th user and the i-th... The evaluation metric for each user belonging to the same shared area can be obtained by comparing the evaluation metrics for each pair of users belonging to the same shared area.

[0109] S402, set the evaluation threshold, merge two users with the same edge storage node whose evaluation index is greater than the evaluation threshold into a shared area, and obtain each shared area.

[0110] In this embodiment, an evaluation threshold of 0.5 is set. Two users with evaluation metrics greater than the threshold of 0.5 and sharing the same edge storage node are grouped into the same shared region, thereby obtaining various shared regions. The evaluation threshold can be set by the implementer according to specific circumstances, and is not specifically limited here.

[0111] In a single shared area, obtain the number of users who are marked as high-hot logical units in the current shared area for a certain logical unit. If the number of users marked as high-hot logical units is greater than half of the total number of users in the current shared area, then the corresponding logical unit is determined to be a group high-hot logical unit, that is, group high-hot access data.

[0112] It should be noted that, for the current user, in addition to the group's high-frequency logic units, the remaining high-frequency logic units corresponding to the user are individual high-frequency access data. Individual high-frequency access data includes, for example, a set of frequently missed questions created by the user themselves.

[0113] Thus, this embodiment groups users with similar access characteristics into the same shared area, resulting in various shared areas.

[0114] S5. Based on the frequency of each teacher's access to the shared area in each monitoring period and the access popularity index of each user for each logical unit, determine the storage priority index of each logical unit at the current moment.

[0115] Here, the storage priority index indicates the priority of accessing data in different logical units when performing cache management.

[0116] In teaching scenarios, there are situations where a large number of users simultaneously access logical unit data that is less popular among a certain group. For example, before the final exam, many students begin to review the knowledge points from the beginning of the semester. At this time, it is necessary to store the relevant data in advance to the nearby edge storage node. Otherwise, a large number of concurrent access requests may lead to a significant reduction in storage efficiency. However, existing data management methods cannot achieve the above operation and rely on the exam time input by the user.

[0117] To achieve the above operations, the storage priority index of each logical unit is analyzed based on timing characteristics and teacher behavior.

[0118] As an exemplary implementation, step S5 described above can be achieved through... Figure 5 Steps S501 to S504 shown are implemented as follows:

[0119] S501, for each shared area, determine the third ratio between the frequency of each teacher's access to the shared area in the current monitoring period and the average frequency of access to the shared area in all monitoring periods.

[0120] In this embodiment, the access frequency of each teacher to the shared area in the current monitoring period refers to the total frequency of access to the shared area by all student users, such as the cumulative value of the access frequency to the student error statistics page in the shared area. The average access frequency refers to the average access frequency of teachers with interactive behavior to the shared area in all monitoring periods.

[0121] The third ratio can represent the proportion of the access frequency of the shared area in the current monitoring period relative to all monitoring periods. For example, the larger the third ratio, the greater the frequency of teachers accessing the student error statistics page in the shared area up to the current moment in the current period compared to historical data.

[0122] S502, based on the third ratio between the access frequency and the average access frequency, combined with the average path distance between every two high-heat logic units accessed by the teacher in the current monitoring period, determine the pre-exam behavioral characteristic indicators of each teacher who has interactive behavior with the shared area.

[0123] Here, the third ratio and the mean path distance are both directly proportional to the pre-exam behavioral characteristic indicators.

[0124] Since teacher access behavior has a significant impact on group access habits in teaching scenarios, we quantify pre-exam behavioral characteristics based on teacher access data to facilitate the determination of the storage priority index of logical units at the current moment.

[0125] In this embodiment, when a teacher who interacts with the shared area accesses the error statistics of student users in the corresponding shared area within a short period of time, and frequently accesses the access data of two logical units with a long path distance, it indicates that for the student users in the shared area at the current moment, the exam may be approaching or the teaching focus of the next learning stage may change. In other words, there is a high probability that the student users in the shared area will access the relevant logical units in large numbers within a short period of time.

[0126] It should be noted that if a teacher accesses a student's incorrect answer page, the teacher can only obtain statistical data on students' incorrect answers within the shared area, and cannot obtain students' personal information; two logical units that are far apart may be knowledge points with low correlation.

[0127] Taking the z-th teacher who interacts with the j-th shared area as an example, if the z-th teacher visits the student error statistics page in the j-th shared area more frequently than in historical data up to the current time in the current monitoring period, and the path distance between each pair of logical units visited by the z-th teacher is relatively far, then the z-th teacher's behavior is more consistent with the pre-exam behavior characteristics, that is, the pre-exam behavior characteristic index is larger.

[0128] As an example, the formula for calculating the pre-exam behavioral characteristic index of the z-th teacher who interacts with the j-th shared area can be:

[0129] In the formula, This represents the pre-exam behavioral characteristics of the z-th teacher who interacts with the j-th shared area. This indicates the frequency of access to the shared area by the z-th teacher who interacts with the j-th shared area during the current monitoring period. This represents the average frequency of visits to the shared area by the z-th teacher who interacts with the j-th shared area across all monitoring periods. This represents the third ratio between the access frequency and the mean access frequency. This represents the average path distance between every two hot logic units visited by the z-th teacher who interacts with the j-th shared area during the current monitoring period.

[0130] Referring to the pre-exam behavioral characteristic indicators of the z-th teacher who interacts with the j-th shared area, we can obtain the pre-exam behavioral characteristic indicators of each teacher who interacts with the shared area.

[0131] S503, obtain the number of students in the shared area who have marked the target logic unit as a high-heat logic unit at the current time, and the total number of students in the shared area.

[0132] Here, the target logic unit is any logic unit.

[0133] In this embodiment, for the kth logic unit, it is the number of students that are marked as high-heat logic units and located in the shared area within the current monitoring period including the current time.

[0134] S504. Based on the fourth ratio of the number of students to the total number of students, the average access popularity index of each teacher who has interactive behavior with the shared area to the target logical unit in the current monitoring period, and the average value of the pre-exam behavioral characteristic indicators, determine the storage priority index of the target logical unit at the current moment.

[0135] In this embodiment, the fourth ratio indicates the degree to which the accessed data of the target logical unit belongs to the group of high-frequency data. The larger the fourth ratio, the greater the proportion of the target logical unit marked as a high-frequency logical unit in the shared area, and the greater the possibility that the target logical unit is accessed simultaneously. Therefore, the storage priority index of the target logical unit at the current moment is greater. The average access popularity index indicates the access popularity of the target logical unit by teachers. The larger the average access popularity index, the higher the storage level of the target logical unit should be, since teachers' access in the teaching system has a guiding effect on students. Students will also access the target logical unit under the influence of teachers. The pre-exam behavior characteristic index indicates the degree to which teachers' access behavior conforms to the pre-exam behavior characteristics. The larger the average pre-exam behavior characteristic index, the greater the influence of teachers' access behavior on the group access habits of student users in the shared area, and the higher the storage priority index of the target logical unit corresponding to the user in the shared area at the current moment.

[0136] Specifically, the time characteristic value of the shared area at the current moment is obtained. Based on the time characteristic value, the fourth ratio, the access popularity index, and the average of the pre-exam behavioral characteristic indicators, the storage priority index of the target logical unit at the current moment is determined. Among them, the time characteristic value is determined by the time from the current moment to the next exam time, and the time characteristic value, the fourth ratio, the average of the access popularity index, and the average of the pre-exam behavioral characteristic indicators are all proportional to the storage priority index.

[0137] As an exemplary implementation, the formula for calculating the storage priority index of the k-th logic unit at the current moment can be:

[0138] In the formula, This represents the storage priority index of the k-th logical unit at the current moment. This represents the time feature value at the current moment. This represents the number of students in the j-th shared region who have marked the k-th logic unit as a high-heat logic unit at the current time. This represents the total number of students in the j-th shared area. This represents the average access popularity index of all teachers who have interacted with the j-th shared area to the k-th logical unit during the current monitoring period. This represents the mean of the pre-exam behavioral characteristics of all teachers who interact with the j-th shared area.

[0139] The time characteristic value is: , This indicates the time remaining until the next exam. This represents a non-zero constant, used to avoid cases where the denominator of a fraction is zero; if no user-inputted exam time exists, then... Regarding the time feature value, its main purpose is to ensure that even if the exam time is zero, it will not affect the subsequent storage priority analysis. That is, even if no user enters the accurate exam time into the system, the priority can still be adjusted based on subsequent calculations.

[0140] Referring to the process of determining the storage priority index of the kth logical unit at the current moment, the storage priority index of each logical unit at the current moment can be obtained.

[0141] Thus, this embodiment obtains the storage priority index of each logical unit in each shared region at the current moment.

[0142] S6 adjusts the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment.

[0143] After determining the storage priority index and access popularity index, high-access data for groups and high-access data for individuals are determined based on the two indices. Since the storage capacity of different educational devices is limited and the types of data they can store vary, data with high access popularity for individuals but low access popularity for groups is cached on the individual terminal devices based on the storage capacity of the terminal devices, data with high access popularity for groups is transferred to edge storage nodes, and the remaining access data is transferred to the cloud.

[0144] As an exemplary implementation, step S6 can be achieved by steps S601 to S603 (not shown in the figures):

[0145] S601: For any user's terminal device, filter out the first target cache data based on the access popularity index, move the first target cache data from the device cache to the cloud storage, and obtain the current remaining capacity of the terminal device.

[0146] In this embodiment, the first target cache data refers to the relevant data that has not been marked as a high-heat logic unit from the beginning of the previous monitoring period to the present moment. The relevant data is all types of teaching data corresponding to the construction of the logic unit, that is, the knowledge points, pictures, exercises, etc. of a certain unit are aggregated into a logic unit, which can also be called all types of learning materials in a logic unit.

[0147] S602, according to the access popularity index of each logical unit from large to small, stores the relevant data of the logical unit that conforms to the storage data type of the terminal device into the terminal device, until the storage amount reaches the preset percentage of the current remaining capacity, and stores the remaining data into the edge storage node corresponding to the user.

[0148] In this embodiment, the preset percentage can be set to 80%, which can be set by the implementer according to the specific actual situation, and no specific limitation is made here.

[0149] S603, for a single shared area, selects the second target cache data based on the storage priority index, moves the second target cache data from the edge storage node to the cloud storage, and stores the third target cache data to the corresponding edge storage node based on differential storage.

[0150] In this embodiment, the second target cache data refers to the relevant data of logical units whose priority index is less than the preset priority threshold stored from the beginning of the previous monitoring period to the current moment, while the third target cache data refers to the relevant data of logical units whose priority index is not less than the preset priority threshold.

[0151] The preset priority threshold can be set to 0.6, which can be set by the implementer according to the specific actual situation, and no specific limitation is made here.

[0152] Once a storage solution is obtained, data can be stored based on that solution to improve storage efficiency.

[0153] Another embodiment of the present invention provides a data storage management system for a cloud computer, comprising:

[0154] Data acquisition module: used to acquire access data of several users on different terminal devices for several consecutive monitoring periods, and integrate it based on teaching data logic to obtain the access data of users in each logical unit and the path distance between every two logical units;

[0155] Data processing module: This module analyzes individual user access habits based on each user's access data in each logical unit and the path distance between any two logical units, determining the access popularity index for each user in each logical unit. For any two users with the same identity tag and interactive behavior, it analyzes the group's access habits based on the number of interactions, access popularity index, and edge storage nodes of the two users, obtaining each shared area. Based on the access frequency of each teacher interacting with the shared area in each monitoring period and the access popularity index of each user in each logical unit, it determines the storage priority index of each logical unit at the current moment.

[0156] The data storage module is used to adjust the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment.

[0157] Another embodiment of the present invention provides a data storage medium for a cloud computer, including a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a data storage management method for a cloud computer.

[0158] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A data storage management method for a cloud computer, characterized in that, Includes the following steps: To obtain access data of several users on different terminal devices for several consecutive monitoring periods; Based on the integration of teaching data logic, the access data of users in each logical unit for each monitoring period and the path distance between every two logical units are obtained; the logical unit is a data unit divided according to the teaching unit and used to aggregate at least one of the knowledge points, pictures and exercises of the teaching unit. Based on the access data of each user in each logical unit and the path distance between each two logical units, analyze the access habits of individual users and determine the access popularity index of each user for each logical unit. For each pair of users with the same identity tag and interactive behavior, the access habits of the group are analyzed based on the number of interactions between the two users, the access popularity index, and the edge storage nodes of the two users to obtain each shared area. Based on the frequency of each teacher's access to the shared area in each monitoring period and the access popularity index of each user for each logical unit, the storage priority index of each logical unit at the current moment is determined. Adjust the storage scheme by considering the storage priority index and access popularity index of each logical unit at the current moment; The step of analyzing a single user's access habits based on each user's access data in each logical unit and the path distance between every two logical units, and determining each user's access popularity index for each logical unit, includes: For each user, the access frequency of the user for each logical unit in each monitoring period is obtained based on the access data. The logical units that are frequently accessed are filtered out by the access frequency of each logical unit and recorded as the logical units to be analyzed. Based on the path distance between any two logical units to be analyzed and the access frequency corresponding to each logical unit to be analyzed, the correlation degree of high-frequency access data of the user on different terminal devices is determined; the correlation degree is inversely proportional to the path distance and directly proportional to the access frequency. Based on the access data, the frequency and duration of user access to each logical unit in the current monitoring period are obtained, and combined with the correlation degree, the access popularity index of user for each logical unit is determined. The method involves analyzing the group's access habits based on the number of interactions between each pair of users, the access popularity index, and the edge storage nodes of the two users to obtain various shared areas, including: The evaluation index for each pair of users belonging to the same shared area is determined based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index of the identical high-heat logical units, and the total number of logical units; the high-heat logical units are logical units whose access popularity index is greater than the set access popularity threshold. Set an evaluation threshold, and merge two users with the same edge storage node whose evaluation index is greater than the evaluation threshold into a shared area to obtain various shared areas.

2. The data storage management method for a cloud computer according to claim 1, characterized in that, Based on the user's access frequency and duration for each logical unit during the current monitoring period, and combined with the aforementioned correlation degree, a user access popularity index for each logical unit is determined, including: Calculate the average access duration of users for all logical units in the current monitoring period, and record it as the average access duration. The user's access popularity index for each logical unit is determined based on the first ratio of the user's access duration for each logical unit in the current monitoring period to the average access duration, the access frequency, and the correlation degree. The first ratio, the access frequency, and the degree of correlation are all directly proportional to the access popularity index.

3. The data storage management method for a cloud computer according to claim 1, characterized in that, The evaluation indicators for determining whether two users belong to the same shared area are based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index corresponding to the identical high-heat logical units, and the total number of logical units. These indicators include: Determine a second ratio between the number of identical high-heat logic units corresponding to every two users and the total number of logic units; The evaluation index for each pair of users belonging to the same shared area is determined based on the number of interactions between each pair of users, the second ratio, and the average of the access popularity index. The average of the number of interactions, the second ratio, and the access popularity index are all directly proportional to the evaluation indicators.

4. The data storage management method for a cloud computer according to claim 1, characterized in that, The method of determining the storage priority index of each logical unit at the current moment based on the access frequency of each teacher to the shared area in each monitoring period and the access popularity index of each user to each logical unit, includes: For each shared area, determine the third ratio between the frequency of each teacher's access to the shared area in the current monitoring period and the average frequency of access to the shared area in all monitoring periods for each teacher who has interactive behavior with the shared area. Based on the third ratio between the access frequency and the average access frequency, and combined with the average path distance between every two high-traffic logic units accessed by the teacher in the current monitoring period, the pre-exam behavioral characteristic indicators of each teacher who has interactive behavior with the shared area are determined; the third ratio and the average path distance are both proportional to the pre-exam behavioral characteristic indicators. Obtain the number of students in the shared area who have marked the target logic unit as a high-heat logic unit at the current time, and the total number of students in the shared area; the target logic unit can be any logic unit. Based on the fourth ratio of the number of students to the total number of students, the average access popularity index of each teacher who interacts with the shared area to the target logical unit in the current monitoring period, and the average value of the pre-exam behavioral characteristic indicators, the storage priority index of the target logical unit at the current moment is determined.

5. The data storage management method for a cloud computer according to claim 4, characterized in that, The method of determining the storage priority index of the target logical unit at the current moment based on the fourth ratio of the number of students to the total number of students, the average access popularity index of each teacher interacting with the shared area to the target logical unit in the current monitoring period, and the average of the pre-exam behavioral characteristic indicators includes: Obtain the time characteristic value of the shared area at the current moment, wherein the time characteristic value is determined by the time elapsed between the current moment and the next exam time; Based on the time feature value, the fourth ratio, the average of the access popularity index, and the average of the pre-exam behavior feature indicators, the storage priority index of the target logic unit at the current moment is determined. The time characteristic value, the fourth ratio, the average access popularity index, and the average pre-exam behavioral characteristic index are all proportional to the storage priority index.

6. The data storage management method for a cloud computer according to claim 1, characterized in that, The adjustment of the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment includes: For any user's terminal device, the first target cached data is selected based on the access popularity index, the first target cached data is moved from the device cache to cloud storage, and the current remaining capacity of the terminal device is obtained. The first target cache data refers to the relevant data that has not been marked as a high-heat logic unit from the beginning of the previous monitoring period to the current moment. The relevant data is all types of teaching data corresponding to the construction of the logic unit. According to the access popularity index of each logical unit in descending order, the relevant data of the logical unit that conforms to the storage data type of the terminal device are stored in the terminal device until the storage amount reaches the preset percentage of the current remaining capacity. The remaining data is then stored in the edge storage node corresponding to the user. For a single shared area, the second target cache data is selected based on the storage priority index, the second target cache data is moved from the edge storage node to the cloud storage, and the third target cache data is stored in the corresponding edge storage node based on differential storage; The second target cache data refers to the relevant data of logical units whose storage priority index is less than the preset priority threshold from the beginning of the previous monitoring period to the present moment, while the third target cache data refers to the relevant data of logical units whose storage priority index is not less than the preset priority threshold.

7. A data storage management system for cloud computers, characterized in that, include: Data acquisition module: used to acquire access data of several users on different terminal devices for several consecutive monitoring periods, and integrate it based on teaching data logic to obtain the access data of users in each logical unit and the path distance between every two logical units; the logical unit is a data unit divided according to the teaching unit and used to aggregate at least one of the knowledge points, pictures and exercises of the teaching unit; Data processing module: This module analyzes individual user access habits based on each user's access data in each logical unit and the path distance between any two logical units, determining the access popularity index for each user in each logical unit. For any two users with the same identity tag and interactive behavior, it analyzes the group's access habits based on the number of interactions, access popularity index, and edge storage nodes of the two users, obtaining each shared area. Based on the access frequency of each teacher interacting with the shared area in each monitoring period and the access popularity index of each user in each logical unit, it determines the storage priority index of each logical unit at the current moment. The data storage module is used to adjust the storage scheme based on the storage priority index and access popularity index of each logical unit at the current moment; The step of analyzing a single user's access habits based on each user's access data in each logical unit and the path distance between every two logical units, and determining each user's access popularity index for each logical unit, includes: For each user, the access frequency of the user for each logical unit in each monitoring period is obtained based on the access data. The logical units that are frequently accessed are filtered out by the access frequency of each logical unit and recorded as the logical units to be analyzed. Based on the path distance between any two logical units to be analyzed and the access frequency corresponding to each logical unit to be analyzed, the correlation degree of high-frequency access data of the user on different terminal devices is determined; the correlation degree is inversely proportional to the path distance and directly proportional to the access frequency. Based on the access data, the frequency and duration of user access to each logical unit in the current monitoring period are obtained, and combined with the correlation degree, the access popularity index of user for each logical unit is determined. The method involves analyzing the group's access habits based on the number of interactions between each pair of users, the access popularity index, and the edge storage nodes of the two users to obtain various shared areas, including: The evaluation index for each pair of users belonging to the same shared area is determined based on the number of interactions between each pair of users, the number of corresponding identical high-heat logical units, the average access popularity index of the identical high-heat logical units, and the total number of logical units; the high-heat logical units are logical units whose access popularity index is greater than the set access popularity threshold. Set an evaluation threshold, and merge two users with the same edge storage node whose evaluation index is greater than the evaluation threshold into a shared area to obtain various shared areas.

8. A data storage medium for a cloud computer, characterized in that, It includes a processor and a memory, the processor being used to process instructions stored in the memory to implement a data storage management method for a cloud computer as described in any one of claims 1-6.

Images