A delay optimization method based on cloud computing

Abstract

The present invention discloses a cloud computing-based latency optimization method, which is applied to a cloud computing-based latency optimization system. The method includes a data collection module, a loading management module, and a server management module. The method is characterized in that: the data collection module is used to collect user historical access data and user operating habits, and enter the data to be processed into the system; the loading management module is used to analyze the characteristics of user access data, and select corresponding data to be pre-loaded in the edge server based on the data characteristics; the server management module is used to optimize the load of the server processing node and the data synchronization between servers; the data collection module, the loading management module, and the server management module are communicatively connected to each other; the data collection module includes a data collection module and a data entry module; the data collection module is used to collect user historical access data. The present invention has the characteristics of reducing latency and improving efficiency.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a delay optimization method based on cloud computing. Background Art

[0002] With the continuous development of Internet technology, the Internet has become an important part of people's daily life and work. The Internet makes people's lives more convenient, but due to the large number of users, the server cannot provide efficient services. In order to further improve the response speed, edge servers are set up to reduce the data transmission distance to reduce the response delay. However, due to the limited resources of the edge server, when the number of user requests is too large, it cannot process user requests in time, which will cause data congestion and lead to a large delay in responding to user requests. Moreover, since the edge server can only store part of the user data, it needs to synchronize data with the central server, but the data transmission distance is long and the data volume is also large, resulting in low data synchronization efficiency. In addition, the edge server is still processing user requests while synchronizing data, resulting in reduced available resources and low synchronization efficiency. Therefore, it is very necessary to design a cloud computing-based latency optimization method to reduce latency and improve efficiency. Summary of the Invention

[0003] The purpose of the present invention is to provide a cloud computing-based latency optimization method to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a cloud computing-based latency optimization method, applied to a cloud computing-based latency optimization system, comprising a data collection module, a load management module, and a server management module, characterized in that: the data collection module is used to collect user historical access data and enter the data to be processed into the system; the load management module is used to analyze the characteristics of user access data and select corresponding data to be pre-loaded in the edge server based on the data characteristics; the server management module is used to optimize the load of server processing nodes and data synchronization between servers; the data collection module, the load management module, and the server management module are communicatively connected to each other;

[0005] The node optimization module includes a bandwidth adjustment submodule, a priority assessment submodule, and a node adjustment submodule. The bandwidth adjustment submodule is used to adjust the bandwidth allocation of processing nodes in the edge server. The priority assessment submodule is used to rate the priority of user access request processing. The node adjustment submodule is used to adjust the load balancing and collaborative processing of nodes in the edge server.

[0006] The data management module includes a path adjustment submodule and a data synchronization submodule. The path adjustment submodule is used to adjust the transmission path in data synchronization to reduce the data packet loss rate. The data synchronization submodule is used to improve the efficiency of data synchronization between servers.

[0007] According to the above technical solution, the data acquisition module includes a data collection module and a data entry module. The data collection module is used to collect user historical access data and user access data habits, and the data entry module is used to enter the user-side data to be processed into the system for processing.

[0008] According to the above technical solution, the loading management module includes a data relationship analysis module and a communication module. The data relationship analysis module is used to analyze the knowledge graph of user access data to obtain the user's access habits, and the communication module is used to send reminders to client users.

[0009] According to the above technical solution, the loading management module also includes a data preloading module, which is used to preload target data in the edge server according to the data relationship in the data knowledge graph.

[0010] According to the above technical solution, the server management module includes a node optimization module and a data management module. The node optimization module is used to adjust the load balancing of the processing nodes in the edge server, and the data management module is used to manage data synchronization and data security during data transmission.

[0011] According to the above technical solution, the delay optimization method mainly includes the following steps:

[0012] Step S1: The data collection module collects the user's historical access data records on the current edge server, and the data entry module enters the user's request data into the system;

[0013] Step S2: When the user terminal receives the user request, the system sends an electrical signal to start the node optimization module, which begins to analyze the data volume, priority level and node load rate of the node to be processed, and adjusts the node load according to the analysis results;

[0014] Step S3: When the edge server and the central server synchronize data, the data management module is started to adjust the transmission path and the efficiency of data synchronization between the servers;

[0015] Step S4: When the system responds to the user's request, the system starts the data relationship analysis module, begins to analyze the user's current request data and user access habits, retrieves the data in the data map according to the analysis results, preloads the data in the edge server, and responds to the user's access request according to the user's access request analysis results.

[0016] According to the above technical solution, step S2 further includes the following steps:

[0017] Step S21: Scan and identify the data volume of the data to be processed in the current edge server processing node. If the data volume of the current processing node is less than a first threshold, the data to be processed in the current processing node is transferred to the adjacent processing node through the virtual chain, the bandwidth allocated to the current processing node is recovered and the node is put into hibernation. If the data volume of the current processing node is greater than the first threshold and less than the second threshold, the node continues to process the data. If the data volume of the current processing node is greater than the second threshold, the bandwidth in the recovery pool is called to compensate the current processing node.

[0018] Step S22: retrieve the user request data, scan and identify the target data in the user request, identify the size of the target data, retrieve the remaining bandwidth of the current processing node, and calculate the time required for the current processing node to load the target data according to the formula Wherein, i=1,2,3......n, T represents the time required for the current processing node to load the target data, M represents the number of bytes of the current target data, α represents the influence coefficient of the number of data bytes on the loading time, q represents the loading speed conversion efficiency of the current bandwidth, D represents the remaining bandwidth of the current processing node, κ represents the influence coefficient of the current bandwidth loading speed, μ represents the influence coefficient of the transmission distance between the user client and the current edge server on the loading time, the processing nodes are sorted in ascending order according to the time of loading the target data, if the time of the processing node to load the target data is less than the first threshold, the processing node is marked as an alternative node, if the time of the processing node to load the target data is greater than the first threshold and less than the second threshold, the node is marked as a combined loading node, otherwise the remaining target nodes are discarded.

[0019] According to the above technical solution, step S22 further includes the following steps:

[0020] Step S221: retrieve the remaining number of CPU cores and memory of the current processing node, calculate the CPU efficiency and memory efficiency of the current processing node respectively through the formula, and perform weighted fusion of the CPU efficiency and memory efficiency to obtain the efficiency of the current processing node. C represents the remaining efficiency of the current processing node, S represents the remaining number of CPUs of the current processing node, W represents the total number of CPU cores, β represents the weight of the node's CPU efficiency, m represents the remaining memory, P represents the total memory of the node, and λ represents the weight of the node's memory efficiency. The candidate nodes are retrieved and their remaining efficiency is identified. If the remaining efficiency of the processing node is less than the system-set threshold, the candidate mark in the node is cleared and the node is blocked. Otherwise, the first-ranked processing node is selected from the remaining candidate nodes in a top-down order as the node to load the target data.

[0021] Step S222: When there is no selectable node among the candidate nodes, search for nodes according to the combined loading node mark, randomly combine the combined loading nodes, retrieve the number of bytes of the target data, and calculate the data loading time for each node combination according to the formula:

[0022]

[0023] Where, i = 1, 2, 3 ... n, j = 1, 2, 3 ... n, T 总 Indicates the data loading time for each node combination, T i represents the time for the current processing node to load the target data under the current weight, C represents the residual efficiency of the current processing node, γ i Indicates the influence coefficient of the current node's remaining efficiency on the loading time. If the time to load the target data is less than the system-set threshold, the weight of the current node combination and the number of data bytes currently allocated to each node are recorded. Otherwise, the current node combination is discarded, and the target data loaded by the combined nodes is integrated and transmitted to the user end through the virtual chain.

[0024] According to the above technical solution, step S3 further includes the following steps:

[0025] Step S31: Retrieve user request data, identify the path code assigned by the system in the user request data, and if the path code does not exist in the user request, clear the user request. Otherwise, the edge server assigns the user request to the target path node according to the path code, anchors the storage module of the target edge processor according to the code of the path node, and detects the memory remaining of the storage module of the target edge server. If the memory remaining of the current target server is less than the number of bytes of the target data, scan the neighboring edge servers, retrieve the target edge server with sufficient memory remaining, change the path code so that the transmission path passes through the target edge server, retrieve the authentication instruction in the user request, and construct a path to connect to the central server according to the authentication instruction and the path code. The central server loads the data to the target edge server through the constructed path, and the user accesses the data loaded by the target edge server through the path.

[0026] Step S32: When synchronizing data between servers, the master edge server sends a synchronization instruction to the target edge server through the communication module. The target edge server recognizes the synchronization instruction, starts the synchronization node according to the synchronization instruction, establishes a coordinate system, calculates the distance between adjacent synchronization nodes according to the distance formula, sorts the records in ascending order, selects the first-ranked synchronization node as the first target synchronization node of the current edge server, builds a virtual chain connecting the current synchronization node and the first target node, transmits data to the first target synchronization node through the virtual chain, and the first target synchronization node packages the data and sends it to the next target synchronization node. The last target is connected to the first target node to build a ring synchronization chain. Data synchronization is performed on the edge server through the ring synchronization chain, and the master edge server transmits the data to the central server through the virtual chain for data synchronization.

[0027] According to the above technical solution, in step S4, the user's historical access data is retrieved, the characteristics of the user's historical access data are identified, the target data accessed by the user is retrieved, the characteristics of the target data are identified, the data relationship map is retrieved according to the characteristics of the target data, the corresponding data is retrieved in the data relationship map according to the characteristics of the user's historical access data, and marked as target loading data, the target loading data is retrieved from the central server using the data preloading module, the target loading data is loaded in the edge server, cached in the storage module of the edge server, and the user request is responded to.

[0028] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, by recycling the bandwidth of some processing nodes and compensating for the large volume of data to be processed, can speed up the data transmission speed of the processing nodes, that is, reduce the loading time of data, greatly reduce the response delay of the system, and by calculating the time taken by the current processing node to load the target data, can quickly screen out the best processing node for loading the target data, thereby reducing the time the system takes to load the user's target data and further reducing the system response delay, and by screening the alternative nodes according to the remaining efficiency of the processing nodes, it can avoid the situation where the nodes do not have sufficient resources to load data, resulting in data congestion, and further reduce the system delay, and by weighted segmentation of the target data according to the remaining efficiency of the processing nodes, the segmented data are respectively sent to the According to the loading time within the node, the time spent on data by different node combinations is calculated respectively. The remaining efficiency of the node can be used to complete the loading of the target data, thereby avoiding data congestion caused by data waiting for resources, greatly reducing the waiting time for loading data, and greatly reducing the delay in system response. By adjusting the data transmission path, it can ensure that the target data requested by the user can be loaded, reducing the time waiting for resources for loading data. By starting the synchronization node to synchronize data and building a ring synchronization chain to synchronize data with the edge server, data synchronization can be performed without affecting the normal operation of the edge server, thereby greatly improving the data synchronization efficiency. By pre-loading the data that the user may access to the edge server, the loading time for the user to access the data can be reduced, thereby greatly reducing the delay in system response. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0030] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0032] See also Figure 1The present invention provides a technical solution: a cloud computing-based latency optimization method, which is applied to a cloud computing-based latency optimization system, including a data collection module, a load management module, and a server management module. The method is characterized in that: the data collection module is used to collect user historical access data and enter the data to be processed into the system; the load management module is used to analyze the characteristics of user access data and select corresponding data to be pre-loaded in the edge server based on the data characteristics; the server management module is used to optimize the load of server processing nodes and data synchronization between servers; the data collection module, the load management module, and the server management module are communicatively connected to each other;

[0033] The node optimization module includes a bandwidth adjustment submodule, a priority assessment submodule, and a node regulation submodule. The bandwidth adjustment submodule is used to adjust the bandwidth allocation of processing nodes in the edge server. The priority assessment submodule is used to rate the priority of user access request processing. The node regulation submodule is used to adjust the load balancing and collaborative processing of nodes in the edge server.

[0034] The data management module includes a path adjustment submodule and a data synchronization submodule. The path adjustment submodule is used to adjust the transmission path in data synchronization and reduce the data packet loss rate. The data synchronization submodule is used to improve the efficiency of data synchronization between servers.

[0035] The data acquisition module includes a data collection module and a data entry module. The data collection module is used to collect user historical access data and user access data habits. The data entry module is used to enter the user-side data to be processed into the system for processing.

[0036] The loading management module includes a data relationship analysis module and a communication module. The data relationship analysis module is used to analyze the knowledge graph of user access data to obtain the user's access habits, and the communication module is used to send reminders to client users.

[0037] The loading management module also includes a data preloading module, which is used to preload target data in the edge server according to the data relationship in the data knowledge graph.

[0038] The server management module includes a node optimization module and a data management module. The node optimization module is used to adjust the load balancing of the processing nodes in the edge server, and the data management node is used to manage data synchronization and data security during data transmission.

[0039] The latency optimization method mainly includes the following steps:

[0040] Step S1: The data collection module collects the user's historical access data records on the current edge server, and the data entry module enters the user's request data into the system;

[0041] Step S2: When the user terminal receives the user request, the system sends an electrical signal to start the node optimization module, which begins to analyze the data volume, priority level and node load rate of the node to be processed, and adjusts the node load according to the analysis results;

[0042] Step S3: When the edge server and the central server synchronize data, the data management module is started to adjust the transmission path and the efficiency of data synchronization between the servers;

[0043] Step S4: When the system responds to the user's request, the system starts the data relationship analysis module, begins to analyze the user's current request data and user access habits, retrieves the data in the data map according to the analysis results, preloads the data in the edge server, and responds to the user's access request according to the user's access request analysis results.

[0044] Step S2 further includes the following steps:

[0045] Step S21: Scan and identify the data volume of the data to be processed in the current edge server processing node. If the data volume of the current processing node is less than the first threshold, the data to be processed in the current processing node is transferred to the adjacent processing node through the virtual chain, the bandwidth allocated to the current processing node is recovered and the node is put into hibernation. If the data volume of the current processing node is greater than the first threshold and less than the second threshold, the node continues to process the data. If the data volume of the current processing node is greater than the second threshold, the bandwidth in the recovery pool is called to compensate the current processing node. By recovering the bandwidth of some processing nodes and compensating for the data with a large volume of data to be processed, the data transmission speed of the processing node can be accelerated, that is, the data loading time is reduced, and the response delay of the system is greatly reduced;

[0046] Step S22: retrieve the user request data, scan and identify the target data in the user request, identify the size of the target data, retrieve the remaining bandwidth of the current processing node, and calculate the time required for the current processing node to load the target data according to the formula Wherein, i=1,2,3......n, T represents the time required for the current processing node to load the target data, M represents the number of bytes of the current target data, α represents the influence coefficient of the number of data bytes on the loading time, q represents the loading speed conversion efficiency of the current bandwidth, D represents the remaining bandwidth of the current processing node, κ represents the influence coefficient of the current bandwidth loading speed, μ represents the influence coefficient of the transmission distance between the user client and the current edge server on the loading time, the processing nodes are sorted in ascending order according to the time of loading the target data, if the time of the processing node to load the target data is less than the first threshold, the processing node is marked as an alternative node, if the time of the processing node to load the target data is greater than the first threshold and less than the second threshold, the node is marked as a combined loading node, otherwise the remaining target nodes are discarded, by calculating the time the current processing node takes to load the target data, the best processing node for loading the target data can be quickly screened out, thereby reducing the time the system takes to load the user's target data and further reducing the system response delay.

[0047] Step S22 further includes the following steps:

[0048] Step S221: retrieve the remaining number of CPU cores and memory of the current processing node, calculate the CPU efficiency and memory efficiency of the current processing node respectively through the formula, and perform weighted fusion of the CPU efficiency and memory efficiency to obtain the efficiency of the current processing node. C represents the remaining efficiency of the current processing node, S represents the remaining number of CPUs of the current processing node, W represents the total number of CPU cores, β represents the weight of the node's CPU efficiency, m represents the remaining amount of memory, P represents the total amount of memory of the node, and λ represents the weight of the node's memory efficiency. The candidate nodes are retrieved and their remaining efficiency is identified. If the remaining efficiency of the processing node is less than the system-set threshold, the candidate mark in the node is cleared and blocked. Otherwise, the first-ranked processing node is selected from the remaining candidate nodes in a top-down order as the node for loading the target data. By screening the candidate nodes according to the remaining efficiency of the processing nodes, it is possible to avoid the situation where the nodes do not have sufficient resources to load data, resulting in data congestion, and further reduce system latency.

[0049] Step S222: When there is no selectable node among the candidate nodes, search for nodes according to the combined loading node mark, randomly combine the combined loading nodes, retrieve the number of bytes of the target data, and calculate the data loading time for each node combination according to the formula:

[0050]

[0051] Where, i = 1, 2, 3 ... n, j = 1, 2, 3 ... n, T 总Indicates the data loading time for each node combination, T i represents the time for the current processing node to load the target data under the current weight, C represents the residual efficiency of the current processing node, γ i It represents the influence coefficient of the current node's remaining efficiency on the loading time. If the time to load the target data is less than the system-set threshold, the weight of the current node combination and the number of data bytes currently allocated to each node are recorded. Otherwise, the current node combination is discarded, and the target data loaded by the combined nodes are integrated and transmitted to the user end through the virtual chain. The target data is weightedly divided according to the remaining efficiency of the processing node, and the loading time of the divided data in the corresponding nodes is calculated respectively. The time for different node combinations to load the data is calculated respectively, and the node remaining efficiency can be used to complete the target data loading, thereby avoiding data congestion caused by data waiting for resources, greatly reducing the waiting time for loading data, and greatly reducing the delay of system response.

[0052] Step S3 further includes the following steps:

[0053] Step S31: Retrieve user request data, identify the path code assigned by the system in the user request data, and if the path code does not exist in the user request, clear the user request. Otherwise, the edge server assigns the user request to the target path node according to the path code, anchors the storage module of the target edge processor according to the code of the path node, and detects the memory remaining of the storage module of the target edge server. If the memory remaining of the current target server is less than the number of bytes of the target data, scan the neighboring edge servers, retrieve the target edge server with sufficient memory remaining, change the path code so that the transmission path passes through the target edge server, retrieve the authentication instruction in the user request, and construct a path to connect to the central server according to the authentication instruction and the path code. The central server loads the data to the target edge server through the constructed path. The user accesses the data loaded by the target edge server through the path. By adjusting the data transmission path, it can be ensured that the target data requested by the user can be loaded, and the waiting time for loading data resources is reduced;

[0054] Step S32: When synchronizing data between servers, the main edge server sends a synchronization instruction to the target edge server through the communication module. The target edge server recognizes the synchronization instruction, starts the synchronization node according to the synchronization instruction, establishes a coordinate system, calculates the distance between adjacent synchronization nodes according to the distance formula, sorts the records in ascending order, selects the first-ranked synchronization node as the first target synchronization node of the current edge server, builds a virtual chain to connect the current synchronization node and the first target node, transmits data to the first target synchronization node through the virtual chain, and the first target synchronization node packages the data and sends it to the next target synchronization node. The last target is connected to the first target node to build a ring synchronization chain. Data synchronization is performed on the edge server through the ring synchronization chain. The main edge server transmits the data to the central server through the virtual chain for data synchronization. Data synchronization is performed by starting the synchronization node. A ring synchronization chain is built to synchronize data on the edge server. Data synchronization can be performed without affecting the normal operation of the edge server, thereby greatly improving the data synchronization efficiency.

[0055] In step S4, the user's historical access data is retrieved, the characteristics of the user's historical access data are identified, the target data accessed by the user is retrieved, the characteristics of the target data are identified, the data relationship map is retrieved according to the characteristics of the target data, the corresponding data is retrieved in the data relationship map according to the characteristics of the user's historical access data, and marked as target loading data, the target loading data is retrieved from the central server using the data preloading module, the target loading data is loaded in the edge server, cached in the storage module of the edge server, and the user request is responded to. By preloading the data that the user may access to the edge server, the loading time of the user access data can be reduced, thereby greatly reducing the delay of the system response.

[0056] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0057] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A cloud computing-based latency optimization method, applied to a cloud computing-based latency optimization system, comprising a data acquisition module, a load management module, and a server management module, characterized in that: The data collection module is used to collect user historical access data and user operation habits, and enter the data to be processed into the system. The loading management module is used to analyze the characteristics of user access data and select corresponding data based on the data characteristics to pre-load in the edge server. The server management module is used to optimize the load of server processing nodes and data synchronization between servers. The data collection module, loading management module and server management module are interconnected and communicated with each other; The server management module includes a node optimization module and a data management module. The node optimization module is used to adjust the load balancing of the processing nodes in the edge server, and the data management module is used to manage data synchronization and data security during data transmission; The node optimization module includes a bandwidth adjustment submodule, a priority assessment submodule, and a node adjustment submodule. The bandwidth adjustment submodule is used to adjust the bandwidth allocation of processing nodes in the edge server. The priority assessment submodule is used to rate the priority of user access request processing. The node adjustment submodule is used to adjust the load balancing and collaborative processing of nodes in the edge server. The data management module includes a path adjustment submodule and a data synchronization submodule. The path adjustment submodule is used to adjust the transmission path in data synchronization to reduce the data packet loss rate. The data synchronization submodule is used to improve the efficiency of data synchronization between servers. The data acquisition module includes a data collection module and a data entry module. The data collection module is used to collect user historical access data, and the data entry module is used to enter the user-side data to be processed into the system for processing; The loading management module includes a data relationship analysis module and a communication module. The data relationship analysis module is used to analyze the knowledge graph of user access data to obtain the user's access habits. The communication module is used to send reminders to client users. The loading management module also includes a data preloading module, which is used to preload target data in the edge server according to the data relationship in the data knowledge graph; The delay optimization method mainly includes the following steps: Step S1: The data collection module collects the user's historical access data records on the current edge server, and the data entry module enters the user's request data into the system; Step S2: When the user terminal receives the user request, the system sends an electrical signal to start the node optimization module, which begins to analyze the data volume, priority level and node load rate of the node to be processed, and adjusts the node load according to the analysis results; Step S3: When the edge server and the central server synchronize data, the data management module is started to adjust the transmission path and the efficiency of data synchronization between the servers; Step S4: When the system responds to the user's request, the system starts the data relationship analysis module, begins to analyze the user's current request data and user access habits, retrieves data from the data map based on the analysis results, preloads the data in the edge server, and responds to the user's access request based on the user's access request analysis results; The step S2 further comprises the following steps: Step S21: Scan and identify the data volume of the data to be processed in the current edge server processing node. If the data volume of the current processing node is less than a first threshold, the data to be processed in the current processing node is transferred to the adjacent processing node through the virtual chain, the bandwidth allocated to the current processing node is recovered and the node is put into hibernation. If the data volume of the current processing node is greater than the first threshold and less than the second threshold, the node continues to process the data. If the data volume of the current processing node is greater than the second threshold, the bandwidth in the recovery pool is called to compensate the current processing node. Step S22: retrieve the user request data, scan and identify the target data in the user request, identify the size of the target data, retrieve the remaining bandwidth of the current processing node, and calculate the time required for the current processing node to load the target data according to the formula Where i = 1, 2, 3, ..., n, T represents the time required for the current processing node to load the target data, M represents the number of bytes of the current target data, α represents the influence coefficient of the number of data bytes on the loading time, q represents the loading speed conversion efficiency of the current bandwidth, D represents the remaining bandwidth of the current processing node, κ represents the influence coefficient of the current bandwidth loading speed, and μ represents the influence coefficient of the transmission distance between the user client and the current edge server on the loading time. The processing nodes are sorted in ascending order according to the time to load the target data. If the time for a processing node to load the target data is less than a first threshold, the processing node is marked as a candidate node. If the time for a processing node to load the target data is greater than the first threshold and less than a second threshold, the node is marked as a combined loading node. Otherwise, the remaining target nodes are discarded. The step S22 further includes the following steps: Step S221: retrieve the remaining number of CPU cores and memory of the current processing node, calculate the CPU efficiency and memory efficiency of the current processing node respectively through the formula, and perform weighted fusion of the CPU efficiency and memory efficiency to obtain the efficiency of the current processing node. C represents the remaining efficiency of the current processing node, S represents the remaining number of CPUs of the current processing node, W represents the total number of CPU cores, β represents the weight of the node's CPU efficiency, m represents the remaining memory, P represents the total memory of the node, and λ represents the weight of the node's memory efficiency. The candidate nodes are retrieved and their remaining efficiency is identified. If the remaining efficiency of the processing node is less than the system-set threshold, the candidate mark in the node is cleared and the node is blocked. Otherwise, the first-ranked processing node is selected from the remaining candidate nodes in a top-down order as the node to load the target data. Step S222: When there is no selectable node among the candidate nodes, search for nodes according to the combined loading node mark, randomly combine the combined loading nodes, retrieve the number of bytes of the target data, and calculate the data loading time for each node combination according to the formula: Where, i = 1, 2, 3 ... n, j = 1, 2, 3 ... n, T 总 Indicates the data loading time for each node combination, T i represents the time for the current processing node to load the target data under the current weight, C represents the residual efficiency of the current processing node, γ i Indicates the influence coefficient of the current node's remaining efficiency on the loading time. If the time to load the target data is less than the system-set threshold, the weight of the current node combination and the number of data bytes currently allocated to each node is recorded. Otherwise, the current node combination is discarded, and the target data loaded by the combined nodes is integrated and transmitted to the user end through the virtual chain; The step S3 further comprises the following steps: Step S31: Retrieve user request data, identify the path code assigned by the system in the user request data, and if the path code does not exist in the user request, clear the user request. Otherwise, the edge server assigns the user request to the target path node according to the path code, anchors the storage module of the target edge processor according to the code of the path node, and detects the memory remaining of the storage module of the target edge server. If the memory remaining of the current target server is less than the number of bytes of the target data, scan the neighboring edge servers, retrieve the target edge server with sufficient memory remaining, change the path code so that the transmission path passes through the target edge server, retrieve the authentication instruction in the user request, and construct a path to connect to the central server according to the authentication instruction and the path code. The central server loads the data to the target edge server through the constructed path, and the user accesses the data loaded by the target edge server through the path. Step S32: When synchronizing data between servers, the master edge server sends a synchronization instruction to the target edge server through the communication module. The target edge server recognizes the synchronization instruction, starts the synchronization node according to the synchronization instruction, establishes a coordinate system, calculates the distance between adjacent synchronization nodes according to the distance formula, sorts the records in ascending order, selects the first synchronization node as the first target synchronization node of the current edge server, builds a virtual chain connecting the current synchronization node and the first target node, transmits data to the first target synchronization node through the virtual chain, and the first target synchronization node packages the data and sends it to the next target synchronization node. The last target is connected to the first target node to build a ring synchronization chain. Data synchronization is performed on the edge server through the ring synchronization chain. The master edge server transmits the data to the central server through the virtual chain for data synchronization; In step S4, the user's historical access data is retrieved, the characteristics of the user's historical access data are identified, the target data accessed by the user is retrieved, the characteristics of the target data are identified, the data relationship map is retrieved according to the characteristics of the target data, the corresponding data is retrieved in the data relationship map according to the characteristics of the user's historical access data, and marked as target loading data, the target loading data is retrieved from the central server using the data preloading module, the target loading data is loaded in the edge server, cached in the storage module of the edge server, and the user request is responded to.

Images