Method and system for data migration and synchronization across cloud storage platforms

Abstract

The application provides a data migration and synchronization method and system across cloud storage platforms, relating to the technical field of cloud computing data management, wherein the method comprises: collecting access characteristic data of each data block in the cross-cloud storage platform, and then generating a heat map representing data access heat; converting the heat map into a multi-channel optical signal and reconstructing the light field intensity distribution; analyzing the light field intensity distribution data through a circuit control model to identify the hot area distribution characteristics; generating a data priority sequence based on the characteristics, converting the time sequence control signal using pulse width modulation technology, driving the circuit control system to accurately adjust the conduction state and transmission timing of the data migration channel, and finally realizing real-time synchronization and migration of hot data across cloud storage platforms. The application improves the real-time performance and consistency of cross-cloud hot data synchronization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud computing data management, and particularly relates to a cross-cloud storage platform data migration and synchronization method and system. BACKGROUND

[0002] With the rapid development of cloud computing technology, enterprises usually adopt multiple cloud storage platforms to meet different business needs. Cross-cloud data migration and synchronization has become a key link to ensure business continuity. Therefore, in application scenarios with high real-time requirements, such as multi-active disaster recovery and global business deployment, it is necessary to ensure that frequently accessed hot data maintains strong consistency among multiple cloud platforms while avoiding the impact of synchronization delay on business performance.

[0003] A current existing solution adopts a dynamic traffic scheduling mechanism based on software-defined networks. By monitoring data access frequency and network status, it combines pre-defined strategies to achieve data synchronization priority allocation. This solution uses a centralized controller to collect access state information of each node, and then dynamically adjusts data flow transmission paths according to real-time network bandwidth and delay, combined with access state information of each node, and allocates synchronization resources in a time slice round robin manner.

[0004] However, this solution has certain limitations in data processing, such as its priority judgment relying on periodic state collection and strategy calculation, and the decision-making process having a delay. Moreover, in the face of sudden access hotspots, the system response is not agile enough, which may lead to lag in hot data synchronization. At the same time, the scheduling mechanism at the software level results in large system overhead when processing large-scale concurrent requests, thereby affecting the overall synchronization efficiency. SUMMARY

[0005] The present application provides a cross-cloud storage platform data migration and synchronization method and system to solve the problems of poor real-time performance and low consistency in cross-cloud hot data synchronization in the prior art.

[0006] To solve the above technical problems, in a first aspect, the present application provides a cross-cloud storage platform data migration and synchronization method, comprising:

[0007] Obtaining access characteristic data of each data block in the cross-cloud storage platform;

[0008] Generating a hotness map based on the access characteristic data;

[0009] Converting the hotness map into a multi-channel optical signal, and reconstructing the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data;

[0010] Identifying hot spot area distribution characteristics from the light field intensity distribution data using a circuit control model;

[0011] According to the hotspot area distribution feature, a data priority sequence is generated, a time sequence control signal is generated based on the data priority sequence using a pulse width modulation technique, and the circuit control system adjusts the conduction state and transmission timing of the data migration channel using the time sequence control signal to realize real-time synchronization and migration of data across cloud storage platforms.

[0012] Optionally, the circuit control model is used to identify the hotspot area distribution feature from the light field intensity distribution data, including:

[0013] The light field intensity distribution data is input into a circuit control model, and an adjustment module of the circuit control model adjusts an energy detection threshold according to the statistical characteristics of the light field intensity distribution data;

[0014] An energy analysis of the light field intensity distribution data is performed based on the adjusted energy detection threshold using a scanning module of the circuit control model to form an energy distribution map;

[0015] An energy gradient distribution and a spatial aggregation degree of each target region in the energy distribution map are calculated using a feature extraction module of the circuit control model;

[0016] Based on the energy gradient distribution and the spatial aggregation degree, a hotspot area distribution feature is generated.

[0017] Optionally, the energy analysis of the light field intensity distribution data based on the adjusted energy detection threshold to form an energy distribution map includes:

[0018] The light field intensity distribution data is divided into a plurality of grid cells according to spatial coordinates, wherein each grid cell contains a fixed number of pixel points;

[0019] The energy accumulation values of all pixel points in each grid cell are calculated, and the energy accumulation values of adjacent grid cells are smoothed using a sliding window algorithm to generate smoothed energy accumulation values;

[0020] The smoothed energy accumulation values of each grid cell are compared with the adjusted energy detection threshold in real time, and based on the comparison result, all grid cells with smoothed energy accumulation values greater than the adjusted energy detection threshold are marked as initial regions, and the spatial position coordinates and energy total values of each initial region are recorded;

[0021] Based on the spatial position coordinates, the Euclidean distance between adjacent initial regions is calculated, and region merging processing is performed on adjacent initial regions with a Euclidean distance less than a preset distance threshold to generate boundary information of the target region;

[0022] Based on the boundary information of all target regions and the corresponding total energy values, an energy distribution atlas is constructed.

[0023] Optionally, the adjacent initial regions with the Euclidean distance less than the preset distance threshold are subjected to region merging processing to generate the boundary information of the target region, including:

[0024] The adjacent initial regions with the Euclidean distance less than the preset distance threshold are marked as a to-be-merged group;

[0025] The energy-weighted fusion processing is performed on all initial regions in each to-be-merged group to obtain the centroid coordinates of the target region;

[0026] According to the centroid coordinates of the target region, a boundary fitting algorithm is used to generate the boundary contour information of the target region;

[0027] Based on the boundary contour information, the boundary information of the target region is generated.

[0028] Optionally, the generating of the heat atlas based on the access feature data includes:

[0029] Based on the access frequency data, the temporal locality data and the spatial locality data in the access feature data, a heat index of each data block is generated;

[0030] Based on the heat indexes of all data blocks, spatial aggregation processing is performed to form a two-dimensional heat distribution map;

[0031] The two-dimensional heat distribution map is subjected to region boundary division to generate a heat atlas.

[0032] Optionally, the converting of the heat atlas into a multi-channel optical signal and the light field intensity distribution reconstruction of the multi-channel optical signal to form light field intensity distribution data include:

[0033] The heat atlas is input into an optical modulator, and an electro-optical conversion is performed through an optical modulation processing mode of the optical modulator to generate a multi-channel optical signal;

[0034] The multi-channel optical signal is subjected to phase modulation and amplitude modulation to obtain a reshaped multi-channel optical signal;

[0035] The reshaped multi-channel optical signal is subjected to wavefront reconstruction through a diffractive optical element of the optical modulator to obtain light field intensity distribution data.

[0036] Optionally, the generating of the data priority sequence based on the heat point region distribution characteristics and the generating of the time sequence control signal based on the data priority sequence using the pulse width modulation technology include:

[0037] generate a region level and an energy intensity parameter based on the hotspot region distribution feature;

[0038] generate a data priority sequence by using a priority calculation algorithm based on the region level and the energy intensity parameter;

[0039] input the data priority sequence into a pulse width modulation controller, and map the priority in the data priority sequence into a corresponding pulse modulation parameter by a mapping module in the pulse width modulation controller;

[0040] generate a corresponding baseband pulse signal based on the pulse modulation parameter by a timing generation module in the pulse width modulation controller;

[0041] perform power amplification and filtering processing on the baseband pulse signal to generate a timing control signal.

[0042] In a second aspect, the present application provides a data migration and synchronization system across cloud storage platforms, comprising:

[0043] an acquisition module configured to acquire access feature data of each data block in the cloud storage platform;

[0044] a generation module configured to generate a heat map based on the access feature data;

[0045] a conversion module configured to convert the heat map into a multi-channel optical signal, and perform light field intensity distribution reconstruction on the multi-channel optical signal to form light field intensity distribution data;

[0046] an identification module configured to identify a hotspot region distribution feature from the light field intensity distribution data by using a circuit control model;

[0047] a driving module configured to generate a data priority sequence based on the hotspot region distribution feature, generate a timing control signal by using a pulse width modulation technique based on the data priority sequence, and drive a circuit control system to adjust the conduction state and transmission timing of a data migration channel by using the timing control signal, so as to realize real-time synchronization and migration of data across cloud storage platforms.

[0048] In a third aspect, the present application provides an electronic device, comprising:

[0049] a memory configured to store a computer program;

[0050] a processor configured to execute the computer program to implement the steps of the data migration and synchronization method across cloud storage platforms according to the first aspect.

[0051] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the data migration and synchronization method across cloud storage platforms as described in the first aspect above.

[0052] This application provides a method for data migration and synchronization across cloud storage platforms. The method first acquires access characteristic data of each data block in the cross-cloud storage platform; then, it generates a heatmap based on the access characteristic data; next, it converts the heatmap into a multi-channel optical signal and reconstructs the optical field intensity distribution of the multi-channel optical signal to form optical field intensity distribution data; then, it uses a circuit control model to identify hotspot distribution characteristics from the optical field intensity distribution data; finally, it generates a data priority sequence based on the hotspot distribution characteristics, uses pulse width modulation (PWM) technology to generate a timing control signal based on the data priority sequence, and uses the timing control signal to drive a circuit control system to adjust the conduction state and transmission timing of the data migration channel, thereby achieving real-time synchronization and migration of data between cross-cloud storage platforms.

[0053] The technical solution provided in this application has the following beneficial effects:

[0054] This application first enables comprehensive collection of data access behavior, providing a complete data foundation for heat analysis; then, it transforms abstract access characteristics into a visual spatial distribution model, thus intuitively reflecting the distribution of data hotspots; next, it utilizes the characteristics of optical parallel processing to significantly improve data processing speed, forming a high-precision light field intensity distribution; then, it achieves rapid conversion from optical signals to electrical signals through a photoelectric conversion mechanism, accurately identifying the distribution of hotspot areas; then, based on the identification results, it transforms the priority strategy into precise level control signals, achieving fine-grained control of data synchronization timing; finally, it directly controls the on / off state of the data migration channel through electrical signals, ensuring the real-time performance and reliability of the thermal data synchronization process.

[0055] Furthermore, this application also adaptively sets the energy detection threshold through the adjustment module of the circuit control model, and completes the energy analysis of the light field data through the scanning module to form an energy distribution map. Then, the energy gradient and spatial aggregation degree are calculated through the feature extraction module, and finally hot spot area distribution data with spatial characteristics are generated.

[0056] Furthermore, the embodiments of this application can achieve accurate conversion of light field data into hotspot features, improve environmental adaptability through adaptive thresholds, and enhance the accuracy of hotspot identification by combining energy gradient and spatial aggregation degree analysis, thereby providing a reliable basis for synchronization priority decision-making.

[0057] These or other aspects of this application will become more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0059] Figure 1 A flowchart of a cross-cloud storage platform data migration and synchronization method provided by an embodiment of the present application;

[0060] Figure 2 A specific implementation schematic diagram of a cross-cloud storage platform data migration and synchronization method provided by an embodiment of the present application;

[0061] Figure 3 A structural schematic diagram of a cross-cloud storage platform data migration and synchronization system provided by an embodiment of the present application. DETAILED DESCRIPTION

[0062] In the cross-cloud storage platform data synchronization scenario, the existing dynamic traffic scheduling scheme based on software-defined network has the following obvious limitations: the priority judgment thereof depends on periodic collection of access state and network indicators, and a synchronization strategy is generated through centralized calculation, thereby causing delay in system response; moreover, when facing a sudden data access hotspot, the strategy update lags behind the actual demand, making it difficult to realize real-time synchronization of hot data; at the same time, the scheduling mechanism at the software level produces high system overhead when processing large-scale concurrent requests, affecting synchronization efficiency and leading to insufficient timeliness of hot data synchronization.

[0063] In view of the above problems, the present application proposes a cross-cloud storage platform data migration and synchronization method, which realizes accurate identification and real-time response of hot data by converting data access features into optical signals and performing high-speed processing; specifically, the method first generates a heat map of data access features, then converts the heat map into optical signals using optical modulation technology and reconstructs it into optical field intensity distribution data, and finally generates an accurate timing control signal based on pulse width modulation technology by quickly identifying the distribution characteristics of the hotspot area through a circuit control model. Therefore, the method can improve the processing speed of data through optical parallel processing, and realize real-time execution of the synchronization strategy through circuit-level control, effectively overcoming the problems of response delay and high system overhead of the existing scheme, thereby improving the timeliness and resource utilization efficiency of hot data synchronization.

[0064] For those skilled in the technical field, the present application will be further described in detail below in conjunction with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0065] The core of the present application is to provide a cross-cloud storage platform data migration and synchronization method, and a specific embodiment process diagram is shown in Figure 1 The method comprises the following steps:

[0066] Step 101: Obtain the access characteristic data of each data block in the cross-cloud storage platform.

[0067] In step 101, the access characteristic data refers to the original data reflecting the access characteristics of the data block collected from the cross-cloud storage platform. The access characteristic data includes access frequency data, temporal locality data and spatial locality data. The access frequency data represents the number of times the data block is accessed per unit time. The temporal locality data describes the time correlation characteristics of the repeated access of the data block in the near future. The spatial locality data reflects the spatial correlation characteristics of the continuous access of adjacent data blocks.

[0068] For example, taking a certain cross-cloud storage platform as an example, the monitoring agent collects the access records of 10,000 data blocks in 10 minutes, in which data block A is accessed 50 times, and the access frequency data is 50. Then the reciprocal average value of the time interval of the last 5 accesses of data block A is calculated as 0.2, and the temporal locality data is 0.2. Then the probability of continuous access of data block A and adjacent data blocks B and C is calculated as 0.8, and the spatial locality data is 0.8. The above three types of data together constitute the access characteristic data of data block A.

[0069] Step 102: Generate a heat map based on the access frequency data, the temporal locality data and the spatial locality data.

[0070] In step 102, the heat map is a two-dimensional distribution map formed by weighted fusion and spatial mapping of the access characteristic data, in which each position corresponds to a heat value of a data block, and the heat value reflects the access activity level of the data block.

[0071] In the embodiments of the present application, the three types of access characteristic data are first normalized, then the heat value of each data block is calculated by a weighted formula, and the heat value is equal to the access frequency data multiplied by the weight coefficient plus the temporal locality data multiplied by the weight coefficient plus the spatial locality data multiplied by the weight coefficient. Finally, the heat value is mapped to a two-dimensional coordinate system according to the actual distribution position of the data block in the storage space to form a heat map.

[0072] For example, the weight coefficients are set as 0.5 for access frequency data, 0.3 for time locality data, and 0.2 for space locality data; then the heat value of the data block A is calculated according to the weight coefficients, such as 50*0.5+0.2*0.3+0.8*0.2=25+0.06+0.16=25.22; finally, the heat values of the 10,000 data blocks are mapped to a 100*100 two-dimensional grid according to their physical storage addresses to generate a visual heat map.

[0073] Step 103: converting the heat map into a multi-channel optical signal, and reconstructing the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data.

[0074] In step 103, the multi-channel optical signal refers to an optical wave signal carrying heat information transmitted through multiple parallel optical channels, the light field intensity distribution reconstruction refers to adjusting the phase and amplitude of the optical wave signal to form a specific spatial intensity distribution, and the light field intensity distribution data refers to the intensity values of each point obtained through photoelectric conversion. The light field intensity distribution data can carry the access feature information in the heat map in the form of light field intensity distribution, providing an input carrier for subsequent photoelectric conversion and hot spot area identification.

[0075] In the embodiments of the present application, the heat map is first input into an optical modulator array to generate a multi-channel optical signal carrying part of the heat information through electro-optical conversion; then the multi-channel optical signal is waveform shaped by a phase modulator and an amplitude modulator; then the shaped light field is wavefront reconstructed by a diffractive optical element; and finally, the light field intensity distribution data is formed through an optical focusing system.

[0076] For example, a 100*100 pixel heat map is first input into an optical modulator array containing 8 modulation channels to generate 8 independent optical signals; then the optical signals are waveform adjusted by a phase modulator and an amplitude modulator in turn; then the modulated light field is wavefront reconstructed by a diffractive optical element to finally form a 1024*768 pixel optical pattern; and finally, the optical pattern is converted into corresponding digital intensity data by a photoelectric sensor, with the intensity value ranging from 0 to 255; wherein the high brightness area corresponds to the hot spot area in the heat map.

[0077] Step 104: identifying the hot spot area distribution feature from the light field intensity distribution data by using a circuit control model.

[0078] In step 104, the hot spot area distribution feature refers to the spatial position and intensity feature of the hot spot area identified from the light field data.

[0079] In the embodiments of the present application, first, the adjustment module in the circuit control model automatically adjusts the detection threshold according to the statistical characteristics of the intensity data, then the scanning module identifies the regions exceeding the threshold, and finally the feature extraction module calculates the energy gradient distribution and spatial aggregation degree of each region to generate the hot region distribution features.

[0080] For example, first, the photoelectric sensor converts the 1024x768 pixel optical pattern into corresponding digital intensity data with intensity value ranging from 0 to 255; then, the circuit control model calculates that the average intensity value is 120, the variance is 3600, and sets the detection threshold to 120 plus the calculation result of 2 times 60, which is 240; then, 15 regions exceeding the detection threshold are identified, and the energy gradient of each region is calculated; finally, the hot region distribution features containing the center coordinates and energy values of the regions are generated.

[0081] Step 105: According to the hot region distribution features, generate a data priority sequence, generate a timing control signal using pulse width modulation technology based on the data priority sequence, and use the timing control signal to drive the circuit control system to adjust the conduction state and transmission timing of the data migration channel to realize real-time synchronization and migration of data across cloud storage platforms.

[0082] In step 105, the data priority sequence refers to a list of data block synchronization order sorted by priority; the timing control signal refers to an electrical signal used to control the timing of the circuit switch; the circuit control system is deployed in the synchronization control unit across the cloud storage platform, used to perform scheduling management of the data migration channel; the data migration channel is deployed in the interconnection network between the cross-cloud storage platforms, used to realize data transmission between different cloud storage platforms.

[0083] In the embodiments of the present application, first, the data priority sequence is generated according to the regional energy intensity in the hot region distribution features, where high-energy regions are set as priority synchronization; then the priority in the data priority sequence is mapped to different pulse parameters by the pulse width modulation controller to generate a baseband pulse signal; then the baseband pulse signal is power amplified and noise filtered; finally, the timing control signal is output to drive the circuit control system, thereby controlling the conduction timing and transmission rate of the data migration channel.

[0084] For example, first, sort the 15 hot regions according to the energy value from high to low, and the region with the highest energy value has a priority of 1; then use the pulse width modulation controller to map the priority 1 to a pulse signal with a duty cycle of 80% and a frequency of 1 kHz; and after the pulse signal is power amplified, a 12V timing control signal is output to control the data migration channel to preferentially transmit the data block with priority 1.

[0085] The application realizes accurate identification and real-time synchronization of hot data across the cloud storage platform through the synergistic processing of optical and electrical technologies. First, the optical modulation technology is used to greatly improve the data processing speed, and then the circuit-level control is used to ensure the accurate execution of the synchronization strategy, thereby effectively solving the problems of response delay and large system overhead in the prior art, and finally improving the timeliness and system resource utilization efficiency of the hot data synchronization process.

[0086] To solve the problems of insufficient hot spot area recognition accuracy and response delay in the prior art, in some embodiments, step 104: the hot spot area distribution feature is identified from the light field intensity distribution data by using a circuit control model, as shown in Figure 2 The circuit control model includes:

[0087] Step 201: input the light field intensity distribution data into the circuit control model, and adjust the energy detection threshold value according to the statistical characteristics of the light field intensity distribution data through the adjustment module of the circuit control model.

[0088] In step 201, the statistical characteristics of the light field intensity distribution data are obtained by mathematical statistical analysis of the optical signal intensity value after photoelectric conversion, including but not limited to the maximum value, minimum value, average value, variance, and intensity distribution histogram feature of the intensity value. These statistical characteristics are used to represent the distribution rule and concentration trend of the light field intensity in the entire space range. The adjustment module refers to a circuit control unit that can automatically adjust the threshold value according to the input data characteristics. The energy detection threshold value refers to the critical intensity value used to distinguish the target area from the ordinary area.

[0089] In the embodiments of the application, after inputting the light field intensity distribution data into the circuit control model, the adjustment module first calculates the average value and variance of all intensity values, then calculates the initial threshold value according to the linear combination of the average value and the variance, and finally dynamically adjusts the threshold value according to the overall characteristics of the intensity distribution, and outputs the energy detection threshold value that adapts to the actual data characteristics.

[0090] Step 202: based on the adjusted energy detection threshold value, the energy distribution map is formed by performing energy analysis on the light field intensity distribution data through the scanning module of the circuit control model.

[0091] In step 202, the scanning module refers to a circuit unit that partitions the light field data; the energy distribution map refers to the map data formed after energy analysis, which can directly reflect the spatial position and intensity characteristics of the energy concentration area in the light field intensity distribution. The map contains the boundary information, spatial coordinate distribution, and energy intensity value of the target area, providing a data basis for subsequent feature extraction and hot spot area identification.

[0092] In the embodiment of the present application, the scanning module first divides the light field intensity distribution data into grid cells according to a fixed size, calculates the energy accumulation value of all points in each cell, then uses a sliding window to smooth the energy values of adjacent cells, and finally compares the smoothed energy values with the adjusted energy detection threshold, marks all cells exceeding the threshold and records their spatial positions and energy values, forming an energy distribution atlas.

[0093] Step 203: Calculate the energy gradient distribution and spatial aggregation degree of each target region in the energy distribution atlas through the feature extraction module of the circuit control model.

[0094] In step 203, the target region is the cell whose energy value exceeds the energy detection threshold, and the feature extraction module refers to the circuit unit that extracts feature parameters from the energy distribution data; the energy gradient distribution refers to the rate of change of energy value in the spatial direction; and the spatial aggregation degree refers to the concentration degree of energy within the target region.

[0095] In the embodiment of the present application, the feature extraction module first calculates the energy gradient of each target region in the horizontal and vertical directions, obtains the gradient distribution graph through difference calculation, then calculates the variance value of energy within each region as the spatial aggregation degree index, and finally outputs the gradient distribution and aggregation degree index as the region features.

[0096] Step 204: Generate hot spot region distribution features based on the energy gradient distribution and the spatial aggregation degree.

[0097] In the embodiment of the present application, based on the energy gradient distribution and spatial aggregation degree data, a feature description containing the center coordinates, boundary vertices, total energy value and distribution characteristics is generated for each target region, and finally a complete hot spot region distribution feature data set is formed.

[0098] The following is a specific example:

[0099] After obtaining the digital intensity data corresponding to the optical pattern of 1024x768 pixels, the intensity value range of the digital intensity data is 0 to 255; first, input the light field intensity distribution data into the circuit control model, calculate the statistical features of all 1024x768 intensity values by the adjustment module, including the maximum value 255, the minimum value 0, the average value 120 and the variance 3600, wherein the average value is calculated by the formula of the sum of all intensity values divided by the total number of pixels, and the variance is calculated by the formula of the sum of the square of the difference between each intensity value and the average value divided by the total number of pixels; then adjust the energy detection threshold according to the statistical features, use the formula of threshold equal to average value plus 2 times standard deviation, wherein the standard deviation is the square root of variance, i.e. , the final calculation of the energy detection threshold is ;

[0100] The scanning module first performs energy analysis on the light field intensity distribution data based on the adjusted energy detection threshold 240, divides 1024*768 pixels into 512*384 grid cells, each of which contains 2*2 pixels, then calculates the energy accumulation value of 4 pixels in each grid cell, for example, the 4 pixel values of a certain cell are 230, 245, 238, and 242, respectively, then ; subsequently, a 3*3 sliding window is used to smooth the energy accumulation values of adjacent grid cells, the smoothing formula is that the smoothing value is equal to the center cell value multiplied by 0.2 plus the sum of the values of the adjacent 8 cells multiplied by 0.8 divided by 8; then all the grid cells with smoothed energy values greater than the adjusted energy detection threshold 240 are marked as target regions, a total of 856 candidate cells are marked; finally, the center coordinates and energy total value of each candidate cell are recorded to form a complete energy distribution map;

[0101] The feature extraction module first calculates the energy gradient distribution and spatial aggregation degree of each target region in the energy distribution map; for a target region with center coordinates [100, 200], the adjacent pixel value difference method is used to calculate the horizontal direction gradient, that is, the right pixel value is subtracted from the left pixel value, and the adjacent pixel value difference method is also used to calculate the vertical direction gradient, that is, the lower pixel value is subtracted from the upper pixel value; finally, the gradient distribution vector [15, 20] of the target region is obtained.

[0102] Then, the spatial aggregation degree calculation method is used to obtain the variance of all pixel values in the target region, which is 85; then, based on the energy gradient distribution data and the spatial aggregation degree data, the hot spot region distribution feature is generated, which includes the center coordinate set, the energy gradient matrix, and the aggregation degree list of 15 target regions, for example, the feature data of region 1 is the center coordinates [100, 200], the gradient vector [15, 20], and the aggregation degree value 85; finally, the extraction process of the hot spot region distribution feature is completed.

[0103] In the embodiments of the present application, through the multi-module collaborative processing of the circuit control model, the fast analysis of the light field intensity distribution data and the accurate identification of the hot spot region are first realized; the adjustment module ensures the adaptability of the system to environmental changes, and the scanning module and the feature extraction module provide accurate spatial feature description; finally, a reliable basis is provided for subsequent data synchronization priority decision, thereby comprehensively improving the accuracy and real-time performance of the hot spot region identification process.

[0104] In order to further improve the accuracy and spatial continuity of the hot spot region identification, in some embodiments, step 202: based on the adjusted energy detection threshold, the energy analysis is performed on the light field intensity distribution data to form an energy distribution map, comprising:

[0105] Step 301: divide the light field intensity distribution data into a plurality of grid cells according to spatial coordinates, wherein each grid cell contains a fixed number of pixel points.

[0106] In step 301, the spatial coordinates refer to the position identifier of each pixel point in the light field intensity distribution data; uniform distribution refers to forming regularly arranged grid cells on the spatial coordinates according to fixed grid size and equal interval division method, each grid cell has the same size and equal interval in the horizontal and vertical directions, ensuring that the entire light field region is completely and non-overlappingly covered; the grid cell refers to a rectangular region divided according to a fixed size.

[0107] In the embodiments of the present application, first, the entire data region is divided into rectangular grid cells of the same size according to the spatial coordinate range of the light field intensity distribution data; then a fixed number of adjacent pixel points are set in each grid cell; finally, it is ensured that all data regions are completely covered and there is no overlap between the grid cells.

[0108] Step 302: calculate the energy accumulation value of all pixel points in each grid cell, and smooth the energy accumulation values of adjacent grid cells by a sliding window algorithm to generate smoothed energy accumulation values.

[0109] In step 302, the energy accumulation value refers to the sum of the intensity values of all pixel points in the grid cell.

[0110] In the embodiments of the present application, first, the sum of the intensity values of all pixel points in each grid cell is calculated to obtain the energy accumulation value of each grid cell; then a sliding window method is used to perform weighted average processing on the energy accumulation values of adjacent grid cells, wherein the weight of the center cell is the highest, and the weights of the surrounding cells decrease in turn; finally, the smoothed energy accumulation values are generated.

[0111] Step 303: real-time dynamic comparison of the smoothed energy accumulation value of each grid cell with the adjusted energy detection threshold value, according to the comparison result, marking all grid cells with smoothed energy accumulation values greater than the adjusted energy detection threshold value as initial regions, and recording the spatial position coordinates and energy total value of each initial region.

[0112] In step 303, real-time dynamic comparison refers to immediate comparison of the smoothed energy value of each grid cell with the threshold value; the initial region refers to the grid cell with the smoothed energy value exceeding the threshold value; the spatial position coordinates refer to the center position coordinates of the grid cell in the light field; the energy total value refers to the sum of the intensity values of all pixel points in the grid cell.

[0113] In the embodiments of the present application, firstly, the smoothed energy accumulation value of each grid cell is compared with the adjusted energy detection threshold in real time; then all grid cells exceeding the adjusted energy detection threshold are marked as initial regions; and the center coordinates and total energy value of each initial region are recorded.

[0114] Step 304: Based on the spatial position coordinates, the Euclidean distance between adjacent initial regions is calculated, and the region merging processing is performed on the adjacent initial regions with the Euclidean distance less than a preset distance threshold, to generate the boundary information of the target region.

[0115] In step 304, the Euclidean distance refers to the straight-line distance of two points in space; the preset distance threshold refers to the upper limit of the distance for judging whether the regions are adjacent; the target region refers to the merged region, and the boundary information refers to the contour boundary coordinate set of the target region.

[0116] In the embodiments of the present application, firstly, the Euclidean distance between the centers of adjacent initial regions is calculated according to the recorded spatial position coordinates; then the initial regions with the Euclidean distance less than a preset merging threshold are merged; subsequently, the center of mass coordinates and the total energy value of the merged target region are recalculated; and finally, the boundary information of the new target region is generated.

[0117] Step 305: Based on the boundary information of all target regions and the corresponding total energy value, an energy distribution map is constructed.

[0118] In the embodiments of the present application, based on the boundary information of all target regions and the corresponding total energy value, a distribution map containing the region contour, center position and energy intensity is constructed.

[0119] The following is a specific example:

[0120] After obtaining the marked 856 initial regions and their spatial position coordinates and total energy values, the Euclidean distance between adjacent initial regions is calculated based on the spatial position coordinates, wherein the Euclidean distance calculation formula is , wherein represents the distance unit pixel, represents the center coordinates of the first region, represents the center coordinates of the second region.

[0121] For example, the distance between the center coordinates [100, 200] of region A and the center coordinates [103, 202] of region B is calculated as , and the preset distance threshold is 5 pixels. Since 3.6 is less than 5, the two regions are merged, and the center of mass coordinates of the target region are calculated as , , wherein respectively represent the total energy value of the region, the total energy value of region A 955, the total energy value of region B 892, which are calculated ;

[0122] The total energy value of the target region is The same method is used to merge all adjacent regions with a Euclidean distance of less than 5 pixels, and a total of 15 target regions are obtained. Based on the boundary information of these regions, the set of polygon vertex coordinates of each region, such as the vertex coordinates of region 1 [98, 198], [102, 198], [102, 202], [98, 202], and the corresponding total energy value, an energy distribution atlas containing spatial distribution characteristics and energy characteristics is constructed, which accurately reflects the distribution of energy concentration regions in the light field data.

[0123] In the embodiments of the present application, the spatial continuity of hot spot region identification is effectively improved through energy analysis and region merging processing, thereby avoiding the problem of region fragmentation; at the same time, the generated distribution atlas accurately reflects the spatial distribution characteristics of the target region; and finally provides high-quality input data for the subsequent feature extraction process.

[0124] In order to further improve the accuracy of region merging processing and the integrity of boundary generation, in some embodiments, step 304: the adjacent initial regions with a Euclidean distance less than a preset distance threshold are subjected to region merging processing to generate boundary information of the target region, including:

[0125] Step 401: mark the adjacent initial regions with a Euclidean distance less than a preset distance threshold as a merging group.

[0126] In step 401, the merging group refers to a set of adjacent initial regions with a Euclidean distance less than a preset distance threshold and needing to be merged.

[0127] In the embodiments of the present application, all adjacent initial regions with a distance less than a preset distance threshold are grouped and marked according to the calculated Euclidean distance results, obtaining a plurality of merging groups, wherein each merging group contains two or more initial regions needing to be merged.

[0128] Step 402: perform energy weighted fusion processing on all initial regions in each merging group to obtain the centroid coordinates of the target region.

[0129] In step 402, the centroid coordinates refer to the center position coordinates of the target region.

[0130] In the embodiments of the present application, for the initial regions in each merging group, the weighted average value of the center coordinates of each initial region is calculated with the total energy value of the initial region as the weight coefficient, and the centroid coordinates of the target region are obtained.

[0131] Step 403: generating the boundary contour information of the target region according to the centroid coordinate of the target region by using a boundary fitting algorithm.

[0132] In step 403, the boundary contour information refers to a continuous coordinate sequence describing the outer boundary of the region.

[0133] In the embodiment of the present application, the minimum convex hull algorithm is used to calculate the minimum convex polygon containing all the original region boundary points according to the centroid coordinate of the target region, and a smooth boundary contour coordinate sequence is generated.

[0134] Step 404: generating the boundary information of the target region based on the boundary contour information.

[0135] In the embodiment of the present application, the polygon vertex coordinates are first extracted based on the boundary contour information, and then the coordinate points are connected in the order of the vertices, and the boundary information containing all the boundary point coordinates and the connection relationship is generated.

[0136] The following is a specific example:

[0137] This embodiment continues the above embodiment, after identifying the region A center coordinate [100, 200], the total energy value 955, and the region B center coordinate [103, 202], the total energy value 892, and the Euclidean distance 3.6 pixels less than the preset distance threshold 5 pixels, the two regions are marked as a to-be-merged group, and the energy weighted fusion processing is performed on all initial regions in each to-be-merged group, wherein the centroid coordinate calculation formula is , wherein and represent the unit pixel of the centroid coordinate of the target region, represents the region A center coordinate, represents the region B center coordinate, represents the region A total energy value 955, represents the region B total energy value 892, and the calculation result is ,

[0138] ​Then, according to the centroid coordinates [101.4, 200.9] of the target region, a boundary fitting algorithm is used to generate the boundary contour information of the target region. Specifically, the minimum convex hull algorithm is used to process the original boundary point set of the two regions, which includes the boundary points [98, 198], [102, 198], [102, 202], [98, 202] of region A and the boundary points [101, 201], [105, 201], [105, 205], [101, 205] of region B. The new boundary contour point sequence is calculated as [98, 198], [102, 198], [105, 201], [105, 205], [101, 205], [98, 202];

[0139] Then, based on the boundary contour information, the boundary information of the target region is generated, which includes the boundary point coordinate sequence and the connection relationship between the points. Finally, the complete boundary information of the target region is formed, which together with the total energy value 1847 of the target region constitutes the feature data of the new target region.

[0140] In the embodiments of the present application, the region merging process effectively solves the problem of region fragmentation; then the energy weighted fusion method ensures the accuracy of the target region position; then the boundary fitting algorithm is used to generate a complete and smooth boundary contour; and finally the boundary information formed provides an accurate spatial description basis for subsequent distribution map construction.

[0141] To further improve the accuracy and visualization effect of the heat map generation, in some embodiments, step 102: generating a heat map based on the access feature data, includes:

[0142] Step 501: based on the access frequency data, time locality data and spatial locality data in the access feature data, generate a heat index of each data block.

[0143] In step 501, the frequency quantization value refers to the numerical value of the access frequency data after standardization processing; the time correlation value refers to the numerical value of the time locality data after normalization processing; the spatial correlation value refers to the numerical value of the spatial locality data after normalization processing; and the heat index refers to the comprehensive evaluation value reflecting the access heat degree of the data block.

[0144] In the embodiment of the present application, firstly, the collected access frequency data, time locality data and space locality data are quantitatively processed, and original data of different dimensions are mapped into the same numerical range to obtain corresponding frequency quantitative values, time correlation values and space correlation values; then the frequency quantitative values, time correlation values and space correlation values are weighted and summed according to preset weight coefficients to generate a heat index of each data block, wherein the heat index comprehensively reflects the access heat degree of the data block.

[0145] Step 502: performing spatial aggregation processing based on the heat indexes of all data blocks to form a two-dimensional heat distribution map.

[0146] In step 502, the two-dimensional heat distribution map refers to a heat distribution map formed by mapping the heat index to a two-dimensional space.

[0147] In the embodiment of the present application, firstly, the heat index value of each data block is mapped to the corresponding two-dimensional coordinate position according to the actual distribution position of all data blocks in the storage space; then a continuous heat distribution map is formed through interpolation processing.

[0148] Step 503: dividing the two-dimensional heat distribution map to generate a heat atlas.

[0149] In the embodiment of the present application, firstly, the two-dimensional heat distribution map is subjected to region segmentation processing; then different heat regions are divided according to the spatial distribution characteristics of the heat values, and a heat atlas containing region boundary information and heat value distribution is generated.

[0150] The following is a specific example:

[0151] After obtaining the access frequency data 50, time locality data 0.2 and space locality data 0.8 of data block A, firstly, the three types of data are subjected to quantitative mapping processing, wherein the access frequency data directly adopts the original value to obtain the frequency quantitative value 50, the time locality data is multiplied by 100 to obtain the time correlation value 20, and the space locality data is multiplied by 100 to obtain the space correlation value 80;

[0152] Then the frequency quantitative value, the time correlation value and the space correlation value are weighted and fused, and the formula heat index equals to frequency quantitative value multiplied by weight coefficient plus time correlation value multiplied by weight coefficient plus space correlation value multiplied by weight coefficient is adopted, wherein the weight coefficients are 0.5, 0.3 and 0.2 respectively, and the heat index of data block A is calculated as 50*0.5+20*0.3+80*0.2=25+6+16=47;

[0153] Based on the heat index of all 10000 data blocks, spatial aggregation processing is performed, and the data block physical storage address is mapped to a 100*100 two-dimensional grid. Each grid point corresponds to a heat index value of a data block. A bilinear interpolation algorithm is used to fill the values between the grid points to form a continuous two-dimensional heat distribution map.

[0154] Finally, the two-dimensional heat distribution map is divided into region boundaries. Using a region growing algorithm with a heat threshold of 30, adjacent grid points with a heat value greater than 30 are aggregated into the same region. A total of 15 hot spot regions are identified, and a complete heat map containing the boundary coordinates and heat values of each region is generated. The boundary coordinates of the region 1 are represented as a polygon vertex sequence, such as [10, 15], [10, 20], [15, 20], [15, 15]. The heat value is the average value of all grid points in the region, which provides input data for subsequent optical modulation processing.

[0155] In the embodiments of the present application, the comparability and comprehensive evaluation effect of different feature data are ensured through quantitative mapping and weighted fusion processing, and then an intuitive heat distribution map is generated through spatial aggregation and region division, which provides accurate data input for subsequent optical processing, thereby effectively improving the accuracy and visualization of hot data recognition.

[0156] To further improve the accuracy and efficiency of optical processing, in some embodiments, step 103: converting the heat map into a multi-channel optical signal and reconstructing the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data, includes:

[0157] Step 601: input the heat map into an optical modulator, and perform electro-optical conversion through the optical modulation processing mode of the optical modulator to generate a multi-channel optical signal.

[0158] In step 601, the optical modulator refers to a device that converts an electrical signal into an optical signal, and the electro-optical conversion refers to the process of converting an electrical signal into an optical signal.

[0159] In the embodiments of the present application, the heat map is first input into an optical modulator array. Then, the heat value of each pixel is converted into a corresponding optical intensity signal through the electro-optical conversion unit in the array, obtaining a plurality of parallel transmission optical channel signals, wherein each optical channel signal carries part of the heat information.

[0160] Step 602: phase modulation and amplitude modulation are performed on the multi-channel optical signal to obtain a shaped multi-channel optical signal.

[0161] In step 602, phase modulation refers to a modulation method that changes the phase parameter of the light wave, and amplitude modulation refers to a modulation method that changes the amplitude parameter of the light wave.

[0162] In the embodiment of the present application, by adjusting the phase angle and amplitude value of the light wave, the light signal waveform can be more regular and smooth, and a multi-channel light signal with optimized waveform can be obtained.

[0163] Step 603: performing wavefront reconstruction on the reshaped multi-channel light signal by a diffractive optical element of the optical modulator to obtain light field intensity distribution data.

[0164] In step 603, the diffractive optical element refers to an optical device that uses diffraction effect to regulate the light field.

[0165] It should be understood that this step can first obtain a light field mode through wavefront reconstruction, and then perform focusing processing on the light field mode to generate light field intensity distribution data corresponding to the optical mode.

[0166] Wherein, the above-mentioned light field mode refers to a light field form with a specific spatial intensity distribution, and the focusing processing refers to an optical processing of concentrating light field energy to a specific region, wherein the specific region refers to a data region with a clear boundary range formed in the heat map through region boundary division processing, and the construction process can be: using a region growing algorithm based on heat value distribution characteristics to aggregate adjacent grid points with heat values exceeding a set threshold into continuous regions, and generating a polygon boundary coordinate sequence through a boundary fitting algorithm, and then obtaining the spatial range of each region.

[0167] In the embodiment of the present application, the diffractive optical element can be used to perform diffraction processing on the reshaped multi-channel light signal to adjust the propagation path and interference characteristics of the light wave, and then reconstruct the wavefront distribution of the light field to form a light field mode with clear spatial intensity distribution; and in the optical focusing processing of the light field mode, the lens group can be used to concentrate the light field energy to a specific region to improve the concentration degree of the light field intensity and generate a clear optical mode.

[0168] In the embodiment of the present application, first, multi-channel optical modulation and wavefront reconstruction processing are used to realize efficient conversion of heat information to light field distribution; then, phase modulation and amplitude modulation are used to optimize the quality of the light signal; at the same time, the diffractive optical element is used to ensure the accuracy of the light field distribution; and finally, the generated optical mode provides high-quality input data for subsequent photoelectric conversion and hot spot recognition.

[0169] In order to further improve the generation accuracy and response speed of the timing control signal, in some embodiments, step 105: generating a data priority sequence according to the hot spot region distribution characteristics, and generating a timing control signal based on the data priority sequence using pulse width modulation technology, includes:

[0170] Step 701: generating a region level and an energy intensity parameter based on the hotspot region distribution feature.

[0171] In step 701, the region level refers to the hierarchical identification of the hotspot region according to the energy intensity, and the energy intensity parameter refers to the quantitative value reflecting the size of the regional energy.

[0172] In the embodiments of the present application, first, based on the energy value information in the hotspot region distribution feature, the hotspot region is divided into different levels according to the preset energy threshold range; and the specific energy intensity value of each hotspot region is recorded.

[0173] Step 702: generating a data priority sequence by using a priority calculation algorithm based on the region level and the energy intensity parameter.

[0174] In the embodiments of the present application, first, according to the region level and the energy intensity parameter, a weighted scoring algorithm is used to calculate the priority score of each region; and then a data priority sequence is generated in the order from high to low according to the score.

[0175] Step 703: inputting the data priority sequence into a pulse width modulation controller, and mapping the priority in the data priority sequence to a corresponding pulse modulation parameter by a mapping module of the pulse width modulation controller.

[0176] In step 703, the mapping module refers to a functional unit for converting priority to modulation parameter, and the pulse modulation parameter refers to a parameter set for controlling the characteristics of the pulse waveform, including duty cycle, frequency and phase, etc.

[0177] In the embodiments of the present application, the data priority sequence is input into the pulse width modulation controller, and each priority value is converted into a corresponding pulse modulation parameter by the mapping module, wherein the duty cycle corresponding to the high priority is greater than the duty cycle corresponding to the low priority, and the frequency corresponding to the high priority is greater than the frequency corresponding to the low priority.

[0178] Step 704: generating a corresponding baseband pulse signal based on the pulse modulation parameter by a timing generation module in the pulse width modulation controller.

[0179] In step 704, the timing generation module refers to a basic functional unit for generating a pulse signal waveform. The baseband pulse signal refers to the original pulse signal without amplification processing.

[0180] In the embodiments of the present application, the corresponding baseband pulse signal is generated by the timing generation module according to the pulse modulation parameter, and the baseband pulse signal is a square wave signal with a specific duty cycle and frequency.

[0181] Step 705: power amplification and filtering processing is performed on the baseband pulse signal to generate a timing control signal.

[0182] In step 705, the driving capability refers to the timing control signal being able to effectively drive the power device in the circuit control system to provide sufficient voltage and current output, and ensure the electrical characteristics of the on-off state and transmission timing of the data migration channel.

[0183] In the embodiments of the present application, the baseband pulse signal is subjected to power amplification processing to improve the voltage and current output capability, and at the same time, high-frequency noise is removed through a filtering circuit to generate a pure timing control signal.

[0184] In the embodiments of the present application, the priority sequence generation and pulse modulation processing are first used to realize accurate conversion of the synchronization strategy to the control signal; then the power amplification and noise filtering are used to ensure the driving quality and stability of the signal, so that the pure timing control signal can accurately drive the data migration channel to realize the priority synchronization of hot data.

[0185] Figure 3 The structural diagram of a data migration and synchronization system across cloud storage platforms is provided in the embodiments of the present application, and the specific implementation part describes:

[0186] The acquisition module 31 is configured to acquire access characteristic data of each data block in the cross-cloud storage platform.

[0187] The generation module 32 is configured to generate a heat map based on the access characteristic data.

[0188] The conversion module 33 is configured to convert the heat map into a multi-channel optical signal, and reconstruct the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data.

[0189] The identification module 34 is configured to identify the hot spot area distribution feature from the light field intensity distribution data by using a circuit control model.

[0190] The driving module 35 is configured to generate a data priority sequence according to the hot spot area distribution feature, generate a timing control signal by using a pulse width modulation technology based on the data priority sequence, and drive the circuit control system to adjust the on-off state and transmission timing of the data migration channel by using the timing control signal, so as to realize real-time synchronization and migration of data between cross-cloud storage platforms.

[0191] The data migration and synchronization system across cloud storage platforms of the embodiments of the present application is used to implement the data migration and synchronization method across cloud storage platforms as described above, and thus the specific embodiments in the data migration and synchronization system across cloud storage platforms can be seen from the embodiments of the data migration and synchronization method of the cloud storage platforms as described above, and the specific embodiments can be referred to the description of the respective embodiments, which will not be repeated here.

[0192] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the data migration and synchronization method of the cloud storage platforms as described above when executing the computer program.

[0193] The present application also provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the data migration and synchronization method of the cloud storage platforms as described above when executed by a processor.

[0194] In an exemplary embodiment, the computer readable storage medium as described above can include, but is not limited to, a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0195] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program implements the steps of the data migration and synchronization method of the cloud storage platforms as described above when executed by a processor.

[0196] The skilled person can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0197] The data migration and synchronization method and system of the cloud storage platforms as provided by the present application are described in detail above. The principles and implementation of the present application are described by specific examples in this paper, and the above description of the examples is only used to help understand the method and its core idea of the present application. It should be noted that for ordinary skilled persons in the technical field, without departing from the principles of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for data migration and synchronization across cloud storage platforms, characterized in that, include: Obtain access characteristic data for each data block across cloud storage platforms; Based on the access feature data, a heat map is generated. The heat map is a two-dimensional distribution map formed by weighted fusion and spatial mapping of the access feature data. Each position corresponds to the heat value of a data block, and the heat value reflects the access activity of the data block. The heat map is converted into a multi-channel optical signal, and the light field intensity distribution of the multi-channel optical signal is reconstructed to form light field intensity distribution data. The distribution characteristics of hotspot regions are identified from the light field intensity distribution data using a circuit control model; Based on the distribution characteristics of the hotspot areas, a data priority sequence is generated. Based on the data priority sequence, a timing control signal is generated using pulse width modulation technology. The timing control signal is then used to drive the circuit control system to adjust the conduction state and transmission timing of the data migration channel, so as to realize real-time synchronization and migration of data between cloud storage platforms. The data priority sequence refers to a list of data block synchronization order sorted according to priority. The method of identifying hotspot region distribution characteristics from the light field intensity distribution data using a circuit control model includes: The light field intensity distribution data is input into the circuit control model, and the energy detection threshold is adjusted by the adjustment module of the circuit control model according to the statistical characteristics of the light field intensity distribution data. The scanning module of the circuit control model performs energy analysis on the light field intensity distribution data based on the adjusted energy detection threshold to form an energy distribution map. The energy gradient distribution and spatial aggregation degree of each target region in the energy distribution map are calculated using the feature extraction module of the circuit control model. Based on the energy gradient distribution and the spatial aggregation degree, hotspot region distribution characteristics are generated; The step of generating a data priority sequence based on the distribution characteristics of the hotspot areas, and generating a timing control signal based on the data priority sequence using pulse width modulation technology, includes: Based on the distribution characteristics of the hotspot areas, regional level and energy intensity parameters are generated; Based on the region level and the energy intensity parameter, a data priority sequence is generated using a priority calculation algorithm; The data priority sequence is input to the pulse width modulation controller, and the priority in the data priority sequence is mapped to the corresponding pulse modulation parameters through the mapping module of the pulse width modulation controller; The timing generation module in the pulse width modulation controller generates a corresponding baseband pulse signal based on the pulse modulation parameters. The baseband pulse signal is amplified and filtered to generate a timing control signal.

2. The method according to claim 1, characterized in that, The step of performing energy analysis on the light field intensity distribution data based on the adjusted energy detection threshold to form an energy distribution spectrum includes: The light field intensity distribution data is divided into multiple grid units according to spatial coordinates, wherein each grid unit contains a fixed number of pixels; Calculate the cumulative energy value of all pixels in each grid cell, and smooth the cumulative energy values ​​of adjacent grid cells using a sliding window algorithm to generate a smoothed cumulative energy value; The smoothed cumulative energy value of each grid cell is dynamically compared with the adjusted energy detection threshold in real time. Based on the comparison results, all grid cells with smoothed cumulative energy values ​​greater than the adjusted energy detection threshold are marked as initial regions, and the spatial coordinates and total energy value of each initial region are recorded. Based on the spatial coordinates, the Euclidean distance between adjacent initial regions is calculated. Adjacent initial regions with an Euclidean distance less than a preset distance threshold are merged to generate boundary information of the target region. An energy distribution map is constructed based on the boundary information of all target regions and their corresponding total energy values.

3. The method according to claim 2, characterized in that, The process of merging adjacent initial regions whose Euclidean distance is less than a preset distance threshold to generate boundary information of the target region includes: The adjacent initial regions whose Euclidean distance is less than a preset distance threshold are marked as groups to be merged; Energy-weighted fusion processing is performed on all initial regions in each group to be merged to obtain the centroid coordinates of the target region. Based on the centroid coordinates of the target region, a boundary fitting algorithm is used to generate the boundary contour information of the target region; Based on the boundary contour information, the boundary information of the target region is generated.

4. The method according to claim 1, characterized in that, The step of generating a heat map based on the access feature data includes: Based on the access frequency data, temporal locality data, and spatial locality data in the access feature data, a heat index is generated for each data block. Spatial aggregation is performed based on the heat index of all data blocks to form a two-dimensional heat distribution map; The two-dimensional heat distribution map is divided into regional boundaries to generate a heat map spectrum.

5. The method according to claim 1, characterized in that, The step of converting the heat map into a multi-channel optical signal and reconstructing the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data includes: The heat map is input to an optical modulator, and electro-optical conversion is performed through the optical modulation processing of the optical modulator to generate a multi-channel optical signal. The multi-channel optical signal is subjected to phase modulation and amplitude modulation to obtain a shaped multi-channel optical signal; The wavefront reconstruction of the shaped multi-channel optical signal is performed using the diffractive optical element of the optical modulator to obtain optical field intensity distribution data.

6. A data migration and synchronization system across cloud storage platforms, characterized in that, include: The acquisition module is used to acquire access characteristic data of each data block in the cross-cloud storage platform; The generation module is used to generate a heat map based on the access feature data. The heat map is a two-dimensional distribution map formed by weighted fusion and spatial mapping of the access feature data. Each position corresponds to the heat value of a data block, and the heat value reflects the access activity of the data block. The conversion module is used to convert the heat map into a multi-channel optical signal and reconstruct the light field intensity distribution of the multi-channel optical signal to form light field intensity distribution data. The identification module is used to identify the distribution characteristics of hotspot regions from the light field intensity distribution data using a circuit control model; The driving module is used to generate a data priority sequence based on the distribution characteristics of the hotspot area, generate a timing control signal based on the data priority sequence using pulse width modulation technology, and use the timing control signal to drive the circuit to control the system to adjust the conduction state and transmission timing of the data migration channel, so as to realize real-time synchronization and migration of data between cloud storage platforms. The data priority sequence refers to the synchronization order list of data blocks sorted according to priority. The method of identifying hotspot region distribution characteristics from the light field intensity distribution data using a circuit control model includes: The light field intensity distribution data is input into the circuit control model, and the energy detection threshold is adjusted by the adjustment module of the circuit control model according to the statistical characteristics of the light field intensity distribution data. The scanning module of the circuit control model performs energy analysis on the light field intensity distribution data based on the adjusted energy detection threshold to form an energy distribution map. The energy gradient distribution and spatial aggregation degree of each target region in the energy distribution map are calculated using the feature extraction module of the circuit control model. Based on the energy gradient distribution and the spatial aggregation degree, hotspot region distribution characteristics are generated; The step of generating a data priority sequence based on the distribution characteristics of the hotspot areas, and generating a timing control signal based on the data priority sequence using pulse width modulation technology, includes: Based on the distribution characteristics of the hotspot areas, regional level and energy intensity parameters are generated; Based on the region level and the energy intensity parameter, a data priority sequence is generated using a priority calculation algorithm; The data priority sequence is input to the pulse width modulation controller, and the priority in the data priority sequence is mapped to the corresponding pulse modulation parameters through the mapping module of the pulse width modulation controller; The timing generation module in the pulse width modulation controller generates a corresponding baseband pulse signal based on the pulse modulation parameters. The baseband pulse signal is amplified and filtered to generate a timing control signal.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the data migration and synchronization method across cloud storage platforms as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the data migration and synchronization method across cloud storage platforms as described in any one of claims 1 to 5.

Images