Transaction data synchronization management system and method for multiple payment platforms

Abstract

The present invention discloses a transaction data synchronization management system and method for multiple payment platforms, which relates to the field of data management technology. The system includes a network construction module, a condition judgment module and a data synchronization module. Based on the transaction network, the preferential form and preferential category of the user node are retrieved, the effect coefficient of the supply node is generated, and the transaction network is updated. In this process, the network environment parameters of the current transaction network are also combined to construct a four-dimensional monitoring vector, and the low-order circle structure and high-order circle structure of the node where the transaction network is located are determined to determine low impact or high impact. Under low impact, a batch data synchronization strategy is executed, and under high impact, a blockchain cross-chain synchronization strategy is executed to achieve comprehensive management of users, transactions and blockchains; in the blockchain cross-chain synchronization strategy, adaptive online sharding is used to achieve safe execution of cross-shard transactions off-chain.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a transaction data synchronization management system for multiple payment platforms. Background Art

[0002] Data transactions are a key link in the market-based allocation of data elements. When conducting transactions on different platforms, information such as payment orders, refund orders, and fund unfreezing must remain consistent. For example, refunds or write-offs require reverse processing to ensure that if consumers no longer meet the promotion conditions after a refund, the gifted coupons will be deducted or invalidated. In addition, a large amount of transaction data, such as payment amounts, settlement amounts, handling fees, discounts, etc., must be managed to ensure the consistency of transaction data across multiple platforms. With the continuous development of blockchain technology, transactions based on blockchain technology are also increasing.

[0003] On the one hand, blockchain sharding divides the blockchain into multiple smaller, more manageable components, or shards, enabling each shard to independently and concurrently process different sets of transactions. However, while this parallel processing architecture significantly increases the system's transaction throughput by distributing the computational load, the effective management of cross-shard transactions remains extremely inefficient, and the computational complexity and cost often lead to performance degradation, undermining the scalability advantages of sharding technology.

[0004] On the other hand, traditional payment processing methods require merchants to connect to various payment platforms separately. Not only is the process cumbersome and costly, but it is also difficult to form a unified user profile, limiting merchants' comprehensive understanding of consumer behavior and service optimization. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a transaction data synchronization management system and method for multiple payment platforms. Based on the transaction network, the preferential forms and preferential categories of user nodes are retrieved, the effect coefficient of the supply node is generated, and the transaction network is updated. In this process, the network environment parameters of the current transaction network are also combined to construct a four-dimensional monitoring vector to determine the low-order circle structure and high-order circle structure of the node where the transaction network is located to determine low impact or high impact. Under low impact, a batch data synchronization strategy is executed, and under high impact, a blockchain cross-chain synchronization strategy is executed, thereby realizing comprehensive management of users, transactions and blockchains, and solving the problems raised in the background technology.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] In a first aspect, the present application provides a transaction data synchronization management system for multiple payment platforms, the system comprising:

[0010] Network construction module: Deploy the transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management;

[0011] Conditional judgment module: Based on the transaction network, it retrieves the preferential form and category of the user node, creates the effect coefficient of the corresponding supply node, inputs the effect coefficient into the transaction network as an attribute matrix, updates the transaction network, and configures the data synchronization node to control each payment platform to transmit the transaction data set according to the optimal transaction path. It implements the batch data synchronization strategy under low-impact conditions and the cross-chain synchronization strategy under high-impact conditions.

[0012] Data synchronization module: updates the storage content of each data synchronization node, ensures data consistency among multiple nodes, and triggers the incremental synchronization process when a new transaction operation is detected.

[0013] Furthermore, the steps of deploying the transaction network include:

[0014] Treat each payment platform as a supply node and each user as a user node;

[0015] When a user interacts on any payment platform, an edge is built to construct a transaction network, which is represented as: Net (G, E, W, X);

[0016] Where G represents the supply node set, and G = {g1, g2, …, gi}, gi represents the i-th supply node, and i∈[1, num1], num1 is the total number of payment platforms, W represents the user node set, and W = {w1, w2, …, wj}, and j∈[1, num2], num2 is the total number of users; E indicates that the user has interactive behavior on the payment platform, and X represents the attribute matrix of all nodes.

[0017] Furthermore, the step of creating the effect coefficient of the corresponding supply node includes:

[0018] The preferential forms include no preferential treatment, unilateral preferential treatment, and bilateral preferential treatment, with the preferential form being assigned the first weight;

[0019] The preferential categories include at least one of instant discount coupons, discount coupons for purchases above a certain amount, points, event discounts, commission reductions, and data subsidies. The preferential categories are assigned the second weight.

[0020] Under the condition of no preferential treatment, it is directly marked as low impact;

[0021] Under the condition of unilateral preferential treatment, the corresponding first weight and second weight are weighted and summed to obtain the effect coefficient;

[0022] Under the condition of bilateral preferential treatment, the preferential categories under the same type of user nodes are collected, the corresponding first weight and second weight are multiplied, and the adjustment factor is introduced to weightedly correct the multiplication result, and the sum is calculated to obtain the effect coefficient.

[0023] Furthermore, the steps of determining low impact or high impact include:

[0024] Obtain the network environment parameters of the current transaction network, construct a four-dimensional monitoring vector, combine it with the effect coefficient of the current node, and perform weighted summation to obtain the circle ratio coefficient; where the network environment parameters include at least one of the transaction synchronization frequency, transaction throughput, and network congestion rate;

[0025] Compare and analyze the circle ratio coefficient with the preset first comparison interval and second comparison interval:

[0026] When the circle ratio coefficient is in the first comparison interval, a low-order circle structure is formed and it is judged to be low-impact. At the same time, the corresponding node combination is marked as the first transaction data set;

[0027] When the circle ratio coefficient is in the second comparison interval, a high-order circle structure is formed and it is determined to be high-impact. At the same time, the corresponding node combination is marked as the second transaction data set;

[0028] The first comparison interval is smaller than the second comparison interval.

[0029] Furthermore, the steps of executing the batch data synchronization strategy include:

[0030] Obtain the transmission link from the payment platform to the data synchronization node and calculate the first transaction data set of each transmission segment; divide the transmission segment into equal data blocks, mark the source identifier of the data point corresponding to each data block, and generate the data source;

[0031] During a preset time period T0, all data sources are collected and grouped into a set Si. Any b0 data sources from Si are randomly selected and partitioned into transaction tasks, generating F(a0, b0) transaction tasks. Transaction tasks are interactions initiated on branches and are obtained through permutations and combinations. a0 represents the number of data sources at any moment during time period T0, and data sources include at least one of click logs, add-to-purchase logs, payment logs, and refund logs. Payment logs have higher priority than refund logs, add-to-purchase logs, and click logs.

[0032] Calculate the corresponding deviations of the timestamps of b0 data sources in each transaction task, select several transaction tasks whose deviations are less than the standard deviation threshold, and obtain the corresponding data blocks;

[0033] The total number of synchronization batches is dynamically preset based on priority, and the processing order and time window of each batch are configured; when a data block enters the associated batch processing window, the corresponding first transaction data set is activated for transmission; when the data block processing is completed or times out, the transmission of the first transaction data set under the current batch is terminated and resources are released.

[0034] Furthermore, the steps of executing the blockchain cross-chain synchronization strategy include:

[0035] A delay mechanism is introduced, with a preset delay time period T1. The second transaction data set is collected and distributed to the shard group. Based on the data synchronization node, off-chain cross-shard transactions are executed. The exponential weighted algorithm is used to dynamically adjust the shard startup time slot, and the weight parameters corresponding to the shard are updated through the feedback mechanism.

[0036] Furthermore, the steps of implementing the feedback mechanism include:

[0037] First, the second transaction dataset is processed synchronously according to the batch data. The throughput and latency of the corresponding time slots in each epoch are retrieved, the corresponding weight parameters are dynamically assigned, and the difference is calculated to obtain the first evaluation coefficient to evaluate the alignment effect. The priority level is set according to the alignment effect, and the second transaction dataset is sorted.

[0038] Use the Metis algorithm to perform sharding, retrieve the initial weight parameters corresponding to the throughput and delay of the corresponding time slots in each epoch, and construct the maximum objective function max(β put , β rt ), to calculate the second evaluation coefficient, ; In the formula, ξ represents the second evaluation coefficient, β put represents the weight parameter corresponding to throughput, β rt Indicates the weight parameter corresponding to the delay;

[0039] The optimal time slot for starting the sharding process in each epoch is selected based on the second evaluation coefficient and the first evaluation coefficient, and the weight parameter is dynamically updated.

[0040] Furthermore, the step of dynamically assigning corresponding weight parameters includes:

[0041] Obtain the sorted historical second transaction data set and aggregate it into a first set, with the elements in the first set sorted from low to high priority; take a value from every third digit of the target set, and aggregate the taken values ​​into a second set;

[0042] A deep learning network is preset, and the second set is used as a test data set and input into the deep learning network for training and learning. The mean square error method is used to measure the linear relationship between the priority level and the weight, and the weight is automatically assigned to complete the dynamic allocation of weight parameters.

[0043] In a second aspect, the present application provides a method for synchronously managing transaction data for multiple payment platforms, the method comprising the following steps:

[0044] Deploy a transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management;

[0045] Based on the transaction network, the preferential forms and categories of user nodes are retrieved, and the effect coefficients of the corresponding supply nodes are created. The effect coefficients are input into the transaction network as attribute matrices, the transaction network is updated, and data synchronization nodes are configured to control each payment platform to transmit transaction data sets according to the optimal transaction path. Under low-impact conditions, batch data synchronization strategies are implemented, and under high-impact conditions, cross-chain synchronization strategies are implemented.

[0046] Update the storage content of each data synchronization node and ensure data consistency among multiple nodes, triggering the incremental synchronization process when a new transaction operation is detected.

[0047] (3) Beneficial effects

[0048] The present invention provides a transaction data synchronization management system and method for multiple payment platforms, which has the following beneficial effects:

[0049] 1. This invention creates a transaction network by treating each payment platform as a supply node and each user as a user node. Based on the transaction network, the preferential forms and preferential categories of the user nodes are retrieved to generate the effect coefficient of the supply node. The effect coefficient is used as an attribute matrix to update the transaction network. The user characteristics are indirectly inferred through the structure of this transaction network, maintaining the rigor and interpretability of the network model.

[0050] 2. This invention combines the network environment parameters of the current transaction network to construct a four-dimensional monitoring vector, which determines the low-order and high-order circle structures of the nodes in the transaction network. This determines whether the impact is low or high, and implements a batch data synchronization strategy under low impact conditions and a cross-chain synchronization strategy under high impact conditions, thereby achieving comprehensive management of users, transactions, and blockchains. As new transactions are added, the data synchronization strategy can be flexibly adjusted based on the current system status and the impact generated.

[0051] 3. This invention utilizes a cross-chain blockchain synchronization strategy to not only effectively respond to real-time changes in transaction data, but also implements adaptive sharding strategies in dynamic environments, significantly improving the overall performance and adaptability of the system. Adaptive online sharding utilizes data synchronization nodes to create a trusted execution environment, enabling secure off-chain execution of cross-shard transactions. This eliminates the need for multiple rounds of communication between shards, ensuring the secure execution of complex smart contract transactions without the overhead of on-chain re-verification. Furthermore, the trusted execution environment synchronizes the transaction status required by related shards through an efficient point-to-point communication mechanism, ensuring transaction accuracy and reliability.

[0052] 4. The present invention adopts a batch data synchronization strategy to configure the processing order and time window of each batch to avoid resource competition, and divides batches according to priority to intelligently allocate resources. It not only refines the monitoring granularity of each batch, but also speeds up processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a module diagram of a transaction data synchronization management system according to an exemplary embodiment;

[0054] Figure 2 is a flowchart illustrating execution of a synchronization strategy according to an exemplary embodiment;

[0055] Figure 3 The figure is a flowchart of a transaction data synchronization management method according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] The following will provide a clear and complete description of 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0057] Example 1:

[0058] The embodiment of the present invention provides a transaction data synchronization management system for multiple payment platforms; Figure 1 is a module diagram of a transaction data synchronization management system according to an exemplary embodiment; Figure 2 is a flowchart of executing a synchronization strategy according to an exemplary embodiment; Figures 1 to 2 The system is applied to multiple payment platforms and includes: a network construction module, a condition judgment module, and a data synchronization module, and the network construction module, the condition judgment module, and the data synchronization module are communicatively connected;

[0059] Network construction module: Deploy the transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management;

[0060] The steps to deploy a trading network include:

[0061] Each payment platform is considered as a supply node, and each user is considered as a user node. When a user interacts on any payment platform, an edge is established to construct a transaction network, which is represented as: Net (G, E, W, X);

[0062] Where G represents the set of supply nodes, and G = {g1, g2, …, gi}, g1, g2, and gi represent the first, second, and i-th supply nodes respectively, and i∈[1, num1], num1 is the total number of payment platforms, W represents the set of user nodes, and W = {w1, w2, …, wj}, and j∈[1, num2], num2 is the total number of users; E indicates that the user has interactive behavior on the payment platform, and X represents the attribute matrix of all nodes;

[0063] In addition, the transaction data set is obtained by collecting data from the payment platform through the server, and the server is set up with several clusters. The payment platform can freely add more data sources as the amount of access resources and visits increases. The transaction data set includes at least transaction time, transaction amount, transaction status and payment platform information. The transaction status includes success, failure and pending. The transaction status is mapped to the transaction network, with S01 indicating success, F01 indicating failure and P01 indicating pending.

[0064] The following is an explanation of the nouns involved:

[0065] User nodes: represent participants in the system who initiate transactions for cryptocurrency exchanges or trigger state transitions of smart contracts, such as merchants or consumers;

[0066] Conditional judgment module: Based on the transaction network, it retrieves the preferential form and category of the user node, creates the effect coefficient of the corresponding supply node, inputs the effect coefficient into the transaction network as an attribute matrix, updates the transaction network, and configures the data synchronization node to control each payment platform to transmit the transaction data set according to the optimal transaction path. It implements the batch data synchronization strategy under low-impact conditions and the cross-chain synchronization strategy under high-impact conditions.

[0067] The steps for creating the effect coefficient corresponding to the supply node include:

[0068] The preferential forms include no preferential treatment, unilateral preferential treatment, and bilateral preferential treatment, with the preferential form being assigned the first weight;

[0069] No discount: means no discount is provided to consumers or merchants;

[0070] Unilateral discount: refers to providing discounts to either consumers or merchants;

[0071] Bilateral discount: means that discounts are provided to both consumers and merchants;

[0072] The preferential categories include at least one of instant discount coupons, discount coupons for purchases above a certain amount, points, event discounts, commission reductions, and data subsidies. The preferential categories are assigned the second weight.

[0073] Under the condition of no preferential treatment, it is directly marked as low impact;

[0074] Under the condition of unilateral preferential treatment, the corresponding first weight and second weight are weighted and summed to obtain the effect coefficient;

[0075] Under the condition of bilateral preferential treatment, the preferential categories under the same type of user nodes are collected, the corresponding first weight and second weight are multiplied, and the adjustment factor is introduced to weight the multiplication result and sum them to obtain the effect coefficient;

[0076] Steps to determine low impact or high impact include:

[0077] Obtain the network environment parameters of the current transaction network, construct a four-dimensional monitoring vector, combine it with the effect coefficient of the current node, and perform weighted summation to obtain the circle ratio coefficient:

[0078] ;

[0079] In the formula, PG represents the circle ratio coefficient, Infact represents the effect coefficient, Class para Represents network environment parameters, including at least one of transaction synchronization frequency, transaction throughput, and network congestion rate. k represents the category of network environment parameters, and k is a positive integer. When it is determined to be any one of the network environment parameters, quantization processing is performed. σ fluct represents the fluctuation area of ​​the four-dimensional monitoring vector corresponding to the vector feature space, σ max represents the area enclosed by the maximum contour map of the four-dimensional monitoring vector corresponding to the vector feature space, μ1 and μ2 are both weight correction coefficients, and (μ1+μ1=1)∪(μ1>0)∪(μ2>0);

[0080] Fluctuation area: Construct a four-dimensional monitoring vector at time t: [transaction synchronization frequency, transaction throughput, network congestion rate, t]. Draw the vector feature space with o1 as the X-axis, o2 as the Y-axis, o3 as the Z-axis, and t as the W-axis. Connect o1, o2, and o3 at the same time to form a contour map. Overlay the contour map with the standard contour map, obtain its horizontal, vertical, and vertical screenshots, and compare them to generate the total overlapping area and the total non-overlapping area. Mark the ratio of the total non-overlapping area to the sum of the total overlapping area and the total non-overlapping area as the fluctuation area.

[0081] The area enclosed by the largest contour: Based on the statistical analysis of historical data, the area enclosed by the largest contour is obtained;

[0082] Compare and analyze the circle ratio coefficient with the preset first comparison interval and second comparison interval:

[0083] When the circle ratio coefficient is in the first comparison interval, a low-order circle structure is formed and it is judged to be low-impact. At the same time, the corresponding node combination is marked as the first transaction data set;

[0084] When the circle ratio coefficient is in the second comparison interval, a high-order circle structure is formed and it is determined to be high-impact. At the same time, the corresponding node combination is marked as the second transaction data set;

[0085] wherein the first comparison interval is smaller than the second comparison interval;

[0086] The first comparison interval and the second comparison interval are based on the statistical range of the mean value pj and standard deviation bz of the historical circle comparison coefficient. Usually, the first comparison interval is [pj-3bz, pj-2bz], and the second comparison interval is [pj+2bz, pj+3bz].

[0087] The steps to implement the batch data synchronization strategy include:

[0088] Segmented transmission: Obtain the transmission link from the payment platform to the data synchronization node and calculate the first transaction data set of each transmission segment; divide the transmission segment into equal data blocks, mark the source identifier of the data point corresponding to each data block, and generate the data source;

[0089] Time alignment: For a preset time period T0, all data sources are collected and grouped into a set Si. Any b0 data sources are randomly selected from Si and partitioned into transaction tasks, generating F(a0, b0) transaction tasks. Transaction tasks are interactions initiated on branches and are obtained through permutations and combinations. a0 represents the number of data sources at any moment in time period T0. Data sources include at least one of click logs, add-to-purchase logs, payment logs, and refund logs. Priority is ranked from highest to lowest in the following order: payment logs, refund logs, add-to-purchase logs, and click logs.

[0090] Deviation calculation: Calculate the corresponding deviations of the timestamps of b0 data sources in each transaction task, select several transaction tasks whose deviations are less than the standard deviation threshold, and obtain the corresponding data blocks;

[0091] It should be noted that the deviation threshold is based on historical deviation values, usually the sum of the average value of the historical deviation value and twice its standard deviation;

[0092] Batch division: The total number of synchronization batches is dynamically preset based on priority, and the processing order and time window of each batch are configured to ensure that the processing periods of different batches do not overlap. When a data block enters the associated batch processing window, the corresponding first transaction data set is activated for transmission. When the data block processing is completed or times out, the transmission of the first transaction data set in the current batch is terminated and resources are released.

[0093] Based on the preset total number of synchronization batches, the following is an explanation of the terms involved:

[0094] Transport link: the end-to-end path (e.g., platform-to-platform transaction path), which determines the underlying physical / logical constraints;

[0095] Transmission segment: It is the path of local segmentation of the transmission link, which can be understood as a local indicator;

[0096] For example, taking online shopping as an example, we can generate refined tasks by dynamically combining data sources:

[0097] In a certain time period, payment logs, refund logs, purchase logs, and click logs are obtained. If a0=4 and b0=2, the task combination is {click log + purchase log, click log + payment log, click log + refund log, purchase log + payment log, purchase log + refund log, payment log + refund log}, for a total of 6 task combinations.

[0098] Execute the blockchain cross-chain synchronization strategy steps, including:

[0099] A delay mechanism is introduced with a preset delay period T1. The second transaction data set is collected and distributed to the shard group. Based on the data synchronization node, cross-shard transactions are executed off-chain. The shard start time slot is dynamically adjusted using an exponential weighting algorithm, and the weight parameters corresponding to the shard are updated through a feedback mechanism.

[0100] Data synchronization nodes: There are several of them, distributed in the transaction data processing network, used to store and synchronize transaction data;

[0101] Shard group: refers to a group of shards, represented as {sliver1, sliver2, ..., sliver m}, where m represents the total number of shards. Each shard group consists of multiple blockchains responsible for maintaining subchains. By adjusting the number of shard groups in each epoch to reduce cross-shard transactions, each smart contract is deployed on a specific shard group, which is responsible for maintaining the contract's state data and processing all transactions interacting with the contract.

[0102] Epoch: It consists of several time slots. The continuous time axis is divided into multiple segments of equal or unequal lengths. Each segment is called a time slot and is used to carry specific tasks or data transmission.

[0103] Alignment processing: First, process the second transaction data set synchronously according to the batch data, retrieve the weight parameters corresponding to the throughput and latency of the corresponding time slots in each epoch, and calculate the difference to obtain the first evaluation coefficient: ;

[0104] In the formula, PG represents the first evaluation coefficient, which is used to evaluate the alignment effect. The larger the first evaluation coefficient is, the better the corresponding evaluation effect is. Put represents throughput, rt represents latency, and β put represents the weight parameter corresponding to throughput, β rt Indicates the weight parameter corresponding to the delay;

[0105] Priority is set according to the alignment effect, and the better the alignment effect, the higher the priority. The second transaction data set is sorted to perform sharding.

[0106] Throughput: refers to the total amount of data transmitted on the transaction network;

[0107] Latency: Indicates the rate of data synchronization, which refers to the time required for a data block (such as a message, packet, bit stream, etc.) to be transmitted from one end of the network to the other end;

[0108] The steps for obtaining weight parameters include:

[0109] Obtain the sorted historical second transaction data set and aggregate it into a first set, with the elements in the first set sorted from low to high priority; take a value from every third digit of the target set, and aggregate the taken values ​​into a second set;

[0110] A deep learning network is preset, and the second set is used as a test data set and input into the deep learning network for training. During the training process, the linear relationship between the priority level and the weight is measured using the mean square error method, and a larger weight is automatically assigned to the high priority level and a smaller weight is assigned to the low priority level. The assigned weight is used as the initial weight parameter for the sharding.

[0111] The steps to perform sorting for sharding include:

[0112] Set the sequence number of the second transaction data set corresponding to the lowest alignment level to 1, and set the sequence numbers of the second transaction data sets corresponding to the alignment levels in ascending order, until the sequence numbers of all second transaction data sets are set; wherein, set the total number of second transaction data sets to Nm, Nm ≥ 2, and the corresponding set sequence numbers are 1 to Nm;

[0113] Set Nm-1 shard groups, and set the block ID of each shard group to Km, where Km is any non-repeating value from 1 to Nm-1. Each block with the shard group ID Km stores and only stores the data of the corresponding transaction data sets with sequence numbers Km and Km+1;

[0114] Cross-shard transactions: Based on the storage calls of the transaction network on the blockchain, the transaction call relationship is obtained and a transaction call graph is constructed. Its representation form is: Call(H,E); where H represents the node set, and each node represents a smart contract address and is paired with the third weight, which represents the number of transactions calling the smart contract; E represents the edge set, and each edge links two smart contracts, indicating the existence of a call relationship between the smart contracts. It is paired with the fourth weight, which represents the number of transactions calling the two smart contracts simultaneously.

[0115] Adaptive sharding: Use the Metis algorithm to perform sharding, retrieve the weight parameters corresponding to the throughput and delay of the corresponding time slots in each epoch, and construct the maximum objective function max (β put , β rt ), calculate the second evaluation coefficient ξ: ;

[0116] Based on the second evaluation coefficient, the optimal time slot for starting the sharding process in each epoch is selected, and the weight parameter corresponding to the delay is updated using the exponential weighting rule: Where, Represents the weight parameter corresponding to the updated delay, p rt represents the probability of time slot selection, |jy| represents an epoch, γ represents the time constant, and γ>0;

[0117] Based on the updated weight parameters, all the first evaluation coefficients in an epoch are calculated again, and the time slot corresponding to the maximum value is taken as the optimal time slot;

[0118] Adaptive online sharding utilizes data synchronization nodes to create a trusted execution environment (TEE), enabling secure off-chain execution of cross-shard transactions. This eliminates the need for multiple rounds of communication between shards and ensures the secure execution of complex smart contract transactions without the overhead of on-chain revalidation. Simultaneously, the TEE synchronizes transaction states required by related shards through an efficient peer-to-peer communication mechanism. Sharding divides the blockchain into multiple smaller, more manageable groups, known as shards, enabling each shard to independently and concurrently process different sets of transaction data. This parallel processing architecture significantly increases the system's transaction throughput by distributing the computational load.

[0119] Data synchronization module: updates the storage content of each data synchronization node and ensures data consistency among multiple nodes. When a new transaction operation is detected, it forces the synchronization window and triggers the incremental synchronization process.

[0120] Specifically, the incremental synchronization process mainly records data change logs. That is, when a new transaction operation is detected, a change log is generated and transmitted to the blockchain. Then, cross-shard transactions are carried out based on the information in the log, and shard reorganization is completed at the same time. The timing synchronization window, for example: forced synchronization every 300ms, can effectively reduce the amount of data transmission and avoid repeated transmission of the same data.

[0121] Example 2:

[0122] The embodiment of the present invention provides a transaction data synchronization management method for multiple payment platforms; Figure 3 is a flowchart of a transaction data synchronization management method according to an exemplary embodiment; Figure 3 , the method comprising:

[0123] Deploy a transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management;

[0124] Based on the transaction network, the preferential forms and categories of user nodes are retrieved, and the effect coefficients of the corresponding supply nodes are created. The effect coefficients are input into the transaction network as attribute matrices, the transaction network is updated, and data synchronization nodes are configured to control each payment platform to transmit transaction data sets according to the optimal transaction path. Under low-impact conditions, batch data synchronization strategies are implemented, and under high-impact conditions, cross-chain synchronization strategies are implemented.

[0125] Update the storage content of each data synchronization node and ensure data consistency across multiple nodes, triggering the incremental synchronization process when a new transaction operation is detected. In the application, the several formulas involved are all dimensionless and calculated numerically. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real-world situation.

[0126] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0127] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0128] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A transaction data synchronization management system for multiple payment platforms, characterized by: The system comprises: Network construction module: Deploy the transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management; Conditional judgment module: Based on the transaction network, it retrieves the preferential form and category of the user node, creates the effect coefficient of the corresponding supply node, inputs the effect coefficient into the transaction network as an attribute matrix, updates the transaction network, and configures the data synchronization node to control each payment platform to transmit the transaction data set according to the optimal transaction path. It implements the batch data synchronization strategy under low-impact conditions and the cross-chain synchronization strategy under high-impact conditions. Data synchronization module: updates the storage content of each data synchronization node and ensures data consistency between multiple nodes. It triggers the incremental synchronization process when a new transaction operation is detected. Among them, the step of creating the effect coefficient of the corresponding supply node includes: the preferential forms include no preferential, unilateral preferential and bilateral preferential, and the preferential forms are assigned a first weight; the preferential categories include at least any one of instant discount coupons, full discount coupons, points, activity discounts, commission reductions and traffic subsidies, and the preferential categories are assigned a second weight; under the condition of no preferential, it is directly marked as low impact; under the condition of unilateral preferential, the corresponding first weight and second weight are weighted and summed to obtain the effect coefficient; under the condition of bilateral preferential, the preferential categories under the same type of user nodes are collected, the corresponding first weight and second weight are multiplied, and the adjustment factor is introduced to weight the multiplication result and corrected, and the sum is obtained to obtain the effect coefficient; The steps of determining whether the impact is low or high include: obtaining network environment parameters of the current transaction network, constructing a four-dimensional monitoring vector, combining the effect coefficient of the current node, and performing weighted summation to obtain a circle ratio coefficient; wherein the network environment parameters include at least one of transaction synchronization frequency, transaction throughput, and network congestion rate; and comparing and analyzing the circle ratio coefficient with a preset first comparison interval and a preset second comparison interval: When the circle ratio coefficient is in the first comparison interval, a low-order circle structure is formed and it is judged to be low-impact. At the same time, the corresponding node combination is marked as the first transaction data set; When the circle ratio coefficient is in the second comparison interval, a high-order circle structure is formed, which is judged to be high-impact, and the corresponding node combination is marked as the second transaction data set; wherein, the first comparison interval is smaller than the second comparison interval.

2. The transaction data synchronization management system for multiple payment platforms according to claim 1, characterized in that: The steps to deploy a trading network include: Treat each payment platform as a supply node and each user as a user node; When a user interacts on any payment platform, an edge is built to construct a transaction network, which is represented as: Net (G, E, W, X); Where G represents the supply node set, and G = {g1, g2, …, gi}, gi represents the i-th supply node, and i∈[1, num1], num1 is the total number of payment platforms, W represents the user node set, and W = {w1, w2, …, wj}, and j∈[1, num2], num2 is the total number of users; E indicates that the user has interactive behavior on the payment platform, and X represents the attribute matrix of all nodes.

3. The transaction data synchronization management system for multiple payment platforms according to claim 1, characterized in that: The steps to implement the batch data synchronization strategy include: Obtain the transmission link from the payment platform to the data synchronization node and calculate the first transaction data set of each transmission segment; divide the transmission segment into equal data blocks, mark the source identifier of the data point corresponding to each data block, and generate the data source; During a preset time period T0, all data sources are collected and grouped into a set Si. Any b0 data sources from Si are randomly selected and partitioned into transaction tasks, generating F(a0, b0) transaction tasks. Transaction tasks are interactions initiated on branches and are obtained based on permutations and combinations. a0 represents the number of data sources at any moment during time period T0, and data sources include at least one of click logs, purchase logs, payment logs, and refund logs. Priority is ranked from highest to lowest in the following order: payment logs, refund logs, purchase logs, and click logs. Calculate the corresponding deviations of the timestamps of b0 data sources in each transaction task, extract several transaction tasks whose deviations are less than the standard deviation threshold, and obtain the corresponding data blocks; The total number of synchronization batches is dynamically preset based on priority, and the processing order and time window of each batch are configured; when a data block enters the associated batch processing window, the corresponding first transaction data set is activated for transmission; when the data block processing is completed or times out, the transmission of the first transaction data set under the current batch is terminated and resources are released.

4. The transaction data synchronization management system for multiple payment platforms according to claim 1, characterized in that: The steps to execute the cross-chain synchronization strategy include: A delay mechanism is introduced, with a preset delay time period T1. The second transaction data set is collected and distributed to the shard group. Based on the data synchronization node, off-chain cross-shard transactions are executed. The exponential weighted algorithm is used to dynamically adjust the shard startup time slot, and the weight parameters corresponding to the shard are updated through the feedback mechanism.

5. The transaction data synchronization management system for multiple payment platforms according to claim 4, characterized in that: Steps to implement a feedback mechanism include: First, the second transaction dataset is processed synchronously according to the batch data. The throughput and latency of the corresponding time slots in each epoch are retrieved, the corresponding weight parameters are dynamically assigned, and the difference is calculated to obtain the first evaluation coefficient to evaluate the alignment effect. The priority level is set according to the alignment effect, and the second transaction dataset is sorted. Use the Metis algorithm to perform sharding, retrieve the initial weight parameters corresponding to the throughput and delay of the corresponding time slots in each epoch, and mark the weight parameter corresponding to the throughput as β put , mark the weight parameter corresponding to the delay as β rt , construct the maximum objective function max(β put , β rt ), by calculating the difference between the weight parameter corresponding to the throughput and the weight parameter corresponding to the delay, dividing the difference by the maximum objective function to calculate a second evaluation coefficient, and the second evaluation coefficient range is [0, 1]; The optimal time slot for starting the sharding process in each epoch is selected based on the second evaluation coefficient and the first evaluation coefficient, and the weight parameter is dynamically updated.

6. The transaction data synchronization management system for multiple payment platforms according to claim 5, characterized in that: The steps of dynamically assigning corresponding weight parameters include: Obtain the sorted historical second transaction data set and aggregate it into a first set, with the elements in the first set sorted from low to high priority; take a value from every third digit of the target set, and aggregate the taken values ​​into a second set; A deep learning network is preset, and the second set is used as a test data set and input into the deep learning network for training and learning. The mean square error method is used to measure the linear relationship between the priority level and the weight, and the weight is automatically assigned to complete the dynamic allocation of weight parameters.

7. A transaction data synchronization management method for multiple payment platforms, characterized in that: The following steps are involved: Deploy a transaction network within the current payment platform and build a corresponding portal; the portal includes user management, transaction management, and blockchain management; Based on the transaction network, the preferential forms and categories of user nodes are retrieved, and the effect coefficients of the corresponding supply nodes are created. The effect coefficients are input into the transaction network as attribute matrices, the transaction network is updated, and data synchronization nodes are configured to control each payment platform to transmit transaction data sets according to the optimal transaction path. Under low-impact conditions, batch data synchronization strategies are implemented, and under high-impact conditions, cross-chain synchronization strategies are implemented. Update the storage content of each data synchronization node and ensure data consistency among multiple nodes, triggering the incremental synchronization process when a new transaction operation is detected; Among them, the step of creating the effect coefficient of the corresponding supply node includes: the preferential forms include no preferential, unilateral preferential and bilateral preferential, and the preferential forms are assigned a first weight; the preferential categories include at least any one of instant discount coupons, full discount coupons, points, activity discounts, commission reductions and traffic subsidies, and the preferential categories are assigned a second weight; under the condition of no preferential, it is directly marked as low impact; under the condition of unilateral preferential, the corresponding first weight and second weight are weighted and summed to obtain the effect coefficient; under the condition of bilateral preferential, the preferential categories under the same type of user nodes are collected, the corresponding first weight and second weight are multiplied, and the adjustment factor is introduced to weight the multiplication result and corrected, and the sum is obtained to obtain the effect coefficient; The steps of determining whether the impact is low or high include: obtaining network environment parameters of the current transaction network, constructing a four-dimensional monitoring vector, combining the effect coefficient of the current node, and performing weighted summation to obtain a circle ratio coefficient; wherein the network environment parameters include at least one of transaction synchronization frequency, transaction throughput, and network congestion rate; and comparing and analyzing the circle ratio coefficient with a preset first comparison interval and a preset second comparison interval: When the circle ratio coefficient is in the first comparison interval, a low-order circle structure is formed and it is judged to be low-impact. At the same time, the corresponding node combination is marked as the first transaction data set; When the circle ratio coefficient is in the second comparison interval, a high-order circle structure is formed, which is judged to be high-impact, and the corresponding node combination is marked as the second transaction data set; wherein, the first comparison interval is smaller than the second comparison interval.

Images