Heterogeneous chain cache incremental updating system and method based on consistency verification

Abstract

This invention discloses a heterogeneous blockchain cache incremental update system and method based on consistency verification, belonging to the field of big data analysis technology. This invention acquires and constructs a heterogeneous blockchain cache incremental analysis set by deploying data listeners, and filters the update benchmark blockchain with the least fluctuation through feature quantification, solving the problem of lack of stable references for heterogeneous blockchain cache incremental updates. It constructs a dedicated monitoring profile, performs dual comparisons to determine update anomalies, sets backtracking anchors at the corresponding block height, and establishes an anchor feature repository, reducing the amount of rollback data and computational resource overhead. After backtracking reconciliation verification is completed, relevant data is retrieved through anchor parsing, and the predicted value of the next cache update feature change is output. A comprehensive feature fluctuation score is calculated by combining historical fluctuation amplitudes, and the benchmark chain is updated to achieve dynamic iteration, shortening the processing and verification cycle and solving the problems of low rollback efficiency and insufficient timeliness of consistency verification.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, specifically to a heterogeneous chain cache incremental update system and method based on consistency verification. Background Technology

[0002] In a heterogeneous blockchain aggregation architecture, each blockchain operates independently and continuously generates incremental on-chain data. Cache incremental updates are a key link in maintaining the efficiency of data access in the architecture and ensuring the consistency between cached data and on-chain data. Efficient and reliable heterogeneous chain cache incremental updates can effectively reduce the pressure of on-chain data queries, avoid problems such as cache invalidation and data deviation, and ensure the accuracy and timeliness of data calls in the architecture. It is the core foundation for supporting the stable operation of the heterogeneous blockchain aggregation architecture and improving the overall processing performance. It directly determines the practicality and industrial application potential of the heterogeneous blockchain aggregation architecture and plays an important supporting role in the large-scale development of heterogeneous blockchain aggregation technology.

[0003] In heterogeneous blockchain aggregation architectures, the consensus mechanisms and data structures of each blockchain differ fundamentally. When some blockchains trigger data rollback, existing technologies generally adopt a one-time rollback to a fixed historical node. This model handles a massive amount of data during rollback, resulting in high computational resource overhead and significantly extending the execution cycle of backtracking reconciliation and incremental cache updates. This severely restricts the timeliness of cache data consistency verification and indirectly affects the efficient operation requirements of heterogeneous blockchain aggregation architectures. Therefore, there is an urgent need for a heterogeneous chain cache incremental update system and method based on consistency verification. Summary of the Invention

[0004] The purpose of this invention is to provide a heterogeneous chain cache incremental update system and method based on consistency verification, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a heterogeneous chain cache incremental update method based on consistency verification, the heterogeneous chain cache incremental update method comprising the following steps:

[0006] Obtain on-chain compliance incremental data for each blockchain in the heterogeneous blockchain aggregation architecture, and construct a heterogeneous chain cache incremental analysis set based on the on-chain compliance incremental data and the chain identifier of the corresponding blockchain.

[0007] Based on the genesis block of each blockchain in the heterogeneous blockchain aggregation architecture, the basic chain information of the corresponding blockchain is initialized. The basic chain information includes the genesis block hash, the initial block height, and the chain identifier.

[0008] On-chain data listeners are deployed on each blockchain. The data listeners establish communication relationships with the full nodes of the corresponding blockchain to capture incremental data that has reached finality on the chain in real time. The incremental data includes block data, transaction data and corresponding block confirmation information.

[0009] The data listener acquires on-chain compliance incremental data of each blockchain, selects the chain identifier of each blockchain as the key, selects the on-chain compliance incremental data captured by the corresponding listener and the data capture timestamp as the value, and stores it in a structured dataset in a key-value pair storage method to construct an on-chain incremental data set of heterogeneous blockchain aggregation architecture, denoted as heterogeneous chain cache incremental analysis set.

[0010] The process involves acquiring compliant incremental data on-chain and constructing a heterogeneous chain cache incremental analysis set. This is achieved by initializing the basic chain information of each blockchain, deploying on-chain data listeners, and establishing full node communication connections. Real-time capture of finalized incremental data is then performed, using the chain identifier as the key and the compliant incremental data and data capture timestamp as the values ​​to construct a structured heterogeneous chain cache incremental analysis set. This implementation method enables standardized collection and structured storage of incremental data from each chain within a heterogeneous blockchain aggregation architecture. It provides complete and traceable data support for subsequent rollback event statistical analysis and updating the baseline blockchain selection, avoiding data deviation issues caused by fragmented data and inconsistent formats from the data source.

[0011] Based on the heterogeneous chain cache incremental analysis set, rollback event statistical analysis is performed to obtain the update benchmark blockchain in the heterogeneous blockchain aggregation architecture.

[0012] Configure a time sliding window of a preset fixed length, and pull incremental synchronization height difference data, single batch verification readiness rate data and single batch verification time data of each blockchain from the heterogeneous chain cache incremental analysis set according to the preset step size;

[0013] For each blockchain, the steady-state baseline corresponding to each of the three features within the time sliding window is calculated. The steady-state baseline is the arithmetic mean of the corresponding feature data within the time sliding window.

[0014] The absolute deviation between the real-time values ​​of each feature within the calculation window and the corresponding steady-state baseline is calculated to generate a deviation sequence for each blockchain corresponding to three features. Based on the three sigma principle, a feature anomaly judgment threshold is preset. When the deviation value of a single feature exceeds the anomaly judgment threshold, it is marked as a corresponding feature anomaly sample.

[0015] Analyze the baseline drift characteristics of each deviation sequence. The baseline drift characteristics include the fluctuation amplitude and the number of consecutive drifts of the deviation sequence. The number of consecutive drifts is the number of samples whose deviation values ​​continuously exceed the anomaly judgment threshold.

[0016] Configure a time sliding window of a preset fixed length, and pull incremental synchronization height difference data, single batch verification readiness rate data and single batch verification time data of each blockchain from the heterogeneous chain cache incremental analysis set according to the preset step size;

[0017] For each blockchain, the steady-state baseline corresponding to each of the three features within the time sliding window is calculated. The steady-state baseline is the arithmetic mean of the corresponding feature data within the time sliding window.

[0018] The absolute deviation between the real-time values ​​of each feature within the calculation window and the corresponding steady-state baseline is calculated to generate a deviation sequence for each blockchain corresponding to three features. Based on the three sigma principle, a feature anomaly judgment threshold is preset. When the deviation value of a single feature exceeds the anomaly judgment threshold, it is marked as a corresponding feature anomaly sample.

[0019] Analyze the baseline drift characteristics of each deviation sequence. The baseline drift characteristics include the fluctuation amplitude and the number of consecutive drifts of the deviation sequence. The number of consecutive drifts is the number of samples whose deviation values ​​continuously exceed the anomaly judgment threshold.

[0020] Based on a heterogeneous blockchain cache incremental analysis set, this process completes feature data retrieval within a time sliding window, steady-state baseline calculation, deviation sequence generation, and baseline drift characteristic analysis. Then, through feature weight allocation, volatility amplitude standardization, and comprehensive volatility score calculation and ranking, a benchmark blockchain for updates is selected. This implementation process can quantitatively evaluate the operational stability of each blockchain, accurately identifying the blockchain with the least volatility as the unified update benchmark. It solves the problem of a lack of stable reference benchmarks for cache incremental updates due to fundamental differences in consensus mechanisms and data structures among blockchains, thus avoiding chaos in cache update order caused by the lack of a benchmark.

[0021] Based on the aforementioned updated benchmark blockchain, the real-time cache incremental update of the heterogeneous blockchain aggregation architecture is monitored, and the cache incremental update characteristics of each blockchain are extracted to construct a blockchain profile for monitoring cache incremental updates.

[0022] Instantaneous disturbance data is removed from the three characteristic data corresponding to each blockchain. The instantaneous disturbance data refers to sample data where the single deviation value exceeds the characteristic anomaly judgment threshold but the number of consecutive drifts does not reach the preset drift threshold.

[0023] For the three feature data after removing instantaneous disturbance data, the historical extreme values ​​of the corresponding features of each blockchain are extracted respectively. The historical maximum value is taken for incremental synchronization height difference, the historical minimum value is taken for single batch verification readiness rate, and the historical maximum value is taken for single batch verification time. These are used as the monitoring calibration values ​​of each feature.

[0024] The three features of each blockchain are collected in real time, and the monitoring calibration value and feature deviation sequence are associated and bound to construct a unique cache incremental update monitoring profile for each blockchain. The monitoring profile corresponds one-to-one with the chain identifier of the corresponding blockchain.

[0025] Based on the monitoring profile of the updated benchmark blockchain, the real-time data of the three features of each blockchain to be judged are compared with their corresponding monitoring calibration values. At the same time, the feature deviation sequence of the blockchain to be judged is compared with the feature deviation sequence of the updated benchmark blockchain, thus completing the real-time monitoring of the cache incremental update of the heterogeneous blockchain aggregation architecture.

[0026] When the real-time data of a single feature of any blockchain to be judged exceeds the corresponding monitoring calibration value, and the fluctuation range of the feature deviation sequence exceeds the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in an abnormal state of cache incremental update.

[0027] When all the real-time data of any single feature of any blockchain to be judged does not exceed the corresponding monitoring calibration value, or when the fluctuation range of any single feature deviation sequence does not exceed the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in a normal state of cache incremental update.

[0028] Using the updated benchmark blockchain as a reference, transient disturbance data in the feature data is removed, and the feature monitoring calibration values ​​for each chain are determined to construct a dedicated cache incremental update monitoring profile. The cache incremental update status is determined by comparing the features of the blockchain to be judged with its own monitoring calibration value and with the deviation sequence of the updated benchmark blockchain. This implementation method can filter out interference from transient disturbance data, achieve accurate identification of the cache incremental update status of each blockchain, effectively solve the problems of inaccurate identification of abnormal states and monitoring failure during cache updates, provide a reliable basis for the accurate triggering of backtracking anchors, and ensure the effectiveness of cache incremental update monitoring.

[0029] Create an anchor feature repository and set back anchor points for each blockchain based on the monitoring results of incremental cache updates;

[0030] When the monitoring result of any blockchain to be judged is determined to be in an abnormal state of cache incremental update, an abnormal backtracking anchor point is set at the current block height position where the abnormality of the corresponding blockchain to be judged occurs; an anchor point snapshot is executed on the cache data, real-time data of single abnormal features, and feature deviation sequence at the current moment, and the anchor point cache snapshot hash corresponding to the current abnormal backtracking anchor point is calculated and generated.

[0031] Assign a unique anchor ID to the set anomaly backtracking anchor point and record the anomaly occurrence timestamp. Bind the anomaly occurrence timestamp, real-time data of individual anomaly features, feature deviation sequence, and anchor point cache snapshot hash with the chain identifier, anchor ID, and current block of the corresponding blockchain to form anomaly anchor point data. Write the anomaly anchor point data into the anchor point feature repository.

[0032] Based on the monitoring results of cache incremental update anomalies, an anomaly backtracking anchor point is set at the current block height where the anomaly occurred. This involves taking an anchor point snapshot, calculating the snapshot hash, and binding and storing the anomaly anchor point data, thus constructing an anchor point feature repository. This implementation method can accurately locate the anomaly node and completely retain the anomaly-related data when a cache update anomaly occurs. It abandons the existing technology's method of rolling back to a fixed historical node all at once, significantly reducing the amount of data to be processed during rollback operations, lowering the high computational resource overhead, and shortening the preparation time for backtracking reconciliation, effectively improving the execution efficiency of backtracking operations.

[0033] The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the backtracking reconciliation blockchain completes the verification and cache incremental update, the data is processed based on the rollback events of each blockchain combined with the neural network model to predict the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and a new update benchmark blockchain is determined based on the predicted feature changes.

[0034] The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the blockchain completes the verification and cache incremental update, the anchor ID corresponding to each blockchain is retrieved from the anchor feature repository. The abnormal anchor data and the feature deviation sequence data associated with historical rollback events of the corresponding blockchain are obtained by parsing the anchor ID.

[0035] The frequency of historical rollback events and the corresponding single-item anomaly feature data bound to the anomaly anchor points are extracted from each blockchain. The extracted data is used as input parameters to train a neural network model. The neural network model outputs the feature change prediction value for the next incremental cache update of each blockchain. The feature change prediction value includes the incremental synchronization height difference change value, the single batch verification readiness rate change value, and the single batch verification time change value. Based on the weights of the corresponding features, a comprehensive feature fluctuation score for each blockchain is calculated.

[0036] The characteristic fluctuation comprehensive score of each blockchain is sorted, and the blockchain with the lowest characteristic fluctuation comprehensive score is selected as the new updated benchmark blockchain of the heterogeneous blockchain aggregation architecture.

[0037] Through a backtracking reconciliation verification unit, the verification of the backtracking reconciliation blockchain and incremental cache updates are completed based on the anchor feature repository. Then, the new benchmark chain prediction unit retrieves historical data from the anchor feature repository and combines it with a neural network model to predict feature changes and calculate a comprehensive score for feature fluctuations, ultimately selecting and determining a new update benchmark blockchain. This implementation method relies on anchor data to achieve efficient backtracking reconciliation verification, further shortening the execution cycle of incremental cache updates and data consistency verification, and solving the problem of insufficient timeliness in consistency verification. At the same time, through model prediction, it enables advance prediction of cache update features and dynamic adjustment of the update benchmark, which can continuously adapt to changes in the blockchain's operating status, ensure the long-term consistency of incremental cache updates, and support the efficient and stable operation of heterogeneous blockchain aggregation architectures.

[0038] Furthermore, a heterogeneous chain cache incremental update system based on consistency verification is provided, which includes an incremental data set construction module, an update benchmark blockchain determination module, a cache incremental update monitoring module, a backtracking anchor point setting module, and a new benchmark blockchain prediction and determination module.

[0039] The incremental data set construction module is used to acquire on-chain compliant incremental data of each blockchain in the heterogeneous blockchain aggregation architecture, and construct a heterogeneous chain cache incremental analysis set based on the on-chain compliant incremental data and the chain identifier of the corresponding blockchain; the update benchmark blockchain determination module is used to perform statistical analysis of blockchain rollback events based on the heterogeneous chain cache incremental analysis set, and then obtain the update benchmark blockchain in the heterogeneous blockchain aggregation architecture; the cache incremental update monitoring module is used to monitor the real-time cache incremental updates of the heterogeneous blockchain aggregation architecture based on the update benchmark blockchain, and extract the cache incremental update features of each blockchain to construct a blockchain profile; the backtracking anchor setting module is used to create an anchor feature repository, and set backtracking anchors for each blockchain based on the monitoring results of cache incremental updates; the new benchmark blockchain prediction and determination module is used to verify the backtracking reconciliation blockchain based on the anchor feature repository, and after the backtracking reconciliation blockchain completes the verification and cache incremental update, it combines the rollback events of each blockchain with the neural network model for data processing, predicts the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and determines the new update benchmark blockchain;

[0040] The incremental data set construction module includes a compliance incremental data acquisition unit and an incremental analysis set construction unit; the compliance incremental data acquisition unit is used to acquire on-chain compliance incremental data of each blockchain in the heterogeneous blockchain aggregation architecture; the incremental analysis set construction unit is used to construct a heterogeneous chain cache incremental analysis set based on the on-chain compliance incremental data and the chain identifier of the corresponding blockchain.

[0041] The updated benchmark blockchain determination module includes a rollback event statistics unit and a benchmark blockchain filtering unit; the rollback event statistics unit is used to perform statistical analysis of blockchain rollback events based on the heterogeneous chain cache incremental analysis set; the benchmark blockchain filtering unit is used to obtain the updated benchmark blockchain in the heterogeneous blockchain aggregation architecture based on the results of the blockchain rollback event statistical analysis.

[0042] The cache incremental update monitoring module includes a real-time update monitoring unit and a blockchain profile building unit; the real-time update monitoring unit is used to monitor the real-time cache incremental updates of the heterogeneous blockchain aggregation architecture based on the update benchmark blockchain; the blockchain profile building unit is used to extract the cache incremental update features of each blockchain and build a blockchain profile.

[0043] The backtracking anchor setting module includes an anchor repository creation unit and a backtracking anchor deployment unit; the anchor repository creation unit is used to create an anchor feature repository; the backtracking anchor deployment unit is used to set backtracking anchors for each blockchain based on the monitoring results of cache incremental updates.

[0044] The new benchmark blockchain prediction and determination module includes a backtracking reconciliation verification unit and a new benchmark blockchain prediction unit. The backtracking reconciliation verification unit is used to verify the backtracking reconciliation blockchain based on the anchor feature repository and ensure that the backtracking reconciliation blockchain completes the cache incremental update. The new benchmark blockchain prediction unit is used to combine the rollback events of each blockchain with the neural network model for data processing, predict the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and determine the new update benchmark blockchain.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] 1. This invention initializes the basic information of heterogeneous blockchains, deploys data listeners, constructs a standardized heterogeneous blockchain cache incremental analysis set, and then selects the update benchmark blockchain with the least operational fluctuation based on feature quantitative analysis. This addresses the problem at its source that heterogeneous blockchains, due to fundamental differences in consensus mechanisms and data structures, lack a stable reference benchmark for cache incremental updates, avoids update deviations caused by fragmented data and inconsistent formats, and eliminates chaotic cache update order.

[0047] 2. This invention is based on constructing a blockchain-specific cache incremental update monitoring profile. It employs a dual comparison method to accurately determine abnormal update states, locates and sets back anchor points at the block height where the anomaly occurs, and establishes an anchor point feature repository. It abandons the traditional one-time rollback to a fixed historical node, significantly reducing the data volume of rollback processing, lowering high computational resource overhead, and shortening the preparation time for backtracking reconciliation. This solves the technical drawbacks of traditional rollback methods, such as high resource consumption and low efficiency in anomaly handling.

[0048] 3. This invention achieves efficient blockchain verification and cache updates through a backtracking reconciliation verification unit, and leverages a new benchmark chain prediction unit combined with a neural network model to realize dynamic iteration of update feature prediction and update benchmark. This effectively shortens the execution cycle of backtracking reconciliation and incremental cache updates, solves the problem of insufficient timeliness in cache data consistency verification, continuously adapts to changes in blockchain operating status, and ensures long-term consistency between cached data and on-chain data, supporting the efficient and stable operation and large-scale deployment of heterogeneous blockchain aggregation architectures. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating a heterogeneous chain cache incremental update method based on consistency verification according to the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1: As Figure 1 As shown, this invention provides a technical solution: a heterogeneous chain cache incremental update method based on consistency verification. The heterogeneous chain cache incremental update method includes the following steps:

[0052] Obtain on-chain compliance incremental data for each blockchain in the heterogeneous blockchain aggregation architecture, and construct a heterogeneous chain cache incremental analysis set based on the on-chain compliance incremental data and the chain identifier of the corresponding blockchain.

[0053] Based on the genesis block of each blockchain in the heterogeneous blockchain aggregation architecture, the basic chain information of the corresponding blockchain is initialized. The basic chain information includes the genesis block hash, the initial block height, and the chain identifier.

[0054] On-chain data listeners are deployed on each blockchain. The data listeners establish communication relationships with the full nodes of the corresponding blockchain to capture incremental data that has reached finality on the chain in real time. The incremental data includes block data, transaction data and corresponding block confirmation information.

[0055] The data listener acquires on-chain compliance incremental data of each blockchain, selects the chain identifier of each blockchain as the key, selects the on-chain compliance incremental data captured by the corresponding listener and the data capture timestamp as the value, and stores it in a structured dataset in a key-value pair storage method to construct an on-chain incremental data set of heterogeneous blockchain aggregation architecture, denoted as heterogeneous chain cache incremental analysis set.

[0056] In practice, after the blockchain for backtracking and reconciliation completes verification and incremental cache updates, it retrieves the anchor IDs of each blockchain using the indexing mechanism of the anchor feature repository. Through standardized parsing logic, it extracts abnormal anchor data and feature deviation sequence data associated with historical rollback events from the anchor IDs. The frequency of historical rollback events and the single abnormal feature data bound to abnormal anchors are organized into a structured input set and fed into the neural network model. The model outputs the predicted values ​​of three feature changes for the next incremental cache update based on the heterogeneous chain feature learning logic. Combined with the fluctuation amplitude of the feature deviation sequence corresponding to historical rollbacks, a weighted accumulation is performed according to a preset fixed ratio to obtain a comprehensive feature fluctuation score. After sorting all blockchain scores in descending order, the node with the lowest score is selected as the new update benchmark blockchain. Its chain identifier, anchor ID, and score are entered into the designated data partition of the anchor feature repository to ensure accurate association between subsequent monitoring and backtracking anchors.

[0057] By relying on the heterogeneous chain cache incremental analysis set to execute the rollback event statistical analysis process, the specific selection results of updating the benchmark blockchain in the heterogeneous blockchain aggregation architecture are clarified.

[0058] Configure a time sliding window of a preset fixed length, and accurately pull incremental synchronization height difference data, single batch verification readiness rate data and single batch verification time data of each blockchain in the heterogeneous chain cache incremental analysis set at preset steps to ensure the completeness and timeliness of data extraction;

[0059] For each blockchain, the steady-state baseline corresponding to each of the above three features within the time sliding window is calculated. The steady-state baseline is obtained by solving the arithmetic mean of the corresponding feature data within the window, so as to eliminate the impact of data fluctuations on the accuracy of the baseline.

[0060] Substitute the real-time values ​​of each feature within the window with the corresponding steady-state baseline one by one, calculate the absolute deviation between the two, and construct the three feature deviation sequences specific to each blockchain one by one;

[0061] Based on the Three Sigma principle, a threshold for feature anomaly detection is set. When the deviation value of a feature exceeds the threshold, it is marked as an abnormal sample of the corresponding feature.

[0062] The baseline drift characteristics of each deviation sequence are analyzed in depth. The baseline drift characteristics specifically include the fluctuation amplitude of the deviation sequence and the number of consecutive drifts. The number of consecutive drifts is clearly defined as the cumulative number of samples whose deviation values ​​continuously exceed the anomaly judgment threshold.

[0063] In practical implementation, for heterogeneous aggregation architecture of consortium blockchains, after the blockchain backtracking reconciliation verification is completed and the cache incremental update is implemented, the anchor IDs of each chain are retrieved in batches from the anchor feature repository. During the parsing process, the binding relationship between the anchor ID and the chain identifier is verified simultaneously to ensure the accuracy of the attribution of abnormal anchor data and feature deviation sequence data. The frequency of historical rollback events is extracted and deduplicated by block height. Individual abnormal feature data retains the original collection dimensions and is then input into the neural network model. The feature change prediction value output by the model covers three core dimensions, including incremental synchronization height difference. The weighting ratio of the feature fluctuation comprehensive score is set based on the previous heterogeneous chain feature analysis results. During the scoring and sorting process, the scoring records of all valid blockchain nodes are retained. After the new benchmark blockchain is determined, its associated information is written to the anchor feature repository in real time, and the updated data can be read instantly by the cache incremental update monitoring module.

[0064] Based on the aforementioned updated benchmark blockchain, the real-time cache incremental update of the heterogeneous blockchain aggregation architecture is monitored, and the cache incremental update characteristics of each blockchain are extracted to construct a blockchain profile for monitoring cache incremental updates.

[0065] Instantaneous disturbance data is removed from the three characteristic data corresponding to each blockchain. The instantaneous disturbance data refers to sample data where the single deviation value exceeds the characteristic anomaly judgment threshold but the number of consecutive drifts does not reach the preset drift threshold.

[0066] For the three feature data after removing instantaneous disturbance data, the historical extreme values ​​of the corresponding features of each blockchain are extracted respectively. The historical maximum value is taken for incremental synchronization height difference, the historical minimum value is taken for single batch verification readiness rate, and the historical maximum value is taken for single batch verification time. These are used as the monitoring calibration values ​​of each feature.

[0067] The three features of each blockchain are collected in real time, and the monitoring calibration value and feature deviation sequence are associated and bound to construct a unique cache incremental update monitoring profile for each blockchain. The monitoring profile corresponds one-to-one with the chain identifier of the corresponding blockchain.

[0068] Based on the monitoring profile of the updated benchmark blockchain, the real-time data of the three features of each blockchain to be judged are compared with their corresponding monitoring calibration values. At the same time, the feature deviation sequence of the blockchain to be judged is compared with the feature deviation sequence of the updated benchmark blockchain, thus completing the real-time monitoring of the cache incremental update of the heterogeneous blockchain aggregation architecture.

[0069] When the real-time data of a single feature of any blockchain to be judged exceeds the corresponding monitoring calibration value, and the fluctuation range of the feature deviation sequence exceeds the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in an abnormal state of cache incremental update.

[0070] When all the real-time data of any single feature of any blockchain to be judged does not exceed the corresponding monitoring calibration value, or when the fluctuation range of any single feature deviation sequence does not exceed the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in a normal state of cache incremental update.

[0071] In practical implementation, for heterogeneous architectures that combine public and private chains, after the blockchain for backtracking and reconciliation completes verification and cache incremental updates, the anchor ID is retrieved based on precise matching of chain identifiers. The abnormal anchor data obtained by parsing the anchor ID includes full information such as the timestamp of the abnormal occurrence and the feature deviation sequence. The frequency statistics of historical rollback events cover the entire operating cycle of the blockchain. After format normalization, the individual abnormal feature data is fed into the neural network model. The predicted value of the feature change in the next cache incremental update output by the model and the fluctuation amplitude of the feature deviation sequence are used together as the scoring dimension. The feature fluctuation comprehensive score is obtained by weighting according to a preset fixed ratio. After sorting the scores from low to high, the top blockchain is selected as the new update benchmark blockchain. Its chain identifier, anchor ID, and score are synchronously written into the anchor feature repository to form a dynamic iterative record of the benchmark chain information.

[0072] Create an anchor feature repository and set back anchor points for each blockchain based on the monitoring results of incremental cache updates;

[0073] When the monitoring result of any blockchain to be judged is determined to be in an abnormal state of cache incremental update, an abnormal backtracking anchor point is set at the current block height position where the abnormality of the corresponding blockchain to be judged occurs; an anchor point snapshot is executed on the cache data, real-time data of single abnormal features, and feature deviation sequence at the current moment, and the anchor point cache snapshot hash corresponding to the current abnormal backtracking anchor point is calculated and generated.

[0074] Assign a unique anchor ID to the set anomaly backtracking anchor point and record the anomaly occurrence timestamp. Bind the anomaly occurrence timestamp, real-time data of individual anomaly features, feature deviation sequence, and anchor point cache snapshot hash with the chain identifier, anchor ID, and current block of the corresponding blockchain to form anomaly anchor point data. Write the anomaly anchor point data into the anchor point feature repository.

[0075] In practical implementation, after the blockchain completes backtracking reconciliation verification and cache incremental update, it retrieves the anchor IDs of each blockchain from the anchor feature repository according to the chain identifier dimension. When parsing the anchor IDs, it simultaneously verifies the data integrity to ensure that there are no missing abnormal anchor data and feature deviation sequence data associated with historical rollbacks. The frequency of historical rollback events is extracted and combined with the block confirmation status to complete the statistics. After the individual abnormal feature data maintains the feature dimension consistent with the monitoring profile, it is input into the neural network model. The predicted values ​​of the three feature changes output by the model are weighted with the fluctuation amplitude of the feature deviation sequence according to a preset fixed ratio to obtain the comprehensive feature fluctuation score. After the scores of each blockchain are sorted, the blockchain with the lowest score is determined as the new updated benchmark blockchain. When its core information is written into the anchor feature repository, it is associated with the original benchmark chain information to facilitate full-process traceability.

[0076] The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the backtracking reconciliation blockchain completes the verification and cache incremental update, the data is processed based on the rollback events of each blockchain combined with the neural network model to predict the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and a new update benchmark blockchain is determined based on the predicted feature changes.

[0077] The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the blockchain completes the verification and cache incremental update, the anchor ID corresponding to each blockchain is retrieved from the anchor feature repository. The abnormal anchor data and the feature deviation sequence data associated with historical rollback events of the corresponding blockchain are obtained by parsing the anchor ID.

[0078] The frequency of historical rollback events and the corresponding single-item anomaly feature data bound to the anomaly anchor points are extracted from each blockchain. The extracted data is used as input parameters to train a neural network model. The neural network model outputs the feature change prediction value for the next incremental cache update of each blockchain. The feature change prediction value includes the incremental synchronization height difference change value, the single batch verification readiness rate change value, and the single batch verification time change value. Based on the weights of the corresponding features, a comprehensive feature fluctuation score for each blockchain is calculated.

[0079] The characteristic fluctuation comprehensive score of each blockchain is sorted, and the blockchain with the lowest characteristic fluctuation comprehensive score is selected as the new updated benchmark blockchain of the heterogeneous blockchain aggregation architecture.

[0080] In practical implementation, after the blockchain for backtracking reconciliation completes verification and cache incremental update, the anchor feature repository responds to the anchor ID retrieval request according to the preset interface protocol. The abnormal anchor data and feature deviation sequence data obtained by parsing the anchor ID are synchronously hash verified. The frequency of historical rollback events is statistically analyzed using a time sliding window as the dimension. After filtering out instantaneous disturbance data, the single abnormal feature data is fed into the neural network model. The predicted value of the feature change for the next cache incremental update output by the model covers three core features. The comprehensive score of feature fluctuation is obtained by weighting and accumulating two indicators by a preset fixed ratio. Based on the score ranking result, the blockchain with the lowest score is selected as the new update benchmark blockchain. When its chain identifier, anchor ID, and score are written into the anchor feature repository, the index label is updated to improve the matching efficiency of subsequent cache incremental update monitoring.

[0081] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

Claims

1. A heterogeneous chain cache incremental update method based on consistency verification, characterized in that: The heterogeneous chain cache incremental update method includes: Obtain on-chain compliance incremental data for each blockchain in the heterogeneous blockchain aggregation architecture, and construct a heterogeneous chain cache incremental analysis set based on the on-chain compliance incremental data and the chain identifier of the corresponding blockchain. Based on the heterogeneous chain cache incremental analysis set, rollback event statistical analysis is performed to obtain the update benchmark blockchain in the heterogeneous blockchain aggregation architecture. Based on the aforementioned updated benchmark blockchain, the real-time cache incremental update of the heterogeneous blockchain aggregation architecture is monitored, and the cache incremental update characteristics of each blockchain are extracted to construct a blockchain profile for monitoring cache incremental updates. Create an anchor feature repository and set back anchor points for each blockchain based on the monitoring results of incremental cache updates; The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the backtracking reconciliation blockchain completes the verification and cache incremental update, the data is processed based on the rollback events of each blockchain and the neural network model. The feature changes for the next cache incremental update in the heterogeneous blockchain aggregation architecture are predicted, and a new update benchmark blockchain is determined based on the predicted feature changes.

2. The heterogeneous chain cache incremental update method based on consistency verification according to claim 1, characterized in that: The obtained updated baseline blockchain in the heterogeneous blockchain aggregation architecture includes: Configure a time sliding window of a preset fixed length, and pull incremental synchronization height difference data, single batch verification readiness rate data and single batch verification time data of each blockchain from the heterogeneous chain cache incremental analysis set according to the preset step size; For each blockchain, the steady-state baseline corresponding to each of the three features within the time sliding window is calculated. The steady-state baseline is the arithmetic mean of the corresponding feature data within the time sliding window. The absolute deviation between the real-time values ​​of each feature within the calculation window and the corresponding steady-state baseline is calculated to generate a deviation sequence for each blockchain corresponding to three features. Based on the three sigma principle, a feature anomaly judgment threshold is preset. When the deviation value of a single feature exceeds the anomaly judgment threshold, it is marked as a corresponding feature anomaly sample. The baseline drift characteristics of each deviation sequence are analyzed. The baseline drift characteristics include the fluctuation amplitude of the deviation sequence and the number of consecutive drifts. The number of consecutive drifts is the number of samples in which the deviation value continuously exceeds the anomaly judgment threshold.

3. The heterogeneous chain cache incremental update method based on consistency verification according to claim 2, characterized in that: The obtained updated baseline blockchain in the heterogeneous blockchain aggregation architecture also includes: Calculate the sum of the number of individual abnormal samples for each of the three features, which is the total number of abnormal samples; divide the number of individual abnormal samples for each feature by the total number of abnormal samples to obtain the weight of the corresponding feature, and the sum of the weights of the three features is 1. The fluctuation amplitudes of the remaining three feature deviation sequences of the blockchain are processed dimensionlessly using the min-max standardization method to obtain the standardized fluctuation amplitudes. The standardized fluctuation amplitudes of each feature are multiplied by their corresponding weights and summed to obtain the comprehensive fluctuation score of each blockchain. The comprehensive volatility scores of each blockchain are ranked, and the blockchain with the lowest comprehensive volatility score is selected as the updated benchmark blockchain for the heterogeneous blockchain aggregation architecture.

4. The heterogeneous chain cache incremental update method based on consistency verification according to claim 1, characterized in that: The process involves extracting cache incremental update features from each blockchain to construct a blockchain profile for monitoring cache incremental updates, including: Instantaneous disturbance data is removed from the three characteristic data corresponding to each blockchain. The instantaneous disturbance data refers to sample data where the single deviation value exceeds the characteristic anomaly judgment threshold but the number of consecutive drifts does not reach the preset drift threshold. For the three feature data after removing instantaneous disturbance data, the historical extreme values ​​of the corresponding features of each blockchain are extracted respectively. The historical maximum value is taken for incremental synchronization height difference, the historical minimum value is taken for single batch verification readiness rate, and the historical maximum value is taken for single batch verification time. These are used as the monitoring calibration values ​​of each feature. The three features of each blockchain are collected in real time, and the monitoring calibration value and feature deviation sequence are associated and bound to construct a unique cache incremental update monitoring profile for each blockchain. The monitoring profile corresponds one-to-one with the chain identifier of the corresponding blockchain. Based on the monitoring profile of the updated benchmark blockchain, the real-time data of the three features of each blockchain to be judged are compared with their corresponding monitoring calibration values. At the same time, the feature deviation sequence of the blockchain to be judged is compared with the feature deviation sequence of the updated benchmark blockchain, thus completing the real-time monitoring of the cache incremental update of the heterogeneous blockchain aggregation architecture. When the real-time data of a single feature of any blockchain to be judged exceeds the corresponding monitoring calibration value, and the fluctuation range of the feature deviation sequence exceeds the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in an abnormal state of cache incremental update. When all real-time data of any single feature of any blockchain to be judged does not exceed the corresponding monitoring calibration value, or when the fluctuation range of any single feature deviation sequence does not exceed the fluctuation range of the corresponding feature deviation sequence of the updated benchmark blockchain, the blockchain to be judged is determined to be in a normal state of cache incremental update.

5. The heterogeneous chain cache incremental update method based on consistency verification according to claim 1, characterized in that: The construction of the heterogeneous chain cache incremental analysis set includes: Based on the genesis block of each blockchain in the heterogeneous blockchain aggregation architecture, the basic chain information of the corresponding blockchain is initialized. The basic chain information includes the genesis block hash, the initial block height, and the chain identifier. On-chain data listeners are deployed on each blockchain. The data listeners establish communication relationships with the full nodes of the corresponding blockchain to capture incremental data that has reached finality on the chain in real time. The incremental data includes block data, transaction data and corresponding block confirmation information. The data listener acquires on-chain compliance incremental data of each blockchain, selects the chain identifier of each blockchain as the key, and selects the on-chain compliance incremental data captured by the corresponding listener and the data capture timestamp as the value. The data is stored in a structured dataset in a key-value pair storage method, thereby constructing an on-chain incremental data set of heterogeneous blockchain aggregation architecture, denoted as the heterogeneous chain cache incremental analysis set.

6. The heterogeneous chain cache incremental update method based on consistency verification according to claim 1, characterized in that: The monitoring results based on cache incremental updates are used to set backtracking anchor points for each blockchain, including: When the monitoring result of any blockchain to be judged is determined to be in an abnormal state of cache incremental update, an abnormal backtracking anchor point is set at the current block height position where the abnormality of the corresponding blockchain to be judged occurs; an anchor point snapshot is executed on the cache data, real-time data of single abnormal features, and feature deviation sequence at the current moment, and the anchor point cache snapshot hash corresponding to the current abnormal backtracking anchor point is calculated and generated. Assign a unique anchor ID to the set anomaly backtracking anchor point and record the anomaly occurrence timestamp. Bind the anomaly occurrence timestamp, real-time data of individual anomaly features, feature deviation sequence, and anchor point cache snapshot hash with the chain identifier, anchor ID, and current block of the corresponding blockchain to form anomaly anchor point data. Write the anomaly anchor point data into the anchor point feature repository.

7. The heterogeneous chain cache incremental update method based on consistency verification according to claim 1, characterized in that: The method for determining a new updated baseline blockchain based on predicted feature changes includes: The blockchain for backtracking reconciliation is verified based on the anchor feature repository. After the blockchain completes the verification and cache incremental update, the anchor ID corresponding to each blockchain is retrieved from the anchor feature repository. The abnormal anchor data and the feature deviation sequence data associated with historical rollback events of the corresponding blockchain are obtained by parsing the anchor ID. The frequency of historical rollback events and the corresponding single-item anomaly feature data bound to the anomaly anchor points are extracted from each blockchain. The extracted data is used as input parameters to train a neural network model. The neural network model outputs the feature change prediction value for the next incremental cache update of each blockchain. The feature change prediction value includes the incremental synchronization height difference change value, the single batch verification readiness rate change value, and the single batch verification time change value. Based on the weights of the corresponding features, a comprehensive feature fluctuation score for each blockchain is calculated. The characteristic fluctuation comprehensive scores of each blockchain are ranked, and the blockchain with the lowest characteristic fluctuation comprehensive score is selected as the new updated benchmark blockchain for the heterogeneous blockchain aggregation architecture.

8. A heterogeneous chain cache incremental update system based on consistency verification, applied to the heterogeneous chain cache incremental update method based on consistency verification as described in any one of claims 1-7, characterized in that: The heterogeneous chain cache incremental update system includes an incremental data set construction module, an update benchmark blockchain determination module, a cache incremental update monitoring module, a backtracking anchor point setting module, and a new benchmark blockchain prediction and determination module. The incremental data set construction module is used to acquire on-chain compliant incremental data of each blockchain in the heterogeneous blockchain aggregation architecture, and construct a heterogeneous chain cache incremental analysis set based on the on-chain compliant incremental data and the chain identifier of the corresponding blockchain; the update benchmark blockchain determination module is used to perform statistical analysis of blockchain rollback events based on the heterogeneous chain cache incremental analysis set, and then obtain the update benchmark blockchain in the heterogeneous blockchain aggregation architecture; the cache incremental update monitoring module is used to monitor the real-time cache incremental updates of the heterogeneous blockchain aggregation architecture based on the update benchmark blockchain, and extract the cache incremental update features of each blockchain to construct a blockchain profile; The backtracking anchor setting module is used to create an anchor feature repository and set backtracking anchors for each blockchain based on the monitoring results of cache incremental updates. The new benchmark blockchain prediction and determination module is used to verify the backtracking reconciliation blockchain based on the anchor feature repository. After the backtracking reconciliation blockchain completes the verification and cache incremental update, it combines the rollback events of each blockchain with the neural network model to process the data, predict the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and determine the new update benchmark blockchain.

9. A heterogeneous chain cache incremental update system based on consistency verification according to claim 8, characterized in that: The incremental data set construction module includes a compliance incremental data acquisition unit and an incremental analysis set construction unit; the compliance incremental data acquisition unit is used to acquire on-chain compliance incremental data of each blockchain in the heterogeneous blockchain aggregation architecture; the incremental analysis set construction unit is used to construct a heterogeneous chain cache incremental analysis set based on the on-chain compliance incremental data and the chain identifier of the corresponding blockchain. The updated benchmark blockchain determination module includes a rollback event statistics unit and a benchmark blockchain filtering unit; The rollback event statistics unit is used to perform statistical analysis of blockchain rollback events based on the heterogeneous chain cache incremental analysis set; the benchmark blockchain filtering unit is used to obtain the updated benchmark blockchain in the heterogeneous blockchain aggregation architecture based on the results of the blockchain rollback event statistical analysis. The cache incremental update monitoring module includes an update real-time monitoring unit and a blockchain profile building unit; the update real-time monitoring unit is used to monitor the cache incremental updates of the heterogeneous blockchain aggregation architecture in real time based on the update benchmark blockchain; The blockchain profiling unit is used to extract the cache incremental update features of each blockchain and construct a blockchain profile.

10. A heterogeneous chain cache incremental update system based on consistency verification according to claim 8, characterized in that: The backtracking anchor setting module includes an anchor repository creation unit and a backtracking anchor deployment unit; the anchor repository creation unit is used to create an anchor feature repository; the backtracking anchor deployment unit is used to set backtracking anchors for each blockchain based on the monitoring results of cache incremental updates. The new benchmark blockchain prediction and determination module includes a backtracking reconciliation verification unit and a new benchmark blockchain prediction unit. The backtracking reconciliation verification unit is used to verify the backtracking reconciliation blockchain based on the anchor feature repository and ensure that the backtracking reconciliation blockchain completes the cache incremental update. The new benchmark blockchain prediction unit is used to combine the rollback events of each blockchain with the neural network model for data processing, predict the feature changes of the next cache incremental update in the heterogeneous blockchain aggregation architecture, and determine the new update benchmark blockchain.

Images