Blockchain-based financial data management system and method

Abstract

The application discloses a kind of financial data management system and method based on block chain, it is related to financial data management technical field, including: based on the target of minimizing network communication delay, maximizing data storage efficiency and maximizing computing capacity utilization, the block chain network is divided into multiple sub-networks;Define multiple trust evaluation indexes, for each node, according to the value of its trust evaluation index, calculate the trust degree of the node;Transaction data is classified as high-frequency transaction data and low-frequency historical data, for high-frequency transaction data, execute full node storage strategy, for low-frequency historical data, execute distributed storage strategy;Collect the load information index of block chain sub-network, calculate the network load value of each sub-network, compare the network load value of each sub-network with network load threshold, according to the comparison result, make corresponding processing, to improve storage efficiency, reduce computing power consumption, enhance data processing speed and real-time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of financial data management, in particular to a financial data management system and method based on blockchain. BACKGROUND

[0002] Financial data management systems not only need to ensure the integrity and accuracy of data, but also need to provide efficient data processing capabilities and real-time performance to support complex financial analysis and decision-making. Blockchain technology, as a distributed ledger technology, has shown great potential in the financial field, especially in ensuring the security and trust of transactions, due to its decentralization, transparency and non-tamperability.

[0003] However, when applying blockchain technology to financial data management systems, a series of challenges are faced. First, blockchain requires each node to store all non-tamperable financial transaction records in the entire network, which puts a very high demand on the storage capacity of the node. With the increase in transaction volume, storage costs rise sharply, and storage efficiency may be affected. In addition, the consensus mechanism in blockchain, such as Proof of Work (PoW), relies on the competition for the right to record new blocks by a large amount of computing power among nodes. This process not only consumes a lot of energy, but also leads to a great waste of computing power resources, because only the nodes that successfully mine can get rewards, and the computing power input of the remaining nodes is all invalid consumption.

[0004] In addition, the scalability problem of the blockchain network is also a major challenge faced by financial data management systems. With the increase in network nodes, network communication delay and data synchronization problems become increasingly prominent. How to improve the scalability and efficiency of the network while maintaining decentralization and security has become a problem that needs to be solved. SUMMARY

[0005] (I) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present application provides a financial data management system and method based on blockchain, which improves storage efficiency, reduces computing power consumption, enhances data processing speed and real-time performance by intelligently dividing the blockchain network, evaluating node trustworthiness, implementing differentiated storage strategies and dynamically adjusting network load, thereby meeting the high standards and strict requirements of financial data management.

[0007] (II) Technical solutions

[0008] To achieve the above purposes, the present application is implemented by the following technical solutions: a financial data management system and method based on blockchain, comprising:

[0009] Collect node information of all nodes in the blockchain network, and divide the blockchain network into multiple sub-networks based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization.

[0010] Historical financial data, node behavior records, and interaction data with other nodes are collected from all nodes in the blockchain network. Multiple trust evaluation indicators are defined, and for each node, the trust level of the node is calculated based on the value of its trust evaluation indicator.

[0011] Extract all transaction data from the blockchain network and categorize the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, implement a full-node storage strategy, and for low-frequency historical data, implement a distributed storage strategy.

[0012] Collect load information metrics of blockchain sub-networks, calculate the network load value of each sub-network, compare the network load value of each sub-network with the network load threshold, and take corresponding actions based on the comparison results.

[0013] Furthermore, we define objective functions, including minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, and set constraints: each node belongs to one and only one subnetwork. We then construct an optimization model.

[0014] Constraints: Where D represents minimizing network communication latency, Represents a node i and nodes j Communication delay between them N Represents a set of nodes. , as well as It is a binary variable. Represents a node i and nodes j Are they in the same subnetwork? Represents a node i Does it belong to a subnetwork? k , Indicates whether the task is assigned to a sub-network. k Nodes in i , E This indicates maximizing data storage efficiency. Represents a node i Storage capacity, R Represents the set of subnetworks. U This represents maximizing the utilization of computing power. Represents a node i Computational power;

[0015] The branch and bound method is used to solve the optimization model and determine the partitioning scheme of the approximate optimal solution, that is, the sub-network to which each node belongs, thus dividing the blockchain network into multiple sub-networks.

[0016] Furthermore, based on the node's historical financial data, node behavior records, and interaction data with other nodes, several trust evaluation indicators are defined, including transmission success rate, node online time rate, battery duration ratio, number of communication connections, available storage capacity ratio, as well as abnormal data ratio, packet loss rate, identity fraud ratio, bit error rate, and link establishment latency.

[0017] Furthermore, the trust evaluation indicators are standardized. For trust evaluation indicators with benefit-related attributes, the standardized calculation formula is as follows: For trust evaluation indicators of cost-related attributes, the standardized calculation formula is as follows: ,in, St_b The standard value of the trust evaluation index representing the benefit-oriented attribute. St_c The standardized value of the trust rating index representing cost-related attributes. X This represents the trust rating index value. Max , Min These represent the maximum and minimum values ​​of the trust rating index, respectively.

[0018] Furthermore, for each node, the trust level of that node is calculated based on the value of its trust evaluation index: ,in, Tr Indicates the trust level of a node. Indicates the first m Standardized values ​​of trust evaluation indicators Indicates the first m The weights corresponding to each trust evaluation indicator, where M represents the number of trust evaluation indicators.

[0019] Furthermore, the number of times each transaction data is accessed is counted, and its average access frequency is calculated. The average access frequency of each transaction data is compared with the access frequency threshold. If the average access frequency of the transaction data is greater than the access frequency threshold, the transaction data is classified as high-frequency transaction data; otherwise, the transaction data is classified as low-frequency historical data.

[0020] Furthermore, a distributed storage strategy is implemented. For each piece of low-frequency historical data, the storage node for that piece of low-frequency historical data is determined. If the number of storage nodes is equal to 1, the sub-network to which that storage node belongs is selected. If the number of storage nodes is not equal to 1, the sub-network to which the storage node with the highest trust level belongs is selected.

[0021] Obtain the trust level of all nodes in the selected sub-network, pre-set a trust level threshold, compare the trust level of each node with the trust level threshold, if the trust level of a node is greater than the trust level threshold, then classify the node as a high-trust node; otherwise, classify the node as a low-trust node.

[0022] For high-trust nodes in the selected sub-network, a complete copy of the low-frequency historical data is saved. For low-trust nodes in the selected sub-network, the low-frequency historical data is hashed to generate a hash value, which is then saved.

[0023] Furthermore, load information metrics of the blockchain sub-networks are collected, including network bandwidth utilization, CPU utilization, and memory utilization. The network load value of each sub-network is calculated through weighted averages. If the network load value of a sub-network exceeds the network load threshold, the number of nodes participating in the accounting of that sub-network is increased by a fixed proportion until the network load value does not exceed the network load threshold or reaches the maximum number of nodes. Conversely, the number of nodes participating in the accounting of that sub-network is reduced by a fixed proportion, while ensuring that the network load value of that sub-network does not exceed the network load threshold.

[0024] Furthermore, when the network load value of a subnetwork exceeds the network load threshold, a fixed proportion of nodes with the highest trust levels in the subnetwork are selected for consensus based on their trust ranking. When the network load value of a subnetwork does not exceed the network load threshold, a full-node consensus mechanism is adopted.

[0025] A blockchain-based financial data management system includes:

[0026] The network partitioning and optimization module collects node information from all nodes in the blockchain network and, based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, divides the blockchain network into multiple sub-networks.

[0027] The node trust assessment module collects historical financial data, node behavior records, and interaction data with other nodes from all nodes in the blockchain network, defines multiple trust evaluation indicators, and calculates the trust level of each node based on the value of its trust evaluation indicators.

[0028] The data storage strategy module extracts all transaction data from the blockchain network and classifies the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, a full-node storage strategy is implemented, and for low-frequency historical data, a distributed storage strategy is implemented.

[0029] The network load optimization module collects load information metrics of the blockchain sub-networks, calculates the network load value of each sub-network, compares the network load value of each sub-network with the network load threshold, and takes corresponding actions based on the comparison results.

[0030] (III) Beneficial Effects

[0031] This invention provides a blockchain-based financial data management system and method, which has the following beneficial effects:

[0032] (1) By dividing the blockchain network into multiple sub-networks, the storage resources of each node can be utilized more effectively. Nodes in each sub-network can share the storage task, thereby dispersing the storage pressure, improving the overall storage efficiency, reducing the communication distance and number of hops between nodes, and thus reducing network communication latency.

[0033] (2) By defining multiple trust evaluation indicators and calculating the trust of nodes, trustworthy nodes can be screened out. These trustworthy nodes are more likely to comply with the rules in financial data management, reducing the risk of data being tampered with or maliciously attacked. Nodes with high trust are more likely to stay online in the network and provide stable services, thereby enhancing the stability of the entire blockchain network.

[0034] (3) By classifying transaction data into high-frequency transaction data and low-frequency historical data, and implementing full-node storage strategy and distributed storage strategy respectively, storage space can be utilized more effectively. Since high-frequency transaction data needs to be accessed frequently, full-node storage ensures the rapid accessibility of the data; while low-frequency historical data reduces unnecessary duplicate storage through distributed storage, thereby reducing the overall storage cost. By storing low-frequency historical data on nodes with high trust and storing the hash value of the data only on nodes with low trust, not only is the waste of storage space reduced, but also the rational use of computing resources and network resources is promoted, avoiding unnecessary computing power consumption and resource waste.

[0035] (4) By dynamically adjusting the number of nodes participating in the accounting, it is helpful to optimize the resource utilization of the blockchain network. When the load is heavy, increasing the number of nodes can make full use of network resources. When the load is light, reducing the number of nodes can reduce operating costs. When the network load is high, selecting a fixed proportion of nodes with high trust ranking for consensus can significantly improve transaction processing speed and efficiency. When the network load is low, adopting a multi-node or full-node consensus mechanism can help maintain the decentralization and security of the blockchain network. Attached Figure Description

[0036] Figure 1 This is a schematic diagram illustrating the steps of the blockchain-based financial data management method of the present invention.

[0037] Figure 2 This is a schematic diagram of the blockchain structure of the present invention;

[0038] Figure 3 This is a schematic diagram of the blockchain-based financial data management system of the present invention. Detailed Implementation

[0039] 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.

[0040] Please see Figures 1-2 This invention provides a blockchain-based financial data management method, comprising the following steps:

[0041] Step 1: Collect node information of all nodes in the blockchain network. Based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, divide the blockchain network into multiple sub-networks.

[0042] Step one includes the following:

[0043] Step 101: Collect node information for all nodes in the blockchain network. Node information includes, but is not limited to, the node's geographical location (such as country, city, data center location, etc.), network bandwidth (including uplink and downlink bandwidth), computing power (such as CPU model, number of cores, memory size, disk read and write speed, etc.), and historical transaction records.

[0044] Step 102: Preprocess the collected node information, including data cleaning (removing duplicate, invalid or abnormal data), data formatting (unifying data format and standards), and data normalization (converting data of different dimensions to the same dimension for subsequent analysis).

[0045] Step 103: Define the objective function, including minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, and set constraints: each node belongs to one and only one subnetwork, and construct the optimization model:

[0046] Constraints: Where D represents minimizing network communication latency, Represents a node i and nodes j Communication delay between them N Represents a set of nodes. , as well as It is a binary variable. Represents a node i and nodes j Are they in the same subnetwork? Represents a node i Does it belong to a subnetwork? k , Indicates whether the task is assigned to a sub-network. k Nodes in i , E This indicates maximizing data storage efficiency. Represents a node i Storage capacity, R Represents the set of subnetworks. U This represents maximizing the utilization of computing power. Represents a node i Computational power;

[0047] It should be noted that computing power can be obtained through benchmark scores. Benchmarking is a method to evaluate the computing power of a node by running specific tasks. Common benchmarks include CPU benchmarks (such as SPECCPU2006), memory benchmarks (such as STREAM), and disk benchmarks (such as DD or FIO). The final computing power score is obtained by weighting the scores of different benchmarks.

[0048] Step 104: Select a solution algorithm (such as branch and bound, genetic algorithm, etc.) to solve the optimization model and obtain the optimal solution or near-optimal solution, that is, the partitioning scheme of which sub-network each node belongs to, and intelligently divide the blockchain network into multiple sub-networks;

[0049] When using this method, refer to steps 101 to 104:

[0050] By dividing the blockchain network into multiple sub-networks, the storage resources of each node can be utilized more effectively. Nodes within each sub-network can share storage tasks, thereby distributing storage pressure, improving overall storage efficiency, reducing communication distance and hop count between nodes, and thus reducing network communication latency.

[0051] Step 2: Collect historical financial data, node behavior records, and interaction data with other nodes from all nodes in the blockchain network, define multiple trust evaluation indicators, and calculate the trust level of each node based on the value of its trust evaluation indicators.

[0052] Step two includes the following:

[0053] Step 201: Collect historical financial data, node behavior records, and interaction data with other nodes from all nodes in the blockchain network, including node transaction success rate, transaction response time, node online duration, abnormal behavior records, etc.

[0054] Step 202: Based on the node's historical financial data, node behavior records, and interaction data with other nodes, define multiple trust evaluation indicators, including transmission success rate, node online time rate, battery duration ratio, number of communication connections, available storage capacity ratio (benefit-oriented attribute, the higher the value, the better), as well as abnormal data ratio, packet loss rate, identity fraud ratio, bit error rate, and link establishment latency (cost-oriented attribute, the lower the value, the better), etc.

[0055] Transmission success rate: refers to the proportion of data successfully transmitted by a node, reflecting the communication stability and reliability of the node; Node online time rate: refers to the proportion of time a node is online to the total time, reflecting the node's activity and availability; Battery life ratio (applicable to mobile devices or IoT nodes): refers to the proportion of time a node's battery lasts to the total charging time, reflecting the node's energy management efficiency; Number of communication connections: refers to the number of communication connections a node establishes simultaneously, reflecting the node's communication capabilities and network participation; Available storage capacity ratio: refers to the proportion of a node's currently available storage capacity to the total storage capacity, reflecting the node's storage capacity and resource utilization.

[0056] Abnormal data ratio: refers to the proportion of abnormal data generated by a node to the total data, reflecting the node's data quality and stability. Packet loss rate: refers to the proportion of data packets lost by a node during transmission, reflecting the node's communication quality and network stability. Identity fraud ratio: refers to the proportion of nodes impersonating other nodes or committing identity fraud in the blockchain network, reflecting the node's integrity and security. Bit error rate: refers to the proportion of erroneous data generated by a node during transmission, reflecting the node's communication accuracy and data integrity. Link establishment latency: refers to the time required for a node to establish a communication link, reflecting the node's response speed and network efficiency.

[0057] Step 203: Standardize the trust evaluation indicators. For trust evaluation indicators with benefit-related attributes, the standardized calculation formula is as follows: For trust evaluation indicators of cost-related attributes, the standardized calculation formula is as follows: ,in, St_b The standard value of the trust evaluation index representing the benefit-oriented attribute. St_c The standardized value of the trust rating index representing cost-related attributes. X This represents the trust rating index value. Max , Min These represent the maximum and minimum values ​​of the trust rating index, respectively.

[0058] Benefit-oriented attributes: These are trust evaluation indicators where higher values ​​are better, such as transmission success rate, node online time rate, battery life duration ratio, number of communication connections, and available storage capacity ratio. Cost-oriented attributes: These are trust evaluation indicators where lower values ​​are better, such as abnormal data ratio, packet loss rate, identity fraud ratio, bit error rate, and link establishment delay.

[0059] Step 204: For each node, calculate its trust level based on the value of its trust evaluation index. This can be done using a weighted average method or other suitable algorithms. For example, a weighted average method can be used to calculate the node's trust level. ,in, Tr Indicates the trust level of a node. Indicates the first m Standardized values ​​of trust evaluation indicators Indicates the first m The weights corresponding to each trust evaluation indicator, M represents the number of trust evaluation indicators;

[0060] It should be noted that the weights can be set by constructing a hierarchical structure model using the analytic hierarchy process (AHP), decomposing the complex problem into multiple levels and factors, and then using comparison matrices and consistency checks to determine the weights, or by determining the weights based on the information entropy of each indicator.

[0061] When using this method, refer to steps 201 to 204:

[0062] By defining multiple trust evaluation metrics and calculating the trust level of nodes, trustworthy nodes can be screened out. These trustworthy nodes are more likely to comply with rules in financial data management, reducing the risk of data tampering or malicious attacks. Nodes with high trust levels are more likely to remain online in the network and provide stable services, thereby enhancing the stability of the entire blockchain network.

[0063] Step 3: Extract all transaction data from the blockchain network and classify the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, implement a full-node storage strategy, and for low-frequency historical data, implement a distributed storage strategy.

[0064] Step three includes the following:

[0065] Step 301: Use a blockchain explorer (such as Etherscan, Blockchain.info, etc.) or API (such as the API provided by Alchemy, Infura, etc.) to extract all transaction data from the blockchain network, count the number of times each transaction data is accessed, and calculate its average access frequency.

[0066] Step 302: Pre-set an access frequency threshold. Compare the average access frequency of each transaction data with the access frequency threshold. If the average access frequency of the transaction data is greater than the access frequency threshold, classify the transaction data as high-frequency transaction data. Otherwise, classify the transaction data as low-frequency historical data. For high-frequency transaction data, implement a full-node storage strategy, that is, each node (or most nodes) will store a complete copy of these data. For low-frequency historical data, implement a distributed storage strategy.

[0067] The access frequency threshold determines which transaction data is considered high-frequency transaction data and which is considered low-frequency historical data. The access frequency threshold should be set according to the specific situation of the blockchain network and business needs. For example, if the blockchain network is mainly used to process high-frequency transactions (such as financial transactions), the access frequency threshold may be set relatively high; if it is mainly used to store historical data (such as document storage), the access frequency threshold may be set relatively low.

[0068] Step 303: Execute the distributed storage strategy. For each low-frequency historical data, determine the storage node of the low-frequency historical data. If the number of storage nodes is equal to 1, select the sub-network to which the storage node belongs. If the number of storage nodes is not equal to 1, select the sub-network to which the storage node with the highest trust level belongs.

[0069] Step 304: Obtain the trust level of all nodes in the selected sub-network, pre-set the trust level threshold, compare the trust level of each node with the trust level threshold, if the trust level of a node is greater than the trust level threshold, then classify the node as a high-trust node; otherwise, classify the node as a low-trust node.

[0070] The trust threshold setting, similar to the access frequency threshold, determines which nodes are considered high-trust nodes and which are considered low-trust nodes. An initial trust threshold is set, which can be a relatively conservative value. Through experiments and tests, the trust threshold is adjusted to determine the final trust threshold.

[0071] Step 305: For high-trust nodes in the selected sub-network, save a complete copy of the low-frequency historical data; for low-trust nodes in the selected sub-network, perform a hash operation on the low-frequency historical data to generate a hash value, and save the hash value.

[0072] When using this method, refer to steps 301 to 305:

[0073] By categorizing transaction data into high-frequency transaction data and low-frequency historical data, and implementing full-node storage and distributed storage strategies respectively, storage space can be utilized more effectively. High-frequency transaction data requires frequent access, and full-node storage ensures rapid data accessibility. Meanwhile, low-frequency historical data reduces unnecessary duplicate storage through distributed storage, thereby lowering the overall storage cost. By storing low-frequency historical data on nodes with high trust levels and only storing the hash values ​​of the data on low-trust nodes, not only is storage space wasted, but the rational use of computing and network resources is also promoted, avoiding unnecessary computing power consumption and resource waste.

[0074] Step 4: Collect load information indicators of blockchain sub-networks, calculate the network load value of each sub-network, compare the network load value of each sub-network with the network load threshold, and take corresponding actions based on the comparison results.

[0075] Step four includes the following:

[0076] Step 401: Use monitoring tools to collect load information indicators of the blockchain sub-network in real time, including network bandwidth utilization, CPU utilization, memory utilization, etc., and calculate the network load value of each sub-network based on the load information indicators through weighted calculation.

[0077] Step 402: Pre-set a network load threshold. Compare the network load value of each sub-network with the network load threshold. If the network load value of a sub-network exceeds the network load threshold, increase the number of nodes participating in the accounting of that sub-network by a fixed percentage (e.g., 5% of the current total number of nodes) until the network load value does not exceed the network load threshold or reaches the maximum number of nodes. Conversely, decrease the number of nodes participating in the accounting of that sub-network by a fixed percentage, and ensure that the network load value of that sub-network does not exceed the network load threshold.

[0078] The setting of network load threshold directly affects the system's performance and resource utilization efficiency. It should be adjusted according to the specific business needs of the financial data management system, such as transaction processing speed and real-time requirements. If the system has high requirements for transaction processing speed, the load threshold can be appropriately reduced to trigger the increase in the number of nodes earlier.

[0079] Step 403: When the network load value of the sub-network exceeds the network load threshold, based on the trust ranking of all nodes in the sub-network, a fixed proportion of the top-ranked nodes are selected for consensus, such as the top 20%, to improve transaction processing speed and efficiency. When the network load value of the sub-network does not exceed the network load threshold, a multi-node or full-node consensus mechanism is adopted to maintain the decentralization and security of the network.

[0080] When using this method, please refer to the content of steps 401 to 403:

[0081] By dynamically adjusting the number of nodes participating in the accounting process, it is helpful to optimize the resource utilization of the blockchain network. Adding nodes when the load is heavy can make full use of network resources, while reducing nodes when the load is light can reduce operating costs. When the network load is high, selecting a fixed proportion of nodes with high trust ranking for consensus can significantly improve transaction processing speed and efficiency. When the network load is low, adopting a multi-node or full-node consensus mechanism helps maintain the decentralization and security of the blockchain network.

[0082] Please see Figure 3 The present invention also provides a blockchain-based financial data management system, comprising: a network partitioning and optimization module, a node trust assessment module, a data storage strategy module, and a network load optimization module; wherein,

[0083] The network partitioning and optimization module collects node information from all nodes in the blockchain network and, based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, divides the blockchain network into multiple sub-networks.

[0084] The node trust assessment module collects historical financial data, node behavior records, and interaction data with other nodes from all nodes in the blockchain network, defines multiple trust evaluation indicators, and calculates the trust level of each node based on the value of its trust evaluation indicators.

[0085] The data storage strategy module extracts all transaction data from the blockchain network and classifies the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, a full-node storage strategy is implemented, and for low-frequency historical data, a distributed storage strategy is implemented.

[0086] The network load optimization module collects load information metrics of the blockchain sub-networks, calculates the network load value of each sub-network, compares the network load value of each sub-network with the network load threshold, and takes corresponding actions based on the comparison results.

[0087] In the application, the various formulas mentioned are all calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The coefficients in the formulas are set by those skilled in the art according to the actual situation.

[0088] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, and combinations thereof. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0089] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0090] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A blockchain-based financial data management method, characterized in that: include: Collect node information of all nodes in the blockchain network, and divide the blockchain network into multiple sub-networks based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization. Historical financial data, node behavior records, and interaction data with other nodes are collected from all nodes in the blockchain network. Multiple trust evaluation indicators are defined, and for each node, the trust level of the node is calculated based on the value of its trust evaluation indicator. Extract all transaction data from the blockchain network and categorize the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, implement a full-node storage strategy, and for low-frequency historical data, implement a distributed storage strategy. Collect load information metrics from blockchain sub-networks, calculate the network load value for each sub-network, compare the network load value of each sub-network with the network load threshold, and take appropriate actions based on the comparison results: When the network load of a subnetwork exceeds the network load threshold, a fixed proportion of nodes with the highest trust levels in the subnetwork are selected for consensus. When the network load of a subnetwork does not exceed the network load threshold, a full-node consensus mechanism is adopted.

2. The blockchain-based financial data management system and method according to claim 1, characterized in that: Define the objective function, including minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, and set constraints: each node belongs to one and only one subnetwork. Construct the optimization model: Constraints: Where D represents minimizing network communication latency, Represents a node i and nodes j Communication delay between them N Represents a set of nodes. , as well as It is a binary variable. Represents a node i and nodes j Are they in the same subnetwork? Represents a node i Does it belong to a subnetwork? k , Indicates whether the task is assigned to a sub-network. k Nodes in i , E This indicates maximizing data storage efficiency. Represents a node i Storage capacity, R Represents the set of subnetworks. U This represents maximizing the utilization of computing power. Represents a node i Computational power; The branch and bound method is used to solve the optimization model and determine the partitioning scheme of the approximate optimal solution, that is, the sub-network to which each node belongs, thus dividing the blockchain network into multiple sub-networks.

3. The blockchain-based financial data management system and method according to claim 1, characterized in that: Based on the node's historical financial data, node behavior records, and interaction data with other nodes, several trust evaluation indicators are defined, including transmission success rate, node online time rate, battery life duration ratio, number of communication connections, available storage capacity ratio, as well as abnormal data ratio, packet loss rate, identity fraud ratio, bit error rate, and link establishment latency.

4. The blockchain-based financial data management system and method according to claim 3, characterized in that: The trust evaluation indicators are standardized. For trust evaluation indicators with benefit-related attributes, the standardized calculation formula is as follows: For trust evaluation indicators of cost-related attributes, the standardized calculation formula is as follows: ,in, St_b The standard value of the trust evaluation index representing the benefit-oriented attribute. St_c The standardized value of the trust rating index representing cost-related attributes. X This represents the trust rating index value. Max , Min These represent the maximum and minimum values ​​of the trust rating index, respectively.

5. The blockchain-based financial data management system and method according to claim 4, characterized in that: For each node, calculate its trust level based on the value of its trust evaluation index: ,in, Tr Indicates the trust level of a node. Indicates the first m Standardized values ​​of trust evaluation indicators Indicates the first m The weights corresponding to each trust evaluation indicator, where M represents the number of trust evaluation indicators.

6. The blockchain-based financial data management system and method according to claim 1, characterized in that: The number of times each transaction data is accessed is counted, and its average access frequency is calculated. The average access frequency of each transaction data is compared with the access frequency threshold. If the average access frequency of the transaction data is greater than the access frequency threshold, the transaction data is classified as high-frequency transaction data; otherwise, the transaction data is classified as low-frequency historical data.

7. A blockchain-based financial data management system and method according to claim 6, characterized in that: Implement a distributed storage strategy. For each piece of low-frequency historical data, determine the storage node for that piece of low-frequency historical data. If the number of storage nodes is equal to 1, select the sub-network to which that storage node belongs. If the number of storage nodes is not equal to 1, select the sub-network to which the storage node with the highest trust level belongs. Obtain the trust level of all nodes in the selected sub-network, pre-set a trust level threshold, compare the trust level of each node with the trust level threshold, if the trust level of a node is greater than the trust level threshold, then classify the node as a high-trust node; otherwise, classify the node as a low-trust node. For high-trust nodes in the selected sub-network, a complete copy of the low-frequency historical data is saved. For low-trust nodes in the selected sub-network, the low-frequency historical data is hashed to generate a hash value, which is then saved.

8. The blockchain-based financial data management system and method according to claim 7, characterized in that: The system collects load information metrics for blockchain sub-networks, including network bandwidth utilization, CPU utilization, and memory utilization. The network load value for each sub-network is calculated using weighted averages. If the network load value of a sub-network exceeds the network load threshold, the number of nodes participating in the accounting process for that sub-network is increased by a fixed percentage until the network load value does not exceed the network load threshold or reaches the maximum number of nodes. Conversely, if the network load value does not exceed the threshold, the number of nodes participating in the accounting process for that sub-network is decreased by a fixed percentage, ensuring that the network load value of that sub-network does not exceed the network load threshold.

9. A blockchain-based financial data management system for implementing the method described in any one of claims 1 to 8, characterized in that: include: The network partitioning and optimization module collects node information from all nodes in the blockchain network and, based on the goals of minimizing network communication latency, maximizing data storage efficiency, and maximizing computing power utilization, divides the blockchain network into multiple sub-networks. The node trust assessment module collects historical financial data, node behavior records, and interaction data with other nodes from all nodes in the blockchain network, defines multiple trust evaluation indicators, and calculates the trust level of each node based on the value of its trust evaluation indicators. The data storage strategy module extracts all transaction data from the blockchain network and classifies the transaction data into high-frequency transaction data and low-frequency historical data. For high-frequency transaction data, a full-node storage strategy is implemented, and for low-frequency historical data, a distributed storage strategy is implemented. The network load optimization module collects load information metrics from the blockchain sub-networks, calculates the network load value for each sub-network, compares the network load value of each sub-network with the network load threshold, and takes appropriate actions based on the comparison results. When the network load of a subnetwork exceeds the network load threshold, a fixed proportion of nodes with the highest trust levels in the subnetwork are selected for consensus. When the network load of a subnetwork does not exceed the network load threshold, a full-node consensus mechanism is adopted.

Images