Cloud-based data processing and storage system and method

By introducing cloud data processing and storage systems into the transaction query platform, using real-time data warehouses and offline data warehouses to process high-frequency query data, the problem of massive data query pressure and time-consuming analysis is solved, and fast and efficient transaction query and data is achieved. statistics.

CN118689947BActive Publication Date: 2024-11-29SHANGHAI FUIOU PAYMENT SERVICE
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411178773.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-29
Estimated Expiration
2044-08-27

AI Technical Summary

Technical Problem

When the existing transaction query platform processes massive data, it leads to excessive reading and writing pressure on the database, and the query takes too long, which affects the system performance and user experience, and the data analysis takes time to drag down the database performance.

Method used

Using cloud-based data processing and storage systems, by constructing query dimension buried point tables, defining high-frequency query dimensions, synchronizing relevant data into real-time data warehouses and offline data warehouses for analysis and summary, saving statistical result data, and building query Dimensional routing tables to optimize query paths.

Benefits of technology

It significantly improves the speed and efficiency of transaction query, reduces database load, improves data analysis performance, and avoids the drag on database performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118689947B_ABST
    Figure CN118689947B_ABST
Patent Text Reader

Abstract

The present application relates to a cloud-based data processing and storage method, including: 1) analyzing and recording the query dimensions of merchants to construct a query dimension tracking table; 2) performing statistical analysis on the query dimension tracking points of merchants to define the query dimensions of high-frequency queries; 3) analyzing and summarizing data in a data warehouse, including: synchronizing all meta-table data involved in the query dimensions defined as high-frequency queries from a self-built database to each data warehouse in the cloud through a data transmission service; executing a query SQL script associated with the query dimensions defined as high-frequency queries at each data warehouse and analyzing and summarizing the query results; 4) saving the analyzed statistical result data in the form of an intermediate table in a corresponding data warehouse; and 5) associating the query dimensions defined as high-frequency queries with the corresponding intermediate table to construct a query dimension routing table.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing and storage, and in particular to a solution for automated processing of massive data and intelligent storage of statistical result data in response to the needs of different business transaction queries. Background Art

[0002] When a transaction query platform or app queries transaction summary data of a specific dimension, it usually directly queries the massive full transaction table data in the self-built database, but this method will result in a long query time (current situation, reasons). This can easily lead to excessive database read and write pressure and performance degradation, which will be reflected in the transaction system as serious delays, no response, or even crashes, which will not only affect the normal operation of the underlying system, but also lead to a poor user experience.

[0003] Taking the QR code scanning and payment business scenario as an example, merchants log in to the query platform or query app to check the total amount and number of revenues last month or this month, the total amount and number of revenues last year and this year, and even the month-on-month and year-on-year data of the number of revenues. When multiple merchants access the database concurrently, a large number of requests will be made to access the database, and the relevant dimension statistics will be queried from the massive transaction data. If a large number of access requests are executed concurrently at the database at the same time, it will cause a single request to take too long, resulting in the database connection being full, and other problems, which will bring great read and write pressure to the database.

[0004] Therefore, it is hoped that a new data storage format can be constructed based on the full database to further simplify the operation of the transaction query and improve the speed and efficiency of the query.

[0005] At the same time, how to analyze and summarize the statistical results is also a problem that needs to be considered. If the data analysis task is placed in a self-built database, the analysis and summary will take a long time, dragging down the database performance and affecting other businesses' access to the database. Therefore, it is necessary to find another platform environment to complete the work of data analysis, summary and statistics.

[0006] In summary, we hope to provide a novel cloud-based data processing and storage system to achieve fast processing of transaction queries and efficient data statistical analysis. Summary of the invention

[0007] The present application relates to an intelligent storage and analysis solution that uses cloud-based processing and storage of massive data based on high-frequency queries.

[0008] According to a first aspect of the present application, a cloud-based data processing and storage method is provided, comprising:

[0009] 1) Analyze and record the merchant's query dimensions to build a query dimension tracking table;

[0010] 2) Conduct statistical analysis on the merchant’s query dimension embedding points to define the query dimensions of high-frequency queries;

[0011] 3) Analyze and summarize data in the data warehouse, including:

[0012] Synchronize all meta-table data involved in the query dimension defined as a high-frequency query from the self-built database to each data warehouse in the cloud through the data transmission service;

[0013] Executing, at each data warehouse, a query SQL script associated with the query dimension defined as a high-frequency query and analyzing and summarizing the query results;

[0014] 4) Save the analyzed statistical results in the form of an intermediate table to the corresponding data warehouse; and

[0015] 5) Associating the query dimension defined as a high-frequency query with the corresponding intermediate table to construct a query dimension routing table.

[0016] According to a second aspect of the present application, a cloud-based data processing and storage system is provided, comprising:

[0017] The client in the front-end server is configured to receive a transaction query request from a merchant;

[0018] A server node in a backend server, configured to process data queries for a merchant query dimension;

[0019] Distributed database module, including self-built database and cloud database, wherein the cloud database includes real-time data warehouse and offline data warehouse;

[0020] The data analyzer is configured to execute the cloud-based data processing and storage method as described in the first aspect.

[0021] This Summary is provided to introduce some concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to describe the manner in which the above and other advantages and features of the present invention can be obtained, a more particular description of the invention briefly described above will be presented by reference to specific embodiments of the invention shown in the accompanying drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings, in which:

[0023] Figure 1The schematic system hierarchy of a traditional transaction query platform that does not use the solution of the present application is shown.

[0024] Figure 2 Shown based on Figure 1 The transaction query processing flow of the transaction query system in .

[0025] Figure 3 A schematic system hierarchy of a cloud-based data processing and storage system according to an embodiment of the present application is shown.

[0026] Figure 4 A schematic flow chart of a cloud-based data processing and storage method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] This application can be applied to fields that require real-time and extremely high-frequency access to massive amounts of data, such as financial payment, stock quotes, futures trading, etc. For convenience, in the following examples, the transaction environment of the payment industry is mainly used as an example for explanation.

[0028] For ease of understanding and comparison, Figure 1 The schematic system hierarchy of a traditional trading platform that does not use the solution of the present application is first introduced.

[0029] As shown in the figure, the figure can be divided into a three-layer hierarchy from top to bottom, namely, the front-end server (also known as the user layer), the back-end server (also known as the business layer) and the distributed database (also known as the data layer, which can also be called the distributed database module). Among them, the front-end server (user layer) includes a variety of clients and their devices that can be used by merchants, such as smartphones, personal computers, tablets, notebooks, etc. installed with the client. The back-end server (business layer) includes multiple business processing systems for processing various transaction queries. The distributed database module (data layer) includes a distributed database that stores the full amount of transaction data.

[0030] based on Figure 1 The traditional transaction processing process of the traditional trading platform is as follows Figure 2 First, the user at the user layer initiates a query to the front-end server through the query interface of the client, and the front-end server distributes the query conditions in the query request to the back-end server. The back-end server directly queries the transaction data table according to the query conditions and returns the statistical result data to the query interface in the original way.

[0031] from Figure 1 and Figure 2It is not difficult to see that all query requests are sent directly to the database, and the query is a massive amount of transaction data. Therefore, when there are massive concurrent query requests, the database will be overloaded with read and write pressure, performance will be reduced, and there will even be a risk of pseudo-death or downtime.

[0032] Therefore, if statistical result data containing only a small number of fields required by the platform query dimension can be summarized from the data in the transaction table in advance and stored for query, the above problem can be solved.

[0033] Therefore, a solution is needed to summarize the statistical result data from the transaction table in advance, which only contains a small number of fields required by the platform query dimension, for merchants or users to query. For example, merchants want to view the transaction data summarized for the day / month on the platform: total number of transactions, total amount, etc., or summarize data by different transaction types, such as WeChat, Alipay, Cloud QuickPass, etc.

[0034] To this end, the technical solution of the present application first uses the concept of "data warehouse" to store data.

[0035] Data warehouse (hereinafter referred to as "data warehouse") is a subject-oriented, integrated, relatively stable data set that reflects historical changes, mainly used to support decision-making in management. It is generated on the basis of the large number of existing databases in order to further explore data resources and meet decision-making needs. The data in the data warehouse is obtained by systematic processing, aggregation and organization based on the extraction and cleaning of the original scattered database data. The inconsistency in the source data must be eliminated to ensure that the information in the data warehouse is consistent global information about the entire enterprise. The data in the data warehouse is mainly used for enterprise decision-making analysis. The data operations involved are mainly data queries. Once a piece of data enters the data warehouse, it will generally be retained for a long time. The data in the data warehouse usually contains historical information, through which quantitative analysis and prediction of the development history and future trends of the enterprise can be made.

[0036] On the other hand, data warehouses can be divided into real-time data warehouses and offline data warehouses. Both real-time data warehouses and offline data warehouses are different types of data warehouses, and both are used to store and manage enterprise data, but they have obvious differences in the time, speed, and purpose of data processing and use.

[0037] Offline Data Warehouse is a system for storing and processing batch data. Its characteristic is that data processing and analysis are based on batch jobs, usually in units of longer time periods. The main application scenarios of offline data warehouses include applications that require historical data analysis and report generation, such as sales reports, monthly and annual financial statements, and applications with large data volumes and where processing time is not a critical issue.

[0038] Therefore, the offline data warehouse has the following features:

[0039] Batch processing: Offline data warehouses process data through batch jobs, which means that data is collected and stored within a certain period of time and then processed all at once.

[0040] High capacity: Offline data warehouses are usually designed to store large amounts of historical data.

[0041] High latency: Since data processing is batch processing, offline data warehouses are not suitable for applications that require real-time or near real-time data.

[0042] On the other hand, a real-time data warehouse is a system for storing and processing real-time data. Its main feature is that data processing and analysis are carried out instantly, and data enters the data warehouse almost immediately and can be used for analysis and decision-making immediately. The applicable scenarios of real-time data warehouses mainly include applications that require real-time monitoring of business indicators, such as financial transaction dashboards, real-time sales reports, online advertising analysis, etc.; and applications that require immediate action to respond to real-time events, such as abnormal monitoring large screens, real-time fraud detection, etc.

[0043] Therefore, the real-time data warehouse has the following features:

[0044] Low latency: Real-time data warehouses can capture and process data quickly after it is generated, usually at a speed of seconds or sub-seconds.

[0045] Data stream processing: Real-time data warehouses often use stream processing techniques to process data, which allows data to be transformed and calculated immediately when it enters the warehouse.

[0046] Real-time analytics: Data can be used for real-time monitoring, dashboards, forecasting, and decision support.

[0047] High throughput: Real-time data warehouses need to process large amounts of data streams and therefore require high throughput performance.

[0048] Complexity: Since real-time data warehouses need to process real-time data streams, their architecture and technology are usually complex.

[0049] Applicable scenarios for real-time data warehouses: Applications that require real-time monitoring of business indicators, such as financial transaction dashboards, real-time sales reports, online advertising analysis, etc. Applications that require immediate action to respond to real-time events, such as anomaly monitoring large screens, real-time fraud detection, etc.

[0050] In order to overcome the problem that all the above query requests focus on querying massive transaction table data,

[0051] This application provides a smart storage and analysis processing solution for cloud data warehouse analysis and aggregation of high-frequency query dimension data. In other words, the solution of this application introduces real-time data warehouse and offline data warehouse in the transaction data processing and storage system to meet different transaction data processing needs, thereby making full use of their respective advantages to achieve fast and efficient transaction query.

[0052] First, in Figure 3 Schematic hierarchical structure of a data processing and storage system based on querying a cloud database (also referred to as “cloud”) according to the present application is shown in FIG.

[0053] The hierarchy includes:

[0054] The client in the front-end server is configured to receive a transaction query request from a merchant; the client may be a mobile device, PC, tablet or notebook, etc., with a corresponding program or APP installed.

[0055] A server node in a backend server, configured to process data queries for a merchant query dimension;

[0056] The database in the (distributed) database includes a self-built database and a cloud database. Among them, the self-built database is a database built by the enterprise itself, where the original sql script and meta table are located, which is used to store all transaction-related data and perform query operations. Unlike the self-built database in the traditional technology, which is used to perform a large number of transaction queries on its own massive data, this application puts the sql scripts corresponding to some high-frequency query dimensions from the self-built library into the data warehouse for execution, and saves the statistical result data after execution in the data warehouse into the intermediate table. The next time the same query dimension can be queried in the intermediate table, there is no need to query in the self-built database. Therefore, the main responsibility of the self-built database of this application is to serve as a backup storage, and transaction query operations will not be frequently performed. Most of the query operations (operations in high-frequency query dimensions) are transferred to the cloud database. The cloud database can be an online data warehouse provided by a third-party platform, which is configured to store the statistical result data after data warehouse analysis and processing and perform corresponding queries. According to the different characteristics of the stored statistical result data, the cloud database can be further divided into a real-time data warehouse and an offline data warehouse, wherein the offline data warehouse is configured to store data with a large amount of data and a relatively low query frequency in the statistical results, while the real-time data warehouse is configured to store data with a small amount of data and a relatively high query frequency in the statistical results.

[0057] In addition to the above structures, the transaction query platform system also includes a data analysis layer and a data analyzer therein, wherein: the data analyzer is configured to perform point tracking and statistical analysis on the merchant's query request, and analyze and summarize the massive transaction data in the corresponding data warehouse. The specific execution method flow of the data analyzer will be described in detail below.

[0058] Generally speaking, the cloud-based data processing and storage method of the present application generally includes the following processes:

[0059] 1. Construct a query dimension tracking table to record query dimensions (query conditions), query SQL scripts and associated meta tables. The query dimension tracking table is also the basis for defining high-frequency query dimensions later.

[0060] 2. According to the data records in the query dimension table, some query dimensions are determined to be high-frequency query dimensions. For each high-frequency query dimension, its SQL script is transferred to the data warehouse (real-time data warehouse and offline data warehouse) for execution; the statistical result data obtained by executing the SQL script is saved in the intermediate table. Different SQL scripts construct different intermediate tables.

[0061] 3. Associate the query dimension with the intermediate table name and its data source one by one to build a query dimension routing table.

[0062] After completing the above method flow, when the merchant at the user level initiates a query again, the query dimension in the query request can be matched with the query dimension routing table. If the match is successful, the query request can be directed to the corresponding intermediate table on the data source recorded in the query dimension routing table to save time and improve efficiency. If the match is not successful, the query request is still executed on the meta-table data in the self-built database, which is the same as the traditional query.

[0063] Figure 4 The schematic flow of the cloud-based data processing and storage method of the present application is further described in detail.

[0064] Step 1: Analyze and record the merchant's query dimensions to build a query dimension tracking table.

[0065] First, the merchant at the user layer initiates a query request through the client on the query platform, and the query request is forwarded by the client to the backend server. The server node at the backend server collects the merchant's query conditions from the received query request, for example: query the total revenue of a certain day, the total revenue of Alipay collection on a certain day, etc. Subsequently, the server node records the collected query conditions and the date of the day in association, and saves the execution SQL script of this query, records the number of executions of this query on the day and the time consumed by this query, for the subsequent statistics and analysis of high-frequency query dimensions.

[0066] These recorded data can be called query dimension tracking. Moreover, the data to be recorded can be saved in the form of a table (which can be called a "query dimension tracking table"), as shown in Table 1:

[0067] Table 1

[0068]

[0069] As shown in Table 1, the query dimension tracking table records the query dimension (query_condition, query condition), the corresponding query SQL script and the associated meta table, which is also the basis for defining high-frequency query dimensions later.

[0070] It should be understood that the fields and field names in the table are merely exemplary and not limited thereto. The technician can add, delete or adjust the query dimension according to actual needs to obtain the required query dimension tracking table, which all fall within the scope of protection required by this application.

[0071] By constructing the query dimension tracking table, subsequent statistical analysis of high-frequency query dimensions is facilitated.

[0072] Step 2: Perform statistical analysis on the merchant’s query dimension tracking points to define high-frequency query dimensions.

[0073] The statistical analysis includes: high-frequency query dimensions in statistical query dimensions. As the name implies, high-frequency query dimensions refer to query dimensions that are queried more frequently. Using the query dimension tracking table constructed in step 1, query dimensions that are determined to have relatively high query time and high query frequency can be defined as high-frequency query dimensions based on the tracking data collected every day. A high query time indicates that the statistical result data volume is large, and a high query frequency means that the number of queries is large.

[0074] Therefore, still taking Table 1 as an example, the definition standard of the high-frequency query dimension can be as follows:

[0075] 1. Queries that take less than or equal to 1000 milliseconds (query_time≤1000ms) and have less than or equal to 10,000 queries per day (query_count≤10,000) are not defined as high-frequency queries;

[0076] 2. Queries that take less than or equal to 1,000 milliseconds and have more than or equal to 10,000 queries per day are not defined as high-frequency queries;

[0077] 3. Queries that take longer than 1,000 milliseconds and have a query count of less than or equal to 10,000 times per day are not defined as high-frequency queries;

[0078] 4. Queries that take longer than 1,000 milliseconds and have more than or equal to 10,000 queries per day are defined as high-frequency queries.

[0079] It should be understood that the above-mentioned time-consuming threshold of 1000 milliseconds and the query number threshold of 10,000 times are given as examples. In actual applications, technicians can make appropriate adjustments based on factors such as transaction volume, server processing capacity, network bandwidth, etc. The adjustments are all within the scope of protection of this application.

[0080] For query dimensions defined as high-frequency query dimensions, the corresponding query SQL scripts need to be placed in the data warehouse for asynchronous execution to perform data analysis and summary processing.

[0081] Step 3: Analyze and summarize data in the data warehouse.

[0082] First, all meta-table data involved in the query dimension defined as a high-frequency query is synchronized from the self-built database to the data warehouse through the data transmission service. The meta-table data refers to the table data involved in the original SQL script (generally associated with multiple tables, such as transaction tables, merchant tables, etc.). According to different data types and query requirements, the synchronization can be divided into:

[0083] Synchronize meta-table data with high real-time requirements, recent data, and relatively small data volume to the real-time data warehouse;

[0084] Synchronize meta-table data with relatively large data volume, historical data, and low real-time requirements to the offline data warehouse.

[0085] Among them, "recent" generally refers to data from this year or the past year (the past 12 months). In actual applications, transaction queries in most scenarios are for recent data, and the amount of data in the meta table is small.

[0086] Among them, "historical" generally refers to data older than one year, and the amount of meta-table data in this case is large. In actual applications, there are of course transaction queries that query transactions from a few years ago, such as transaction summary queries within 5 or 10 years. This type of transaction query analysis is slow to execute, but the query scenario does not occur so frequently.

[0087] Next, the query SQL script associated with the query dimension defined as a high-frequency query is executed at each data warehouse, and the query results are analyzed, summarized and counted.

[0088] By transferring the query SQL script to the data warehouse, the data processing performance advantages of the data warehouse can be fully utilized to make SQL execution faster.

[0089] Step 4: Save the statistical results of the data warehouse analysis to the corresponding data warehouse in the cloud database, where:

[0090] The statistical results have a large amount of data, and data with relatively low query frequency is saved in the offline data warehouse;

[0091] The amount of statistical results is small, and data with relatively high query frequency is saved in the real-time data warehouse;

[0092] Among them, the statistical result data is also called intermediate table data. Different scripts obtain different intermediate tables, which contain the query dimensions of the merchant and the required query results, such as the merchant's business number, transaction date, payment method and total amount, etc. The data format can be referred to in the following Table 2:

[0093] Table 2

[0094]

[0095] Likewise, it should be understood that the fields and field names in the table are merely exemplary and not limiting. Technicians can add, delete or adjust the query dimensions and query results in the table according to actual needs, which all fall within the scope of protection required by this application.

[0096] By constructing an intermediate table corresponding to the high-frequency query dimension based on the statistical result data after executing the SQL script of different high-frequency query dimensions, it can be used to replace the massive meta-table data to execute the query to improve the query efficiency.

[0097] Step 5: Associate the query dimension defined as a high-frequency dimension with the corresponding intermediate table to build a query dimension routing table

[0098] The query dimension defined as a high-frequency query, the table name of the corresponding intermediate table, and the data source of the intermediate table (for example, a real-time data warehouse or an offline data warehouse) are associated and uniformly placed in the query dimension routing table to maintain their association relationship. For an example of the query dimension routing table, see Table 3:

[0099] Table 3

[0100]

[0101] By using the query dimension routing table, query requests for high-frequency query dimensions can be conveniently routed to the data source where the corresponding intermediate table is located to execute scripts on the intermediate table, thereby reducing massive queries on the self-built database.

[0102] For each query dimension defined as a high-frequency query, steps 3-5 are executed in sequence to construct a corresponding query dimension routing table for each high-frequency query dimension. At this point, the schematic flow of the cloud-based data processing and storage method of the present application is basically completed.

[0103] After the cloud-based data processing and storage method is used to realize features such as real-time data warehouse, offline data warehouse, query dimension tracking table, intermediate table, and query dimension routing table, when receiving subsequent query requests, the server can process the merchant query request according to the following process, that is, the back-end system verifies the query conditions in this query request to determine whether it is necessary to read the data in the intermediate table, including:

[0104] If there is no matching query dimension routing table, this query does not need to query the intermediate table data, and the process still follows the original logic, that is, directly querying the transaction table and other data in the self-built database;

[0105] If there is a matching query dimension routing table, this query can be implemented by querying the intermediate table data, that is, obtaining the corresponding intermediate table name and its data source from the matching query dimension routing table, connecting to the data source, and directly executing the query on the intermediate table on the data source to replace querying the massive data in the self-built database.

[0106] Since the intermediate table data only contains statistical result data of high-frequency query dimensions, its data volume is much smaller than the meta-table data in the self-built database. Therefore, the query speed and efficiency based on the intermediate table data are much higher than the corresponding operation on the self-built database.

[0107] Although various embodiments are described above, it should be understood that they are intended to be exemplary rather than limiting. Those skilled in the relevant art(s) will appreciate that various modifications may be made in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims. Therefore, the breadth and scope of the present invention disclosed herein should not be limited by the exemplary embodiments disclosed above, but should be defined only in accordance with the appended claims and their equivalents.

Claims

1. A cloud-based data processing and storage method, comprising: 1) Analyze and record the merchant's query dimensions to build a query dimension tracking table; 2) Conduct statistical analysis on the merchant’s query dimension embedding points to define the query dimensions of high-frequency queries; 3) Analyze and summarize data in the data warehouse, including: Synchronize all meta-table data involved in the query dimension defined as a high-frequency query from the self-built database to each data warehouse in the cloud through the data transmission service; Executing, at each data warehouse, a query SQL script associated with the query dimension defined as a high-frequency query and analyzing and summarizing the query results; 4) Save the analyzed statistical results in the form of an intermediate table to the corresponding data warehouse; and 5) associating the query dimension defined as a high-frequency query with the corresponding intermediate table to construct a query dimension routing table; Wherein, the method further comprises: When a subsequent query request is received, the query condition in the query request is verified to determine whether the data in the intermediate table needs to be read, including: If there is no matching query dimension routing table, the query request does not need to query the data in the intermediate table, and directly performs transaction query in the self-built database; If there is a matching query dimension routing table, the corresponding intermediate table name and the data source where the intermediate table is located are obtained from the matching query dimension routing table, the data source is connected, and the query on the intermediate table is directly executed on the data source.

2. The method according to claim 1, characterized in that The data warehouse includes a real-time data warehouse and an offline data warehouse.

3. The method according to claim 1, characterized in that The query dimension tracking table records the query conditions and is used to record the query SQL script and the associated meta table.

4. The method according to claim 1, characterized in that The intermediate table contains the merchant's query dimensions and the required query results.

5. The method according to claim 1, characterized in that The query dimension routing table includes the query dimension defined as a high-frequency query, the corresponding intermediate table name, and the data source where the intermediate table is located.

6. The method according to claim 1, wherein: For each query dimension defined as a high-frequency query, steps 3) to 5) are performed.

7. A cloud-based data processing and storage system, comprising: The client in the front-end server is configured to receive a transaction query request from a merchant; A server node in a backend server, configured to process data queries for a merchant query dimension; Distributed database module, including self-built database and cloud database, wherein the cloud database includes real-time data warehouse and offline data warehouse; A data analyzer is configured to execute the cloud-based data processing and storage method as claimed in claim 1.

Citation Information

Patent Citations

  • Data query method and device and electronic equipment

    CN112784136A

  • High-frequency library and data hierarchical storage and query method and system using high-frequency library

    CN116610680A