Unified management method and system for integrating time-series databases and data platforms

Abstract

The application relates to the technical field of time series databases, and discloses a unified management method and system for integrating a time series database and a data platform, which comprises the following steps: matching tree model path data with storage group list data by using a longest prefix matching algorithm; performing hierarchical splitting on the tree model path data based on storage group identification data; generating tree model table creation statements and table model table creation statements based on relational model structure data, and performing verification based on preset table creation constraints; performing syntax analysis and semantic verification on structured query language query statements, detecting time conditions based on a syntax tree that passes the verification, and performing rewriting processing to obtain optimized structured query language statement data; listening to permission change events and generating synchronization instructions to obtain synchronization result data, and generating unified management state data. The application solves the problems of difficulty in integrating a time series database and a data platform, lack of standardized mapping between a tree model and a relational model, and scattered operation and maintenance management.

Description

Technical Field

[0001] This invention relates to the field of time-series database technology, specifically to a unified management method and system for integrating time-series databases and data platforms. Background Technology

[0002] With the rapid development of IoT technology, massive amounts of time-series data have been generated in fields such as industrial manufacturing, energy and power, smart cities, and transportation. Time-series databases, due to their storage structures optimized for time-series data, high write performance, and compression capabilities, have become the preferred solution for storing and managing time-series data. However, in practical applications, significant architectural differences and integration gaps exist between time-series databases and enterprise data platforms.

[0003] While some solutions exist for converting tree models to table models, these solutions primarily focus on internal database model conversion, achieving bidirectional conversion between tree and table models through virtual models or mounted models. They do not address the issue of deep integration between time-series databases and enterprise data platforms. At the database and platform integration level, there is still a lack of standardized metadata mapping mechanisms, unified table creation specifications, intelligent query optimization methods, and a unified access control system.

[0004] Therefore, there is an urgent need for a unified management method and system that can achieve deep integration of time-series databases and data platforms to reduce integration costs, improve management efficiency, and ensure data security. Summary of the Invention

[0005] In view of this, the present invention provides a unified management method and system for integrating time-series databases and data platforms, in order to solve the problems in the prior art such as difficulty in integrating time-series databases with enterprise data platforms, lack of standardization in metadata mapping, fragmented operation and maintenance management, low development efficiency, and invisible data assets.

[0006] In a first aspect, the present invention provides a unified management method for integrating a time-series database and a data platform. The method includes: acquiring time-series database instance configuration information; establishing a data connection channel based on the instance configuration information, wherein the data connection channel connects the data platform and the time-series database; acquiring storage group list data and tree model path data from the time-series database based on the data connection channel; matching the tree model path data and storage group list data using a longest prefix matching algorithm to obtain storage group identifier data; performing hierarchical splitting of the tree model path data based on the storage group identifier data to obtain relational model structure data; generating tree model table creation statement data and table model table creation statement data based on the relational model structure data; and performing processing on the tree model table creation statement data and table model table creation statement data based on preset table creation constraints. The process involves: row validation to obtain validated table structure definition data; acquiring structured query language (SCL) query statement data, parsing the SCL query statement data to obtain syntax tree data, performing semantic validation on the syntax tree data based on the table structure definition data, obtaining validated syntax tree data; detecting time conditions based on the validated syntax tree data, rewriting the SCL query statement data based on the detection results, obtaining optimized SCL statement data; monitoring permission change event data on the data platform side, generating permission synchronization instruction data based on the permission change event data, sending the permission synchronization instruction data to the time-series database via the data connection channel, obtaining synchronization result data; and generating unified management status data based on the optimized SCL statement data and the synchronization result data.

[0007] The unified management method for integrating time-series databases and data platforms provided in this embodiment has the following advantages: It establishes a data connection channel by acquiring the time-series database instance configuration information, achieving standardized integration between the time-series database and the data platform, and solving the problem of inconsistent access methods for heterogeneous data sources; it matches tree model path data with storage group list data using the longest prefix matching algorithm, automatically identifying the storage group to which the path belongs, avoiding the defect that fixed-position path splitting cannot handle inconsistent hierarchical depths; and it performs hierarchical splitting of the tree model path data based on storage group identifier data to obtain relational model structure data, achieving a standardized mapping from tree model to relational model. This laid the foundation for subsequent unified table creation and querying; it generated dual-model table creation statements based on relational model structured data and performed validation, ensuring the standardization and consistency of table creation; it performed syntax parsing and semantic validation on structured query language statements to ensure the legality of query statements; it detected time conditions based on syntax trees and automatically rewrote them, avoiding system performance issues caused by full table scans without time ranges and improving query response speed; it listened for permission change events and automatically synchronized them, realizing unified permission management and reducing operation and maintenance costs; and it generated unified management status data based on optimized query statement data and synchronization result data, realizing global control over the integration status.

[0008] In one optional implementation, obtaining the time-series database instance configuration information and establishing a data connection channel based on the instance configuration information includes: obtaining the address information, port information, and authentication configuration information of the time-series database; determining the authentication method based on the authentication configuration information, including username / password authentication and Kerberos authentication; establishing a data connection channel between the data platform and the time-series database based on the address information, port information, and authentication configuration information; performing connectivity verification on the data connection channel, and storing the data connection channel in the connection pool after the verification passes.

[0009] In one optional implementation, the step of matching the tree model path data and the storage group list data using the longest prefix matching algorithm to obtain storage group identifier data includes: calling the time-series database metadata interface to obtain the storage group list data, which contains the complete path names of each storage group; traversing each tree model path data and performing prefix matching on each storage group path name in the storage group list data; determining the storage group path name with the longest matching length as the storage group to which the current tree model path data belongs, and generating storage group identifier data.

[0010] In one optional implementation, the step of hierarchically splitting the tree model path data based on storage group identifier data to obtain relational model structure data includes: removing the prefix portion corresponding to the storage group identifier data from the tree model path data to obtain the remaining path data; splitting the remaining path data according to the hierarchical separator to obtain a hierarchical node data sequence; performing boundary case processing on the hierarchical node data sequence, the boundary case processing being performed based on preset processing rules; and classifying the processed hierarchical node data sequence according to the database layer, table layer, and field layer to obtain relational model structure data.

[0011] In one optional implementation, the step of generating tree model table creation statement data and table model table creation statement data based on relational model structure data, and validating the tree model table creation statement data and table model table creation statement data based on preset table creation constraints to obtain valid table structure definition data includes: obtaining table creation model type identifier data; when the table creation model type identifier data is a tree model, generating tree model table creation statement data based on relational model structure data, the tree model table creation statement data including storage group definition statements and measurement point definition statements; when the table creation model type identifier data is a table model, generating table model table creation statement data based on relational model structure data, the table model table creation statement data including table definition statements and column definition statements; performing validation on the table creation statement data based on preset table creation constraints, and outputting table structure definition data after successful validation.

[0012] In one optional implementation, the steps of obtaining structured query language query statement data, parsing the structured query language query statement data to obtain syntax tree data, and performing semantic verification on the syntax tree data based on table structure definition data to obtain verified syntax tree data include: performing lexical analysis on the structured query language query statement data based on a lexical analyzer to obtain a sequence of lexical units; performing syntactic analysis on the sequence of lexical units based on a syntax analyzer to generate syntax tree data; performing semantic verification on the syntax tree data based on table structure definition data, including table name verification, column name verification, and data type verification; and outputting verified syntax tree data after successful verification.

[0013] In one optional implementation, the step of detecting time conditions based on validated syntax tree data and rewriting the structured query language query statement data based on the detection results to obtain optimized structured query language statement data includes: traversing the validated syntax tree data and detecting whether the conditional clause contains a time column condition; if the conditional clause does not contain a time column condition, appending a preset time range condition to the conditional clause to generate the rewritten structured query language statement data; if the conditional clause contains a time column condition, using the structured query language query statement data as the rewritten structured query language statement data; and replacing the configuration parameters in the rewritten structured query language statement data with actual values ​​to obtain the optimized structured query language statement data.

[0014] In one optional implementation, the process of monitoring permission change event data on the data platform side, generating permission synchronization instruction data based on the permission change event data, and sending the permission synchronization instruction data to a time-series database via a data connection channel to obtain synchronization result data includes: registering a permission change listener on the data platform side; capturing permission change event data, including authorization event data and revocation event data, based on the permission change event data; parsing permission subject information, permission object information, and permission operation information from the permission change event data; generating permission synchronization instruction data based on the permission subject information, permission object information, and permission operation information; and sending the permission synchronization instruction data to a time-series database for execution via a data connection channel, receiving the execution return result, and generating synchronization result data.

[0015] In one optional implementation, the step of generating unified management status data based on optimized structured query language statement data and synchronization result data includes: extracting query configuration information from the optimized structured query language statement data; extracting permission synchronization status information from the synchronization result data; combining the query configuration information and permission synchronization status information into unified management status data; and writing the unified management status data into a management status table.

[0016] Secondly, the present invention provides a unified management system integrating a time-series database and a data platform. The system includes: a connection module for acquiring time-series database instance configuration information and establishing a data connection channel based on the instance configuration information, the data connection channel connecting the data platform and the time-series database; a collection module for acquiring storage group list data and tree model path data from the time-series database based on the data connection channel; a matching module for matching the tree model path data and storage group list data using a longest prefix matching algorithm to obtain storage group identifier data; a splitting module for performing hierarchical splitting of the tree model path data based on the storage group identifier data to obtain relational model structure data; and a table creation module for generating tree model table creation statement data and table model table creation statement data based on the relational model structure data, and configuring the tree model table creation statement data and table model table creation statement data based on preset table creation constraints. The system is divided into several modules: a query tree module and a management module. The query tree is validated to obtain a valid table structure definition. The parsing module acquires the structured query language query statement data, performs syntactic parsing to obtain a syntax tree, and performs semantic validation on the syntax tree based on the table structure definition data to obtain a valid syntax tree. The optimization module checks time conditions based on the valid syntax tree data and rewrites the structured query language query statement data based on the detection results to obtain optimized structured query language statement data. The synchronization module monitors permission change events on the data platform side, generates permission synchronization instructions based on these events, and sends these instructions to the time-series database via the data connection channel to obtain synchronization result data. The management module generates unified management status data based on the optimized structured query language statement data and the synchronization result data. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a specific example of a unified management method integrating a time-series database and a data platform based on embodiments of the present invention. Figure 2 This is a schematic diagram of a specific example of a unified management system integrating a time-series database and a data platform based on embodiments of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0020] To address the technical problems mentioned in the background section, this application provides a unified management method integrating time-series databases and data platforms, as detailed in the embodiments below. Figure 1 As shown, the method includes: Step S101: Obtain the time series database instance configuration information, and establish a data connection channel based on the instance configuration information. The data connection channel connects the data platform and the time series database.

[0021] Specifically, step S101 includes: Step a1: Obtain the address information, port information, and authentication configuration information of the time series database.

[0022] Furthermore, the configuration information of the time-series database instance to be managed is read from the resource management module of the data platform. The instance configuration information includes three core fields: address information, port information, and authentication configuration information. The address information is the network address of the time-series database instance, used to identify the database instance's location on the network, supporting both IP address and domain name formats. The port information is the network port number that the time-series database service listens on, used to identify the network access point of the database service. The authentication configuration information includes the authentication method identifier and corresponding authentication parameters, used for authentication when establishing a connection. The authentication configuration information can be obtained in two ways: manual entry and batch import. Manual entry is suitable for scenarios requiring rapid single-instance access, while batch import is suitable for scenarios requiring centralized management of multiple instances.

[0023] Step a2: Determine the authentication method based on the authentication configuration information. The authentication methods include username / password authentication and Kerberos authentication.

[0024] Further, the authentication method identifier field in the authentication configuration information is read, and the authentication method to be used is determined based on the value of the identifier field. When the authentication method identifier is the first preset value, the username-password authentication method is selected, and the authentication parameters include two fields: username and password. When the authentication method identifier is the second preset value, the Kerberos authentication method is selected, and the authentication parameters include fields such as principal name, key table file path, and domain name. The two authentication methods are suitable for different security level scenarios: the username-password authentication method is simple to configure and suitable for development and testing environments and internal secure network environments; the Kerberos authentication method is based on a ticket mechanism, has higher security, and is suitable for production environments and cross-network domain integration scenarios.

[0025] Step a3: Establish a data connection channel between the data platform and the time-series database based on address information, port information, and authentication configuration information.

[0026] Furthermore, a network connection endpoint is constructed based on the address and port information, and the corresponding authentication driver is loaded based on the authentication configuration information. The connection establishment interface of the database driver is called, passing in the connection endpoint parameters and authentication parameters, to initiate a connection establishment request. The connection establishment process includes two phases: network layer connection establishment and authentication layer authentication. The network layer connection establishment phase completes the TCP three-way handshake to ensure network reachability; the authentication layer authentication phase verifies the user's identity to ensure the access party has legitimate access rights. Once both phases are successfully completed, the data connection channel is established. The data connection channel supports various types of data interaction, including data querying, metadata retrieval, and permission operations.

[0027] Step a4: Perform a connectivity check on the data connection channel. If the check passes, store the data connection channel in the connection pool.

[0028] Furthermore, once the data connection channel is established, connectivity verification is performed immediately. Connectivity verification is achieved by sending a pre-defined probe statement to the time-series database. This probe statement is a lightweight query statement used only to verify connection availability and does not involve any actual data reading. If the probe statement returns a successful response within the pre-defined timeout period, the connectivity verification is considered successful; if the probe statement times out or returns an error response, the connectivity verification is considered unsuccessful, the reason for the failure is recorded, and an alarm is triggered. Verified data connection channels are stored in a connection pool for unified management. The connection pool maintains a set of available connections, supporting connection reuse, expiration detection, and automatic reconstruction, avoiding the performance overhead of frequently creating and destroying connections. Parameters such as the maximum number of connections, the minimum number of idle connections, and the connection timeout period in the connection pool can be configured.

[0029] Step S102: Obtain the storage group list data and tree model path data from the time series database based on the data connection channel.

[0030] Specifically, available data connection channels are obtained from the connection pool, and the metadata query interface of the time-series database is called based on these channels. First, a storage group list query is executed to obtain the complete path names of all storage groups in the database, forming a storage group list. This list is stored as an array, with each element representing the complete path string for a storage group. Then, a tree model path query is executed, traversing the test point paths under all storage groups to obtain all tree model path data. This data is also stored as an array, with each element representing a complete path string from the root node to a leaf node, with nodes at different levels separated by periods. The obtained storage group list and tree model path data are temporarily stored in a metadata cache for use in subsequent matching and splitting steps. The metadata cache supports both timed refresh and event-triggered refresh mechanisms to ensure that the cached data remains consistent with the actual database state.

[0031] Step S103: The longest prefix matching algorithm is used to match the tree model path data and the storage group list data to obtain the storage group identifier data.

[0032] Specifically, step S103 includes: Step c1: Call the time series database metadata interface to obtain the storage group list data, which includes the complete path name of each storage group.

[0033] Furthermore, the `SHOW STORAGE GROUPS` statement or an equivalent metadata query interface of the time-series database is invoked to retrieve information on all storage groups in the current database instance. The query results include attributes such as the full path name, TTL configuration, and number of replicas for each storage group. The full path name field of each storage group is extracted to construct a storage group list. The storage group list is a one-dimensional string array, with the array length equal to the number of storage groups. To improve subsequent matching efficiency, the storage group list is sorted in descending order of path length, with longer storage group paths appearing first. The sorting operation is performed once when the storage group list is first retrieved, and re-sorted whenever the storage group list changes.

[0034] Step c2: Traverse each tree model path data and perform prefix matching on each storage group path name in the current tree model path data and the storage group list data.

[0035] Furthermore, each tree model path data is sequentially extracted as the current path to be matched. For the current path to be matched, prefix matching is performed on each storage group path name in the storage group list data in sorted order. The logic of prefix matching is as follows: the storage group path name is used as a prefix string, and it is checked whether the current path to be matched starts with this prefix string, and the next character after the prefix string is the path separator. The separator check is added to avoid false matches. For example, when the storage group path is "root.abc", the path "root.abcd" should not be matched as a path under this storage group. During the traversal, all successfully matched storage group path names and their matching lengths are recorded.

[0036] Step c3: Determine the storage group path name with the longest matching length as the storage group to which the current tree model path data belongs, and generate storage group identifier data.

[0037] Furthermore, among all successfully matched storage groups, the storage group with the longest match length is selected as the storage group to which the current tree model path belongs. The match length is the number of characters in the storage group path name; the more characters, the more accurate the match. The longest prefix matching strategy can correctly handle scenarios with nested storage groups. For example, when two storage groups, "root.energy" and "root.energy.wind", exist simultaneously, the path "root.energy.wind.turbine01.speed" will be correctly matched to the "root.energy.wind" storage group, not the "root.energy" storage group. After determining the belonging storage group, the corresponding storage group identifier data is generated. The storage group identifier data includes fields such as storage group name, storage group path, and match length, used to identify the storage group to which the current path belongs. The above matching process is repeated for all tree model path data, ultimately obtaining the storage group identifier data set corresponding to each path.

[0038] Step S104: Based on the storage group identifier data, the tree model path data is hierarchically split to obtain relational model structure data.

[0039] Specifically, step S104 includes: Step d1: Remove the prefix portion corresponding to the storage group identifier data from the tree model path data to obtain the remaining path data.

[0040] Furthermore, based on the storage group path in the storage group identifier data, the storage group path prefix is ​​removed from the original tree model path. This removal operation includes removing the storage group path string and any subsequent path separators, ensuring that the remaining part begins with a valid node name. The remaining path data represents the hierarchical paths below the storage group, including device dimension nodes and measurement point nodes. For example, if the original path is "root.energy.wind.turbine01.speed" and the storage group path is "root.energy.wind", then the remaining path is "turbine01.speed". This remaining path data serves as input for subsequent hierarchical splitting.

[0041] Step d2 involves splitting the remaining path data according to the hierarchical delimiter to obtain a sequence of hierarchical node data.

[0042] Furthermore, using periods as level separators, the remaining path data is split into multiple node names, forming a hierarchical node data sequence. This sequence is a one-dimensional string array, where the order of elements corresponds to the left-to-right level order in the path, and the array's length is the level depth of the remaining path. For example, the remaining path "turbine01.speed" is split into ["turbine01", "speed"], with a level depth of 2. After splitting, basic validity checks are performed on each node name, including non-empty checks, special character checks, and length checks, to ensure that the node names conform to naming conventions.

[0043] Step d3 involves performing boundary condition processing on the hierarchical node data sequence, which is based on preset processing rules.

[0044] Further, the hierarchical depth and node characteristics of the hierarchical node data sequence are examined to determine if boundary conditions exist. If boundary conditions exist, corresponding processing operations are performed based on preset processing rules. The preset processing rules include the following five: Rule 1: When the path depth is less than the preset split depth, the already split hierarchical data is filled into the corresponding model level, and unidentified levels are filled with placeholder data. When the remaining path's hierarchical depth is less than the preset split depth, it indicates that the path lacks some dimension levels. The processing method is: existing nodes are filled into the corresponding model levels in sequence, and insufficient levels are filled with preset placeholders. The placeholder data is a uniform default value used to indicate that the level lacks corresponding dimension information. Rule 2: When the storage group nesting level exceeds one level, the complete storage group name is used as the database name data. When the storage group path contains multiple levels of nesting, such as "root.energy.wind" containing two levels of nesting, the innermost storage group name is not used as the database name; instead, the complete storage group path is used as the database name. This avoids naming conflicts between child storage groups with the same name under different parent storage groups. Rule 3: When the path level exceeds the preset split depth, merge the node names exceeding the level into field name data. When the level depth of the remaining path exceeds the preset split depth, it indicates that the path contains additional sub-levels. The processing method is: the first N-1 nodes are split according to the normal level, and all nodes from the Nth to the next are reconnected with a dot and merged into a single field name. Here, N is the preset split depth. This ensures that the table structure of the relational model has a fixed number of levels while not losing path information. Rule 4: When nodes of the same path in the same storage group correspond to different data types, select a unified data type according to the preset priority. When the same field name in the same storage group and the same table level path corresponds to different data types, a type conflict exists. The processing method is: select the highest priority data type as the unified data type according to the preset data type priority list. The data type priorities from high to low are string type, floating-point type, integer type, and boolean type. Selecting a high-priority type can avoid data truncation or precision loss. Rule 5: When a path node name contains special characters, the special characters are escaped to obtain the escaped node name data. Directly splitting path node names containing special characters such as periods, spaces, or Chinese punctuation may lead to hierarchical recognition errors. The solution is to escape the special characters using a pre-defined escape sequence to ensure that the split node names can be correctly recognized and used subsequently.

[0045] Step d4: The processed hierarchical node data sequence is categorized into database layer, table layer, and field layer to obtain relational model structure data.

[0046] Furthermore, after boundary condition processing, the hierarchical node data sequence possesses a regular hierarchical structure. According to preset mapping rules, each hierarchical node is categorized into different levels of the relational model. The database layer corresponds to storage group identifier data, used to identify the database to which the data belongs; the table layer corresponds to device dimension nodes below the storage group, used to identify the table and dimension columns to which the data belongs; the field layer corresponds to the leaf nodes of the path, used to identify the specific measurement value column. The database layer information, table layer information, field layer information, and corresponding data type information are combined to form complete relational model structure data. The relational model structure data is stored in the form of structured objects, containing fields such as database name, table name, column name list, and column data type list, providing a unified model foundation for subsequent table creation and queries.

[0047] Step S105: Generate tree model table creation statement data and table model table creation statement data based on the relational model structure data. Validate the tree model table creation statement data and table model table creation statement data based on preset table creation constraints to obtain the validated table structure definition data.

[0048] Specifically, step S105 includes: Step e1: Obtain the table creation model type identifier data.

[0049] Furthermore, the model type field is read from the table creation request parameters to obtain the table creation model type identifier data. This identifier data indicates the model type used in the current table creation operation, and can be either a tree model type or a table model type. The tree model type is suitable for time-series data modeling scenarios centered on hierarchical paths, while the table model type is suitable for relational modeling scenarios centered on two-dimensional table structures. Both model types share the same set of relational model structure data, ensuring metadata consistency between the two models.

[0050] Step e2: When the table creation model type identifier data is a tree model, the tree model table creation statement data is generated based on the relational model structure data. The tree model table creation statement data includes storage group definition statements and measurement point definition statements.

[0051] Furthermore, when the table creation model type is a tree model, the corresponding tree model table creation statement is constructed based on the relational model structure data. The tree model table creation statement consists of two parts: the first part is the storage group definition statement, which generates a statement to create a storage group based on database layer information, specifying the storage group's path name and related attributes; the second part is the measurement point definition statement, which generates a statement to create a measurement point based on table and field layer information, specifying the measurement point's path and data type. During the generation process, the dimension columns of the relational model are mapped to intermediate-level nodes of the tree model, and the measurement value columns are mapped to leaf nodes of the tree model. The generated tree model table creation statement conforms to the tree model syntax specification of time-series databases.

[0052] Step e3: When the table model type identifier data is a table model, the table model creation statement data is generated based on the relational model structure data. The table model creation statement data includes table definition statements and column definition statements.

[0053] Furthermore, when the table model type is a table model, the corresponding table creation statement is constructed based on the relational model structure data. The table model creation statement uses the standard CREATE TABLE syntax structure, including a table definition section and a column definition section. The table definition section is generated based on database layer information and table layer information, specifying the database name and table name to which the table belongs. The column definition section is generated based on field layer information, with each column corresponding to one field, including the column name and data type. Column definitions are categorized and labeled according to column type, including four types: time columns, label columns, measurement value columns, and attribute columns. The generated table model creation statement conforms to the table model syntax specifications of time-series databases while also being compatible with the syntax conventions of standard relational databases.

[0054] Step e4: Perform validation on the table creation statement data based on preset table creation constraints. After successful validation, output the table structure definition data.

[0055] Furthermore, after the table creation statement is generated, compliance checks are performed on the statement based on preset table creation constraints. These preset constraints include the following three core rules: Constraint 1: Each table contains one and only one time column. The time column is the core dimension of time-series data, used to identify the timestamps of data points. The validation logic is: count the number of columns in the table creation statement whose category is time; if the number is not equal to one, the validation fails. Constraint 2: Each table defines at least one label column and one measurement value column. The label column is used to identify the dimensional information of the data, and the measurement value column is used to store the actual monitoring data; both are indispensable to form a valid time-series data model. The validation logic is: count the number of label columns and measurement value columns in the table creation statement; if either number is zero, the validation fails. Constraint 3: The column name of the time column is a fixed value. The column name of the time column must be a preset fixed name and cannot be customized, ensuring consistency of time columns between different tables and facilitating unified querying and processing. The validation logic is: check whether the column name of the time column is equal to the preset fixed value; if not, the validation fails. During the validation process, if all constraints are met, the validation is considered successful, and the table structure definition data is output. If any constraint is not met, the validation is considered unsuccessful, and the specific reason for the failure is returned for the user to correct and resubmit. The table structure definition data is a standardized table structure description after successful validation, containing complete information such as table name, column names, column data types, and column categories, serving as the basis for subsequent query optimization and metadata management.

[0056] Step S106: Obtain the structured query language query statement data, perform syntax parsing on the structured query language query statement data to obtain syntax tree data, and perform semantic verification on the syntax tree data based on the table structure definition data to obtain the verified syntax tree data.

[0057] Specifically, step S106 includes: Step f1: Perform lexical analysis on the structured query language query statement data based on the lexical analyzer to obtain a sequence of lexical units.

[0058] Furthermore, the user-input structured query language query statement is taken as the input string and fed into the lexical analyzer for processing. Based on the principle of finite state automata, the lexical analyzer decomposes the input character stream into a meaningful sequence of lexical units. Lexical units include keywords, identifiers, literals, operators, delimiters, etc. Each lexical unit contains a type identifier and value information. During lexical analysis, meaningless characters such as spaces and newlines are skipped, and comments are filtered out. After lexical analysis is completed, a structured sequence of lexical units is output as input for subsequent syntax analysis.

[0059] Step f2: Perform syntax analysis on the lexical unit sequence based on the parser to generate syntax tree data.

[0060] Furthermore, the sequence of lexical units is fed into the parser for processing. Based on context-free grammars and pushdown automata principles, the parser organizes the sequence of lexical units into a hierarchical syntax tree structure according to preset grammar rules. The syntax tree represents the grammatical structure of the query statement in a tree structure. Each node in the tree represents a grammatical construct; leaf nodes correspond to lexical units, and non-leaf nodes correspond to grammar rules. The root node of the syntax tree is the query statement node, and child nodes include SELECT clauses, FROM clauses, WHERE clauses, ORDER BY clauses, etc. During syntax analysis, if a syntax error is found, a syntax exception is thrown, indicating the error location and type. After successful syntax analysis, the complete syntax tree data is output.

[0061] Step f3 involves performing semantic validation on the syntax tree data based on the table structure definition data. The validation includes table name validation, column name validation, and data type validation.

[0062] Furthermore, after syntax validation passes, semantic-level validation is performed on the syntax tree based on the table structure definition data. Semantic validation starts from the root node of the syntax tree and recursively traverses each child node, executing corresponding validation logic for different types of nodes. Table name validation: Checks if the table name referenced in the FROM clause exists in the table structure definition data. If the referenced table name is not registered in the table structure definition, it is determined that the table does not exist. Column name validation: Checks if the column names referenced in the SELECT, WHERE, ORDER BY, etc., exist in the column definitions of the corresponding tables. If the referenced column names are not registered in the table structure definition, it is determined that the column does not exist. Data type validation: Checks whether the data types of operands in expression operations and function calls match. For example, arithmetic operations require operands to be numeric, string functions require operands to be string, and comparison operations require data type compatibility on both sides. If the data types do not match, it is determined that there is a type incompatibility error. During semantic validation, if any semantic error is found, the validation immediately terminates and an error message is returned; if all validation items pass, the semantic validation is considered successful.

[0063] Step f4: After successful validation, output the validated syntax tree data.

[0064] Furthermore, after both syntax analysis and semantic validation pass, the validated syntax tree data is output. This validated syntax tree data is a valid syntax tree, ensuring that the query statement conforms to specifications at both the syntactic and semantic levels. The validated syntax tree data serves as input for subsequent query optimization steps, used for time condition checks and statement rewriting. The syntax tree data uses a standardized internal representation format, facilitating subsequent traversal and modification operations.

[0065] Step S107: Based on the valid syntax tree data, the time condition is checked, and the structured query language query statement data is rewritten based on the detection result to obtain the optimized structured query language statement data.

[0066] Specifically, step S107 includes: Step g1: Traverse the validated syntax tree data and check whether the conditional clauses contain time column conditions.

[0067] Further, the WHERE clause node is located from the validated syntax tree data. If no WHERE clause exists, it is directly determined that no time condition is included. If a WHERE clause exists, all condition nodes under the WHERE subtree are recursively traversed. For each condition node, it is checked whether the column name involved in the condition is a time column. The time column is identified based on the column category label in the table structure definition data; columns with the category of time are time columns. During the traversal, as long as any condition node involving a time column is found, it is determined that a time condition is included; if no time column condition is found after traversing all condition nodes, it is determined that no time condition is included. The detection results are recorded in Boolean form for use in subsequent rewrite logic.

[0068] Step g2: If the condition clause does not contain a time column condition, add a preset time range condition to the condition clause to generate the rewritten structured query language statement data.

[0069] Furthermore, when the detection result does not include a time condition, the query statement needs to be automatically rewritten to add a time range restriction. The rewriting operation is performed at the syntax tree level: First, a new time range condition node is constructed, with the condition that the time column is greater than or equal to the start time and less than or equal to the end time. The start and end times are determined based on preset time range parameters, which support granular configuration by day, hour, minute, etc. Then, the newly constructed time range condition node is combined with the original WHERE clause condition node using a logical AND operator to form a new WHERE clause. If the original query statement does not have a WHERE clause, a new WHERE clause node is added. After the syntax tree rewriting is complete, the syntax tree is reserialized into a query statement string, resulting in the rewritten Structured Query Language (SCL) statement data.

[0070] Step g3: When the conditional clause contains a time column condition, the structured query language query data is used as the rewritten structured query language query data.

[0071] Furthermore, when the detection result includes a time condition, it indicates that the user has already specified a time range in the query statement, and no additional time condition is needed. In this case, the original Structured Query Language (SCL) query statement data is directly used as the rewritten statement data without any modification. This ensures that the user-specified time condition remains unaffected and avoids logical redundancy caused by duplicate conditions. The rewritten statement data is identical in content to the original statement data, but has undergone syntax and semantic validation to ensure its legality.

[0072] Step g4: Replace the configuration parameters in the rewritten Structured Query Language (SCL) statement data with the actual values ​​to obtain the optimized SCL statement data.

[0073] Furthermore, the rewritten statement may contain configuration parameters, which exist as placeholders and need to be replaced with actual values ​​before execution. Configuration parameters include two categories: time range parameters and maximum row count limits. Time range parameters control the time span of the query, while maximum row count limits control the maximum number of rows returned by the query, preventing memory overflow due to an excessively large result set. The replacement process involves finding the placeholder identifiers for the configuration parameters in the rewritten statement and replacing each placeholder with its corresponding actual configuration value. The values ​​of the configuration parameters can be user-defined, allowing users to adjust the time range and maximum row count limits according to business needs. After the parameter replacement is complete, the final optimized Structured Query Language (SCL) statement data is obtained and can be directly submitted to the time-series database for execution.

[0074] Step S108: Monitor permission change event data on the data platform side, generate permission synchronization instruction data based on the permission change event data, and send the permission synchronization instruction data to the time series database based on the data connection channel to obtain synchronization result data.

[0075] Specifically, step S108 includes: Step h1: Register a permission change listener on the data platform side.

[0076] Furthermore, an event listener is registered in the data platform's permission management module to monitor permission-related changes. Listener registration is achieved by calling the event subscription interface provided by the permission management module, and the subscribed event types include authorization and revocation events. After registration, when a permission change occurs on the data platform side, the permission management module proactively pushes an event notification to the listener. Using an event-driven listening approach, rather than a polling approach, allows for real-time detection of permission changes, reduces unnecessary query overhead, and ensures timely permission synchronization.

[0077] Step h2: Capture permission change event data based on the permission change listener. The permission change event data includes authorization event data and revocation event data.

[0078] Furthermore, upon receiving an event notification, the permission change listener extracts detailed information from the notification and constructs permission change event data. This data includes fields such as event type, event occurrence time, operator, permission subject, permission object, and permission operation. The event type field distinguishes between authorization and revocation events: authorization events indicate the addition of new permissions, while revocation events indicate the revocation of previously assigned permissions. The event occurrence time records the execution time of the permission change operation, used for synchronization sequence control and audit traceability. The operator field records the user identifier who performed the permission change, used for accountability tracking.

[0079] Step h3: Parse the permission subject information, permission object information, and permission operation information based on the permission change event data.

[0080] Furthermore, the permission change event data is parsed to extract three core types of information required for permission synchronization. The permission subject information identifies the object to which the permission is granted, including user identifier, role identifier, or user group identifier. The permission object information identifies the resource scope to which the permission applies, including database name, table name, or field name. The permission operation information identifies the specific permission type, including six types: all permissions, write permissions, query permissions, create permissions, modify permissions, and delete permissions. During the parsing process, the permission model on the data platform side is mapped to the permission model on the time-series database side, ensuring a one-to-one correspondence between the permission concepts on both sides and avoiding synchronization deviations caused by differences in permission models.

[0081] Step h4: Generate permission synchronization instruction data based on permission subject information, permission object information, and permission operation information.

[0082] Furthermore, based on the event type and the parsed permission information, corresponding permission synchronization instructions are constructed. When the event type is an authorization event, an authorization instruction is generated, granting the specified permissions on the specified object to the specified subject. When the event type is a revocation event, a revocation instruction is generated, revoking the specified permissions on the specified object from the specified subject. The permission synchronization instructions adopt the permission management syntax format of the time-series database to ensure that the instructions can be correctly recognized and executed by the time-series database. During the generation process, the legality of the permission subject, permission object, and permission operation is verified to ensure that the instruction content is complete and conforms to the syntax specification. After the verification is successful, the permission synchronization instruction data is output.

[0083] Step h5: Send the permission synchronization instruction data to the time-series database for execution based on the data connection channel, and receive the execution return results to generate synchronization result data.

[0084] Further, a data connection channel for the corresponding time-series database instance is obtained from the connection pool, and the permission synchronization command data is sent to the time-series database for execution through the data connection channel. After the command is sent, the system waits for the time-series database to return the execution result. The execution result includes two parts: execution status and execution information. The execution status indicates whether the command was executed successfully or failed; the execution information contains specific execution details, including the reason for failure if it fails. Upon receiving the execution result, the execution status, execution information, event type, and synchronization time are combined into synchronization result data. The synchronization result data is used to record the execution status of each permission synchronization, facilitating subsequent querying, auditing, and troubleshooting. When synchronization fails, a retry mechanism is automatically triggered, performing a limited number of retries according to a preset retry strategy. If the retry still fails, a failure log is recorded and an alarm notification is triggered.

[0085] Step S109: Generate unified management status data based on the optimized structured query language statement data and synchronization result data.

[0086] Specifically, step S109 includes: Step i1: Extract query configuration information from the optimized Structured Query Language statement data.

[0087] Furthermore, query configuration-related information is extracted from the optimized Structured Query Language (SCL) statement data. This information includes the table names involved in the query, the time range of the query, the maximum number of rows allowed, and the list of columns to be queried. This information reflects the current query configuration status and is an important component of the unified management status. The extraction process is achieved by parsing the optimized statement or directly obtaining relevant node information from the syntax tree.

[0088] Step i2: Extract permission synchronization status information from the synchronization result data.

[0089] Furthermore, status information related to permission synchronization is extracted from the synchronization result data. This status information includes the time of the most recent synchronization, the result of the most recent synchronization, the cumulative number of successful synchronizations, and the cumulative number of failed synchronizations. This information reflects the operational status of permission synchronization and is another important component of the unified management status. The extraction process is achieved by reading the corresponding fields from the synchronization result data.

[0090] Step i3 combines the query configuration information and permission synchronization status information into unified management status data.

[0091] Furthermore, the extracted query configuration information and permission synchronization status information are integrated to form unified management status data. This unified management status data is stored in the form of structured objects, containing two sub-objects: query management sub-status and permission management sub-status, corresponding to the status information of query optimization and permission synchronization, respectively. In addition to query configuration and permission synchronization status, the unified management status data also includes other dimensions of management information such as connection status and metadata synchronization status, forming a comprehensive characterization of the time-series database integration management status.

[0092] Step i4: Write the unified management status data into the management status table.

[0093] Furthermore, the generated unified management status data is persisted to a management status table. The management status table is stored in the data platform's metadata database, with each record corresponding to the management status of a time-series database instance. The table structure includes fields such as instance identifier, connection status, query configuration status, permission synchronization status, metadata synchronization status, and last update time. Status data is written using an update-insert pattern; if the instance already exists, the existing record is updated; if the instance does not exist, a new record is inserted. The management status table provides a unified status view for operations and maintenance management, allowing administrators to quickly understand the integrated operational status of each time-series database instance by querying the management status table.

[0094] In an optional implementation, the method further includes a heterogeneous data migration step: obtaining source data source information and target data source information, determining a migration mode based on the source model type and target model type, constructing data read statements and data write statements based on the migration mode, performing data synchronization, and generating migration result data; the migration mode includes three types: tree model to relational model migration, relational model to tree model migration, and tree model to tree model migration.

[0095] Specifically, including: Step j1: Obtain source data source information and target data source information.

[0096] Furthermore, the source and target data source identifiers are read from the migration task configuration, and the corresponding connection and model type information is obtained from the data source management module based on the data source identifier. Both the source and target data sources support multiple types, including time-series databases, relational databases, and data warehouses. The model type information identifies the data model type used by the data source, including tree model types and table model types.

[0097] Step j2: Determine the migration mode based on the source model type and the target model type.

[0098] Furthermore, based on the combination of source and target model types, the corresponding migration mode is determined. There are three migration modes: the first is a migration from a tree model to a relational model, where the source is a tree-model time-series database and the target is a relational database or data warehouse; the second is a migration from a relational model to a tree model, where the source is a relational database or data warehouse and the target is a tree-model time-series database; and the third is a migration from one tree model to another, where both the source and target are tree-model time-series databases. Different migration modes correspond to different data transformation logic and field mapping rules.

[0099] Step j3: Construct data read statements and data write statements based on the migration mode.

[0100] Furthermore, based on the determined migration pattern, source-side data read statements and target-side data write statements are constructed. When the tree model is used as the source, the data read statements are constructed based on the storage path and data retrieval conditions, supporting data filtering by storage group, device, time range, and other dimensions. When the table model is used as the source, the data read statements are constructed based on the table name and data retrieval conditions, supporting data filtering by column and by condition. When the tree model is used as the target, the data write statements are constructed based on the measurement point path; when the relational model is used as the target, the data write statements are constructed based on the table name and column name. During the construction process, model conversion is automatically performed to convert the source-side data format to a target-side compatible data format.

[0101] Step j4: Perform data synchronization and generate migration result data.

[0102] Furthermore, based on the constructed read and write statements, batch data synchronization is performed. Data synchronization employs paginated reads and batch writes to control the amount of data processed in a single operation and avoid memory overflow. During synchronization, the amount of synchronized data, the amount of failed data, and the reasons for failure are recorded. Two execution modes are supported: interface machine mode and non-interface machine mode. In interface machine mode, data is relayed through an independent interface machine node, suitable for network isolation scenarios; in non-interface machine mode, data is transmitted directly, suitable for network interconnection scenarios. After synchronization is complete, synchronization statistics, details of failed data, and execution time are combined to form the migration result data.

[0103] In an optional implementation, the method further includes a metadata collection and data sharing step: collecting metadata information from the time-series database based on the data connection channel, converting the metadata information into unified asset model data based on preset mapping rules, incorporating the unified asset model data into the data asset catalog and performing structured display; and uploading the time-series data table to the sharing and exchange portal through an application programming interface or a wide table based on the data asset catalog.

[0104] Specifically, including: Step k1: Collect metadata information from the time-series database based on the data connection channel.

[0105] Furthermore, a data connection channel is obtained from the connection pool, and a metadata query statement is executed via the JDBC driver to retrieve the raw metadata of the time-series database. The raw metadata includes a database list, table list, field list, data types, and comment information. The collection process supports filtering by storage model: when selecting a tree model, the storage group range is specified for collection; when selecting a table model, the database name is specified for collection. The collected raw metadata is temporarily stored in a metadata cache.

[0106] Step k2: Convert metadata information into unified asset model data based on preset mapping rules.

[0107] Furthermore, according to preset mapping rules, the raw metadata of the time-series database is converted into a unified asset model format. The mapping rules for tree model metadata are as follows: storage groups are mapped to the database layer, device dimension nodes are mapped to the table layer, and measurement point nodes are mapped to the field layer; when the path depth is insufficient, table names are set to empty or placeholders are used, and fields are directly taken from the measurement point nodes; when storage groups are nested more than one level, the complete storage group name is used as the database name. The mapping rules for table model metadata are as follows: databases are mapped to the database layer, tables are mapped to the table layer, and columns are mapped to the field layer. The converted unified asset model data includes standardized fields such as asset name, asset type, asset level, field list, data type, and source system.

[0108] Step k3 involves incorporating the unified asset model data into the data asset catalog and performing a structured display.

[0109] Furthermore, the transformed unified asset model data is written into the asset catalog table, completing the asset registration of metadata. The asset catalog supports multiple query methods, including browsing by hierarchy, searching by keyword, and filtering by type. The asset details page displays basic asset information, field information, data examples, lineage information, and permission information, forming a searchable, browsable, and understandable data asset catalog.

[0110] Step k4: Based on the data asset catalog, upload the time-series data tables to the shared exchange portal via application programming interfaces or wide tables.

[0111] Next, select the time-series data tables to be shared from the data asset catalog, and configure the sharing method and scope. There are two sharing methods: Application Programming Interface (API) and Wide Table. The API method encapsulates the data tables as RESTful interfaces, allowing external systems to retrieve data via API calls. The Wide Table method publishes the data tables as directly accessible wide table views, allowing external systems to read them directly using standard query languages. After configuration, the data tables are uploaded to the sharing and exchange portal for other users to browse and apply for. Users who submit applications and receive approval will obtain the corresponding data access permissions.

[0112] The unified management method for integrating time-series databases and data platforms provided in this embodiment has the following beneficial effects: By establishing standardized data connection channels, unified access and management of time-series databases are achieved, solving the problem of inconsistent access methods for heterogeneous data sources; By automatically identifying storage groups through the longest prefix matching algorithm, accurate attribution determination of tree model paths to storage groups is achieved, correctly handling complex scenarios such as nested storage groups and avoiding the inherent defects of fixed-position path splitting; Through hierarchical splitting and boundary case handling, standardized mapping from tree models to relational models is achieved, automatically handling boundary cases such as insufficient path depth, exceeding hierarchical limits, and type conflicts, significantly reducing manual intervention; Through dual-mode... Unified table creation and constraint validation ensure the standardization and consistency of table creation, reducing the probability of modeling errors. Syntax parsing and semantic validation ensure the legality of query statements, providing a reliable foundation for subsequent optimization. Automatic detection of time conditions and the addition of time range limits effectively avoid full table scans without time ranges, significantly improving query response speed and ensuring system stability. An event-driven permission synchronization mechanism enables automatic synchronization of permissions between the data platform and the time-series database, greatly reducing the operational costs of permission management and ensuring data security. Unified management of state data generation and persistence enables global control and visual management of the integrated state.

[0113] The above are embodiments of the unified management method for integrating time-series databases and data platforms provided in this application. The following describes embodiments of the unified management system for integrating time-series databases and data platforms provided in this application. Please refer to the following for details.

[0114] This invention also discloses a unified management system integrating a time-series database and a data platform, such as... Figure 2 As shown, the system includes: The connection module is used to obtain the time series database instance configuration information and establish a data connection channel based on the instance configuration information. The data connection channel connects the data platform and the time series database.

[0115] The acquisition module is used to acquire storage group list data and tree model path data from the time series database based on the data connection channel.

[0116] The matching module is used to match the tree model path data and the storage group list data using the longest prefix matching algorithm to obtain the storage group identifier data.

[0117] The splitting module is used to perform hierarchical splitting of tree model path data based on storage group identifier data to obtain relational model structure data.

[0118] The table creation module is used to generate tree model table creation statement data and table model table creation statement data based on relational model structure data. It validates the tree model table creation statement data and table model table creation statement data based on preset table creation constraints to obtain the table structure definition data that passes the validation.

[0119] The parsing module is used to obtain the query statement data of the Structured Query Language, perform syntax parsing on the query statement data of the Structured Query Language to obtain the syntax tree data, and perform semantic verification on the syntax tree data based on the table structure definition data to obtain the syntax tree data that passes the verification.

[0120] The optimization module is used to detect time conditions based on the validated syntax tree data, and rewrite the structured query language query data based on the detection results to obtain optimized structured query language query data.

[0121] The synchronization module is used to monitor permission change event data on the data platform side, generate permission synchronization instruction data based on the permission change event data, and send the permission synchronization instruction data to the time series database through the data connection channel to obtain synchronization result data.

[0122] The management module is used to generate unified management status data based on optimized structured query language statement data and synchronization result data.

[0123] The unified management system integrating time-series databases and data platforms provided in this embodiment offers the following advantages: It establishes standardized data connection channels through a connection module, enabling unified access to time-series databases; it achieves adaptive mapping from tree models to relational models through acquisition, matching, and splitting modules, resolving the issue of inconsistent metadata between heterogeneous models; it implements unified table creation and constraint verification for both models through a table creation module, ensuring standardized modeling; it intelligently optimizes query statements through parsing and optimization modules, improving query performance; it automatically synchronizes permissions through a synchronization module, reducing operational costs; and it achieves unified management of the integration status through a management module, enhancing the system's observability and manageability.

[0124] In this embodiment, the unified management system integrating the time-series database and the data platform is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other programmable logic devices that can provide the above functions.

Claims

1. A unified management method integrating time-series databases and data platforms, characterized in that, The method includes: Obtain the time-series database instance configuration information, and establish a data connection channel based on the instance configuration information. The data connection channel connects the data platform and the time-series database. Based on the data connection channel, obtain the storage group list data and tree model path data from the time series database; The longest prefix matching algorithm is used to match the tree model path data and the storage group list data to obtain the storage group identifier data; Based on the storage group identifier data, the tree model path data is hierarchically split to obtain relational model structure data; Based on the relational model structure data, tree model table creation statement data and table model table creation statement data are generated respectively. Based on the preset table creation constraints, the tree model table creation statement data and table model table creation statement data are validated to obtain the table structure definition data that passes the validation. Obtain the query statement data from the Structured Query Language (SCL), perform syntax parsing on the SCL query statement data to obtain syntax tree data, and perform semantic validation on the syntax tree data based on the table structure definition data to obtain the validated syntax tree data. Based on the valid syntax tree data, the time condition is checked, and the structured query language query statement data is rewritten based on the detection result to obtain the optimized structured query language statement data. Listen to permission change event data on the data platform side, generate permission synchronization instruction data based on the permission change event data, and send the permission synchronization instruction data to the time series database through the data connection channel to obtain synchronization result data; Unified management status data is generated based on the optimized structured query language statement data and the synchronization result data.

2. The method based on claim 1, characterized in that, The step of obtaining the time-series database instance configuration information and establishing a data connection channel based on the instance configuration information includes: Obtain the address information, port information, and authentication configuration information of the time-series database; The authentication method is determined based on the authentication configuration information. The authentication methods include username / password authentication and Kerberos authentication. A data connection channel between the data platform and the time-series database is established based on address information, port information, and authentication configuration information; Perform connectivity verification on the data connection channel. Once the verification passes, store the data connection channel in the connection pool.

3. The method based on claim 2, characterized in that, The method of matching the tree model path data and the storage group list data using the longest prefix matching algorithm to obtain the storage group identifier data includes: calling the time series database metadata interface to obtain the storage group list data, which contains the complete path name of each storage group; Iterate through the tree model path data and perform prefix matching on each storage group path name in the current tree model path data and the storage group list data; The storage group path name with the longest matching length is determined as the storage group to which the current tree model path data belongs, and storage group identification data is generated.

4. The method based on claim 3, characterized in that, The hierarchical splitting of the tree model path data based on storage group identifier data yields relational model structure data, including: Remove the prefix portion corresponding to the storage group identifier data from the tree model path data to obtain the remaining path data; The remaining path data is split according to the hierarchical delimiter to obtain a hierarchical node data sequence; Boundary condition processing is performed on the hierarchical node data sequence, and the boundary condition processing is performed based on preset processing rules; The processed hierarchical node data sequence is categorized into database layer, table layer, and field layer to obtain relational model structure data.

5. The method based on claim 4, characterized in that, The relational model structure data is used to generate tree model table creation statement data and table model table creation statement data respectively. Based on preset table creation constraints, the tree model table creation statement data and table model table creation statement data are validated to obtain validated table structure definition data, including: Retrieve the table creation model type identifier data; When the table creation model type identifier data is a tree model, tree model table creation statement data is generated based on relational model structure data. The tree model table creation statement data includes storage group definition statement and measurement point definition statement. When the table model type identifier data is a table model, the table model creation statement data is generated based on the relational model structure data. The table model creation statement data includes table definition statements and column definition statements. The table creation statement data is validated based on preset table creation constraints. After the validation passes, the table structure definition data is output.

6. The method based on claim 5, characterized in that, The process of obtaining structured query language query statement data, parsing the structured query language query statement data to obtain syntax tree data, and performing semantic validation on the syntax tree data based on table structure definition data to obtain validated syntax tree data includes: Lexical analysis is performed on the query statement data of the structured query language based on the lexical analyzer to obtain a sequence of lexical units; The parser performs syntactic analysis on the sequence of lexical units to generate syntax tree data. Based on the table structure definition, semantic validation is performed on the syntax tree data. The validation includes table name validation, column name validation, and data type validation. After successful validation, output the validated syntax tree data.

7. The method based on claim 6, characterized in that, The time condition for detecting the syntax tree data that has passed verification is used to rewrite the structured query language query statement data based on the detection results, resulting in optimized structured query language statement data, including: Traverse the validated syntax tree data and check whether the conditional clauses contain time column conditions; When the condition clause does not contain a time column condition, a preset time range condition is added to the condition clause to generate the rewritten structured query language statement data; When the conditional clause includes a time column condition, the data from the Structured Query Language query statement will be used as the rewritten Structured Query Language query statement data. Replace the configuration parameters in the rewritten Structured Query Language (SCL) statement data with the actual values ​​to obtain the optimized SCL statement data.

8. The method based on claim 7, characterized in that, The monitoring data platform receives permission change event data, generates permission synchronization instruction data based on the permission change event data, and sends the permission synchronization instruction data to the time-series database via the data connection channel to obtain synchronization result data, including: Register a permission change listener on the data platform side; Based on the permission change listener, permission change event data is captured. Permission change event data includes authorization event data and revocation event data. Based on the permission change event data, the permission subject information, permission object information, and permission operation information are parsed out. Generate permission synchronization instruction data based on permission subject information, permission object information, and permission operation information; Based on the data connection channel, the permission synchronization command data is sent to the time-series database for execution, and the execution return result is received to generate synchronization result data.

9. The method based on claim 8, characterized in that, The generation of unified management status data based on optimized structured query language statement data and synchronization result data includes: Extract query configuration information from the optimized structured query language statement data; Extract permission synchronization status information from the synchronization result data; Combine query configuration information and permission synchronization status information into unified management status data; Write the unified management status data into the management status table.

10. A unified management system integrating a time-series database and a data platform, characterized in that, The system includes: The connection module is used to obtain the time-series database instance configuration information and establish a data connection channel based on the instance configuration information. The data connection channel connects the data platform and the time-series database. The acquisition module is used to acquire storage group list data and tree model path data from the time series database based on the data connection channel; The matching module is used to match the tree model path data and the storage group list data using the longest prefix matching algorithm to obtain the storage group identifier data; The splitting module is used to perform hierarchical splitting of tree model path data based on storage group identifier data to obtain relational model structure data. The table creation module is used to generate tree model table creation statement data and table model table creation statement data based on relational model structure data, and to validate the tree model table creation statement data and table model table creation statement data based on preset table creation constraints to obtain the validated table structure definition data. The parsing module is used to obtain the structured query language query statement data, perform syntax parsing on the structured query language query statement data to obtain syntax tree data, and perform semantic verification on the syntax tree data based on the table structure definition data to obtain the syntax tree data that passes the verification. The optimization module is used to detect time conditions based on the validated syntax tree data, and rewrite the structured query language query statement data based on the detection results to obtain optimized structured query language statement data. The synchronization module is used to monitor permission change event data on the data platform side, generate permission synchronization instruction data based on the permission change event data, and send the permission synchronization instruction data to the time series database through the data connection channel to obtain synchronization result data; The management module is used to generate unified management status data based on optimized structured query language statement data and synchronization result data.

Images