Distributed task intelligent scheduling method and system based on dynamic resource perception

Abstract

The invention belongs to the field of distributed task intelligent scheduling, and discloses a distributed task intelligent scheduling method and system based on dynamic resource awareness, and the method comprises the steps: enumerating a newly added data source, and then carrying out the database dialect customization of the newly added source type; each node is bound with an actuator with the same name; platform task creation: a plurality of components can be created, and the execution mode is component flow sequence execution; a platform task is executed through an xxl-job distributed timed scheduling framework; establishing an ods library corresponding to each golden butterfly financial library; a report is generated and reported for monitoring; and loading and storing the template.

Description

Technical Field

[0001] The present invention belongs to, but is not limited to, the technical field of distributed task intelligent scheduling, and in particular relates to a distributed task intelligent scheduling method and system based on dynamic resource perception. Background Art

[0002] Distributed task intelligent scheduling is a task management technology based on a distributed system architecture. It optimizes task execution efficiency through dynamic resource allocation, intelligent decision-making algorithms, and automated processes. The core functional elements are as follows:

[0003] Intelligent Decision-Making and Dynamic Allocation: Dynamically adjust task distribution strategies based on real-time metrics like node load and network latency, supporting predictive scheduling based on machine learning. Automatically expand computing resources through elastic scaling to handle sudden traffic bursts or large-scale data processing needs.

[0004] Multi-dimensional Task Management: Supports task priority setting, cross-device dependency configuration, and distributed lock control to ensure the orderly execution of complex task chains. Provides a visual workflow orchestration tool that allows you to define task execution logic and data flow paths through a graphical interface.

[0005] High reliability and fault tolerance: Multi-copy storage and automatic failover ensure seamless task switching in the event of node failure. Built-in retry strategies and exception monitoring trigger real-time alarm notifications. Summary of the Invention

[0006] In response to the problems existing in the prior art, the present invention provides a distributed task intelligent scheduling method and system based on dynamic resource perception.

[0007] The present invention is implemented as follows: a distributed task intelligent scheduling method based on dynamic resource perception, characterized in that the distributed task intelligent scheduling method based on dynamic resource perception specifically includes:

[0008] S1: Add a new data source enumeration, and then customize the database dialect for the new source type;

[0009] S2: Bind an executor with the same name to each node;

[0010] S3: Platform task creation, which can create multiple components and execute them in the order of component processes;

[0011] S4: Execute platform tasks through the xxl-job distributed timing scheduling framework;

[0012] S5: Create a corresponding ODS database for each Kingdee financial database;

[0013] S6: Generate reports and submit them for monitoring;

[0014] S7: Load the template and save it.

[0015] Furthermore, the S2 platform is divided into two types of nodes, local nodes and front-end nodes. The local node appName is LocalTaskHandler, and the executor is compHandler; the front-end node appName is configured during specific implementation, and the executor is datax (data synchronization tool) RemoteHandler.

[0016] Furthermore, in S4, the scheduling platform sends the task ID to be executed to the locally registered node (LocalTaskHandler), and then calls the task execution class (com.xmjl.xmcloud.datai.component.comm.core.task.TaskFunc) through its bound executor (compHandler). All component process execution processes are executed in its execute().

[0017] Furthermore, in S5, mapping management is placed in the ODS library. Each ODS library needs to be configured with its corresponding mapping relationship. Adding, deleting, modifying, and checking mapping relationships all require a database connection. The backend uses JDBC to operate on the transmitted database. The code is:

[0018] (com.xmjl.xmcloud.datai.component.datacleansing.*).

[0019] Furthermore, S6, report monitoring, allows relevant departments to review the report submission status of their respective enterprises, including reports that are unsubmitted, submitted, and overdue. Monitoring is performed by report batch, which consists of a six-digit year and month plus the period type (monthly, quarterly, or annual). Report monitoring relies on the report pre-generation feature. Pre-generated report generation generates the required reports (rp_fina_report_batch) and report batch table (rp_fina_batch) based on the overdue date of the period type.

[0020] Furthermore, in S7, the template is saved using the create method of ReportTemplateController.

[0021] Another object of the present invention is to provide a distributed task intelligent scheduling system based on dynamic resource perception, the system specifically comprising:

[0022] Data exchange system, including data integration module, used for data source management, data integration, and data collection;

[0023] The reporting system includes a report monitoring module and a report management module.

[0024] Furthermore, for data source management, the platform only supports database connections in the form of jdbc. When configuring the source type, only four parameters are allowed: jdbcUrl (Java database connection URL), jdbcDriver (JDBC driver), username, and password. The platform only supports two database types: mysql and sqlserver. The platform database connection supports two modes: online and offline mode. The online database is mainly the platform ods library and the report business database; the offline database is mainly the Kingdee financial library.

[0025] Furthermore, the report monitoring module is provided with a one-key reporting configuration function.

[0026] Furthermore, the report management module includes report filling, report query, and report deletion.

[0027] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0028] 1. Core applications in asset supervision:

[0029] (1) Dynamic resource perception and asset scheduling optimization:

[0030] Application scenarios:

[0031] An enterprise's asset allocation usually involves multiple resources (such as office buildings, equipment, vehicles, etc.). The existing manual management method may lead to resource waste or unreasonable scheduling.

[0032] The dynamic resource perception capability of the present invention can monitor asset usage in real time and intelligently optimize asset allocation based on task priorities and resource availability.

[0033] The value it brings:

[0034] Improve asset utilization: reduce idle resources, optimize asset allocation, and improve asset operation efficiency.

[0035] Reduce operation and maintenance costs: Reduce unnecessary resource expenditure and improve capital utilization efficiency through intelligent scheduling.

[0036] Improve asset transparency: Systematize the asset allocation process, reduce human intervention, and improve fairness and transparency.

[0037] (2) Fund usage monitoring and intelligent approval:

[0038] Application scenarios:

[0039] Enterprises need to strictly adhere to budget and approval processes when using funds. Traditional approval methods are usually time-consuming and prone to inefficiency or violations.

[0040] Through the intelligent task scheduling function of the present invention, fund use applications can be intelligently allocated and approved, and fund flows can be analyzed based on historical data to optimize budget management.

[0041] The value it brings:

[0042] Improve approval efficiency: reduce manual approval links, speed up capital turnover, and improve work efficiency.

[0043] Strengthen fund compliance: Use data analysis to automatically detect the rationality of fund use and prevent illegal use of funds.

[0044] Intelligent budget optimization: Dynamically adjust funding allocation plans to ensure that key project funds receive priority support.

[0045] (3) Enterprise big data analysis and risk warning:

[0046] Application scenarios:

[0047] Asset supervision departments need to conduct comprehensive monitoring of a company's financial data, asset operations, contract performance, etc., but the data is scattered and the processing volume is large, making it difficult to achieve efficient supervision using traditional methods.

[0048] The intelligent scheduling method of the present invention can dynamically allocate computing resources, optimize large-scale data processing tasks, and provide real-time risk warnings.

[0049] The value it brings:

[0050] Improve data analysis capabilities: automate the processing of large-scale enterprise data and improve regulatory efficiency.

[0051] Real-time risk warning: Through intelligent scheduling of computing resources, rapid discovery and early warning of financial anomalies, asset loss and other issues can be achieved.

[0052] Auxiliary decision support: Based on data analysis results, optimize decisions such as enterprise reform and resource allocation.

[0053] (4) Asset Supervision Intelligent Audit and Compliance Management:

[0054] Application scenarios:

[0055] Enterprises need to conduct regular asset inventories, financial audits, and compliance checks. Traditional methods rely on manual statistics and audits, which are time-consuming and prone to errors.

[0056] The intelligent scheduling system of the present invention can automatically allocate computing resources according to the complexity and urgency of audit tasks, optimize audit processes, and improve the level of compliance management.

[0057] The value it brings:

[0058] Improve audit efficiency: Intelligently allocate audit tasks, reduce manual review time, and improve audit quality.

[0059] Reduce compliance risks: Automatically identify potential violations and mitigate financial and legal risks.

[0060] Improve audit transparency: Record the audit task scheduling throughout the entire process to ensure audit fairness.

[0061] 2. Practical implementation plan for asset supervision business scenarios:

[0062]

[0063] Second, as auxiliary evidence of the invention's inventiveness, it is also reflected in the following important aspects:

[0064] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:

[0065] Currently, this invention has been applied to corporate asset supervision, but in the future it can be further expanded to a wider range of asset management fields, such as:

[0066] ① Asset supervision platform: By connecting to a wider range of asset management systems, cross-regional and cross-industry intelligent asset supervision can be achieved.

[0067] ② Enterprise group management: Applicable to large enterprise groups, it can intelligently dispatch and supervise the assets, capital flows, and project execution of its subsidiaries.

[0068] ③ Fund management system: can be combined with budget management, procurement and other systems to improve the efficiency of fund use.

[0069] ④ Public infrastructure resource management: Applicable to urban public utilities (such as water supply, power supply, and transportation facilities), improving the scheduling efficiency of public resources and reducing operating costs.

[0070] The application of this invention in the field of asset supervision is primarily reflected in asset scheduling optimization, fund management, data analysis, and intelligent auditing. Its core value lies in improving supervision efficiency, optimizing resource allocation, reducing operating costs, and enhancing asset security. In the future, its market value can be further expanded by extending it to a wider range of supervision platforms, enterprise group management, and public infrastructure management. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of a distributed task intelligent scheduling method based on dynamic resource perception provided by an embodiment of the present invention.

[0072] Figure 2 This is a flow chart of platform scheduling tasks provided by an embodiment of the present invention.

[0073] Figure 3 This is a schematic diagram of mapping relationship management provided by an embodiment of the present invention.

[0074] Figure 4 This is the report preset generation provided by the embodiment of the present invention.

[0075] Figure 5 This is a schematic diagram of template loading provided by an embodiment of the present invention.

[0076] Figure 6 This is a schematic diagram of template storage provided by an embodiment of the present invention.

[0077] Figure 7 This is a module diagram of a distributed task intelligent scheduling system based on dynamic resource perception provided by an embodiment of the present invention.

[0078] Figure 8 This is a design diagram of the collection platform scheduling process provided by an embodiment of the present invention.

[0079] Figure 9 This is a Kingdee financial database transmission diagram provided by an embodiment of the present invention.

[0080] Figure 10 This is a flow chart of report filling provided by an embodiment of the present invention.

[0081] Figure 11 This is a report deletion flow chart provided by an embodiment of the present invention.

[0082] Figure 12 This is a report query flow chart provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0084] like Figure 1 As shown, an embodiment of the present invention provides a distributed task intelligent scheduling method based on dynamic resource perception, which specifically includes:

[0085] S1: Add a new data source enumeration, and then customize the database dialect for the new source type;

[0086] S2: Bind an executor with the same name to each node;

[0087] S3: Platform task creation, which can create multiple components and execute them in the order of component processes;

[0088] S4: Execute platform tasks through the xxl-job distributed timing scheduling framework;

[0089] S5: Create a corresponding ODS database for each Kingdee financial database;

[0090] S6: Generate reports and submit them for monitoring;

[0091] S7: Load the template and save it.

[0092] In S2, the platform is divided into two types of nodes: local nodes and front-end nodes. The appName of local nodes is LocalTaskHandler, and the executor is compHandler. The appName of front-end nodes is configured during implementation, and the executor is RemoteHandler, a datax (data synchronization tool). Platform executors are bound to nodes. For front-end nodes, their executor names are the same, and each node must be bound to an executor with the same name (this can be optimized to have only one executor and multiple nodes).

[0093] In S4, the platform task is composed of component processes. A task contains at least one component. The execution mode of the component process is mainly serial execution, and parallel execution is not supported yet.

[0094] The component is mainly divided into two parts: component configuration and component execution, corresponding to two abstract classes respectively

[0095] (Component configuration: com.xmjl.xmcloud.datai.component.comm.core.conf.ConfFuncDefault).

[0096] (Component implementation: com.xmjl.xmcloud.datai.component.comm.core.task.TaskFuncDefault).

[0097] (1) New components:

[0098] When adding a new component, you only need to inherit the TaskFuncDefault (default class for component execution) and ConfFuncDefault (default class for component configuration storage). The interfaces for adding, deleting, modifying, and querying component configuration parameters do not need to be changed (the interface accepts parameters in the form of a map). The front end must provide these four fields, and the remaining parameters can be customized:

[0099] "confId":"node3JSf3IgPIk1ce", #Component configuration id (generated by the front end);

[0100] "confName":"datax (data synchronization tool) test file", #Component configuration name (fill in when creating the component);

[0101] "confType":"rpCleanFunc", #Storage class name for component configuration (customized during development, managed by Spring);

[0102] "compType":"rpDataCleansing", #The execution class name of the component configuration (customized during development and managed by spring).

[0103] (2) Existing components of the platform:

[0104] Datax (data synchronization tool) component: (under the Datax (data synchronization tool) module):

[0105] "confId":"", #Component configuration id (generated by the front end);

[0106] "confName":"", #Component configuration name (fill in when creating a component);

[0107] "confType":"datax (data synchronization tool) LocalFunc / datax (data synchronization tool) RemoteFunc", #Component processing class name (local / remote);

[0108] "compType":"datax (Data Synchronization Tool) Comp", #datax (Data Synchronization Tool) component.

[0109] Financial data cleaning component: (under the datacleansing module):

[0110] "confId":"", #Component configuration id (generated by the front end);

[0111] "confName":"", #Component configuration name (fill in when creating a component);

[0112] "confType":"rpCleanFunc", #Financial data cleaning class name;

[0113] "compType":"rpDataCleansing", #Financial data cleaning component;

[0114] Report preset generation component: (under datacleansing module):

[0115] "confId":"", #Component configuration id (generated by the front end);

[0116] "confName":"", #Component configuration name (fill in when creating a component);

[0117] "confType":"rpReportPreFunc", #Financial report preset generation class name;

[0118] "compType":"rpRportPreset", #Financial report preset generation component.

[0119] The platform scheduling task flow chart of S4 is as follows Figure 2 As shown in the figure, the scheduling platform (data integration system) sends the task ID to be executed to the locally registered node (LocalTaskHandler), and then calls the task execution class (com.xmjl.xmcloud.datai.component.comm.core.task.TaskFunc) through its bound executor (compHandler). All component process execution processes are performed in its execute() method.

[0120] The above S5, mapping management is mainly because the codes of fields with the same meaning in different databases are different, and these codes need to be converted into unified codes in the platform library. Figure 3 As shown, mapping management is placed in the ODS library, and each ODS library needs to be configured with its corresponding mapping relationship.

[0121] For each Kingdee financial database, a corresponding ODS database must be established. The two have a one-to-one correspondence. Therefore, adding, deleting, modifying, and querying mapping relationships require a database connection. The backend uses JDBC to operate on the transmitted database. The code is as follows:

[0122] (com.xmjl.xmcloud.datai.component.datacleansing.*).

[0123] The aforementioned S6, reporting monitoring is for the relevant departments to check the reporting status of the companies under their charge, including reports on unreported, reported, and overdue status. Monitoring is based on report batches, which are the year and month (6 digits) plus the period type (monthly, quarterly, and annual reports). The implementation of the reporting monitoring function relies on the report preset generation function. The report preset generation will generate the report (rp_fina_report_batch) and report batch table (rp_fina_batch) that needs to be filled in the current period in advance based on the overdue date of the period type ( Figure 4 ).

[0124] The S7 template is loaded as Figure 5 As shown, the template is saved using the create method of ReportTemplateController, such as Figure 6 shown.

[0125] like Figure 7 As shown, an embodiment of the present invention provides a distributed task intelligent scheduling system based on dynamic resource perception, which specifically includes:

[0126] Data exchange system, including data integration module, used for data source management, data integration, and data collection;

[0127] The reporting system includes a report monitoring module and a report management module.

[0128] For data source management, the platform only supports JDBC database connections. When configuring the source type, only four parameters are allowed: jdbcUrl (Java database connection URL), jdbcDriver (JDBC driver), username, and password. The platform only supports MySQL and SQL Server database types.

[0129] 1. MySQL:

[0130] jdbc_url: jdbc:mysql: / / ${ip}:${port} / ${database}?useUnicode(directive set)=true&characterEncoding=utf8;

[0131] jdbc_driver: com.mysql.cj.jdbc.Driver;

[0132] name: (fill in when configuring the connection);

[0133] password: (fill in when configuring the connection).

[0134] 2. sqlserver:

[0135] jdbc_url:jdbc:sqlserver: / / ${ip}:${port};DatabaseName=${database};

[0136] jdbc_driver: com.microsoft.sqlserver.jdbc.SQLServerDriver;

[0137] name: (fill in when configuring the connection);

[0138] password: (fill in when configuring the connection);

[0139] (Note: ${} acts as a station and needs to be replaced when configuring the connection).

[0140] New source types:

[0141] You need to add a new data source enumeration to the data source type enumeration class (com.xmjl.xmcloud.common.databasequery.constant.DatabaseType) and customize the database dialect for the new source type. This can be achieved by inheriting the com.xmjl.xmcloud.common.databasequery.meta.BaseDatabaseMeta class and overriding methods. This is mainly used to query database structure, table information, and field information.

[0142] The platform database connection supports two modes, online and offline modes. Online means that the acquisition system can directly connect to the database, while offline means that the acquisition system cannot connect and can only access it indirectly through other means.

[0143] The online database is mainly the platform ODS database and the report business database; the offline database is mainly the Kingdee financial database.

[0144] The data collection and collection platform scheduling process design diagram is as follows Figure 8 shown.

[0145] 1. Local node:

[0146] (1) The platform has an embedded xxl-job admin. When the platform starts, a local node (built-in) is also started to register the admin (user name). This means that the platform registers itself.

[0147] (2) The local built-in node appname is LocalTaskHandler, and the task component executor is compHandler.

[0148] (3) All automatic tasks of the platform are configured in the form of components in the task, and then the task is bound to the scheduler. The scheduler passes the task ID to the executor (compHandler) in the local node (LocalTaskHandler), and the task is executed through the local node executor.

[0149] 2. Pre-node:

[0150] (1) The front-end node registration uses the node registration method of xxl-job, and the datax (data synchronization tool) remote call communication mechanism also uses the xxl-job communication mechanism (see xxl-job communication for details).

[0151] (2) The front-end node requires a front-end user. When starting, it obtains the access token from the platform through the front-end user. The username and password of the front-end user need to be configured in the front-end yml file.

[0152] 3. Kingdee financial statement data collection:

[0153] (1) Collection process: In the front end, the report data in the Kingdee database is exported through datax (data synchronization tool) to form a data text file, and then uploaded to the platform. The platform imports the data text file into the corresponding ODS library, and then cleans the data in the ODS library into the report business library according to the mapping relationship. Finally, fill in the report to view the required files and data.

[0154] (2) In actual business, a collection task consists of three components: the first is a remote call of datax (data synchronization tool), the second is a local call of datax (data synchronization tool), and the third is a financial data cleaning component.

[0155] (3) Each pre-processor needs to be configured with an ODS library on the platform. This is mainly to avoid the same code (instruction set) having different meanings in different Kingdee libraries, which would cause the platform to be unable to map (see Mapping Management for details).

[0156] Report data flow is as follows Figure 9 shown.

[0157] The report monitoring module sets a one-click reporting configuration function, which mainly binds the scheduling of collection tasks with enterprise users, so that users can realize the one-click reporting (collection) function.

[0158] The report management module includes:

[0159] Report filling process Figure 10 shown.

[0160] Note: The entire saving process involves the following tables:

[0161] 1. rp_fina_report_data (report data);

[0162] 2. rp_fina_report_data_map (report data mapping, used for deletion);

[0163] 3. rp_fina_report_batch (report record);

[0164] 4. rp_fina_report_batch_exp (report instance related cross-reference formula);

[0165] 5. Mongodb (non-relational database) structured data.

[0166] The report deletion process is as follows Figure 11 shown.

[0167] Note: Delete logic:

[0168] In the report data table (rp_fina_report_data), data with the same meaning (such as cash and cash equivalents, end-of-period balance) exists in different reports and is only saved once in the data table. Therefore, it cannot be deleted directly from the data table. Therefore, a data mapping table (rp_fina_report_data_map) is established to assist in deletion. This table stores data from each table. If the number of times data with the same meaning appears in this table is greater than 1 (used by multiple tables), the data in the data table will not be deleted. If it is equal to 1 (used only by the current table), the data can be deleted.

[0169] Report query process is as follows Figure 12 shown.

[0170] Note: The report query idea is:

[0171] First, query the cache in MongoDB (a non-relational database). If the cache exists, return the report directly. If not, query the database, assemble the report according to the data template, and store the result in MongoDB (a non-relational database).

[0172] 1. Specific application fields or related products of the present invention.

[0173] 1. Core applications in asset supervision:

[0174] (1) Dynamic resource perception and asset scheduling optimization:

[0175] Application scenarios:

[0176] An enterprise's asset allocation usually involves multiple resources (such as office buildings, equipment, vehicles, etc.). The existing manual management method may lead to resource waste or unreasonable scheduling.

[0177] The dynamic resource perception capability of the present invention can monitor asset usage in real time and intelligently optimize asset allocation based on task priorities and resource availability.

[0178] The value it brings:

[0179] Improve asset utilization: reduce idle resources, optimize asset allocation, and improve asset operation efficiency.

[0180] Reduce operation and maintenance costs: Reduce unnecessary resource expenditure and improve capital utilization efficiency through intelligent scheduling.

[0181] Improve asset transparency: Systematize the asset allocation process, reduce human intervention, and improve fairness and transparency.

[0182] (2) Fund usage monitoring and intelligent approval:

[0183] Application scenarios:

[0184] Enterprises need to strictly adhere to budget and approval processes when using funds. Traditional approval methods are usually time-consuming and prone to inefficiency or violations.

[0185] Through the intelligent task scheduling function of the present invention, fund use applications can be intelligently allocated and approved, and fund flows can be analyzed based on historical data to optimize budget management.

[0186] The value it brings:

[0187] Improve approval efficiency: reduce manual approval links, speed up capital turnover, and improve work efficiency.

[0188] Strengthen fund compliance: Use data analysis to automatically detect the rationality of fund use and prevent illegal use of funds.

[0189] Intelligent budget optimization: Dynamically adjust funding allocation plans to ensure that key project funds receive priority support.

[0190] (3) Enterprise big data analysis and risk warning:

[0191] Application scenarios:

[0192] Asset supervision departments need to conduct comprehensive monitoring of a company's financial data, asset operations, contract performance, etc., but the data is scattered and the processing volume is large, making it difficult to achieve efficient supervision using traditional methods.

[0193] The intelligent scheduling method of the present invention can dynamically allocate computing resources, optimize large-scale data processing tasks, and provide real-time risk warnings.

[0194] The value it brings:

[0195] Improve data analysis capabilities: automate the processing of large-scale enterprise data and improve regulatory efficiency.

[0196] Real-time risk warning: Through intelligent scheduling of computing resources, rapid discovery and early warning of financial anomalies, asset loss and other issues can be achieved.

[0197] Auxiliary decision support: Based on data analysis results, optimize decisions such as enterprise reform and resource allocation.

[0198] (4) Asset Supervision Intelligent Audit and Compliance Management:

[0199] Application scenarios:

[0200] Enterprises need to conduct regular asset inventories, financial audits, and compliance checks. Traditional methods rely on manual statistics and audits, which are time-consuming and prone to errors.

[0201] The intelligent scheduling system of the present invention can automatically allocate computing resources according to the complexity and urgency of audit tasks, optimize audit processes, and improve the level of compliance management.

[0202] The value it brings:

[0203] Improve audit efficiency: Intelligently allocate audit tasks, reduce manual review time, and improve audit quality.

[0204] Reduce compliance risks: Automatically identify potential violations and mitigate financial and legal risks.

[0205] Improve audit transparency: Record the audit task scheduling throughout the entire process to ensure audit fairness.

[0206] 2. Relevant evidence of the technical effects obtained by the embodiments of the present invention.

[0207] Actual implementation plan for asset supervision business scenarios:

[0208]

[0209] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0210] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A distributed task intelligent scheduling method based on dynamic resource perception, characterized in that: The method specifically includes: S1: Add a new data source enumeration, and then customize the database dialect for the new source type; S2: Bind an executor with the same name to each node; S3: Platform task creation, which can create multiple components and execute them in the order of component processes; S4: Execute platform tasks through the xxl-job distributed timing scheduling framework; S5: Create a corresponding ODS database for each Kingdee financial database; S6: Generate reports and submit them for monitoring; S7: Load the template and save it.

2. The distributed task intelligent scheduling method based on dynamic resource perception according to claim 1 is characterized in that: In the S2 platform, there are two types of nodes: local nodes and front-end nodes. The appName of the local node is LocalTaskHandler, and the executor is compHandler. The appName of the front-end node is configured during the specific implementation, and the executor is dataxRemoteHandler.

3. The distributed task intelligent scheduling method based on dynamic resource perception as claimed in claim 1, characterized in that: In S4, the scheduling platform sends the task ID to be executed to the locally registered node (LocalTaskHandler), and then calls the task execution class (com.xmjl.xmcloud.datai.component.comm.core.task.TaskFunc) through its bound executor (compHandler). All component process execution processes are executed in its execute().

4. The distributed task intelligent scheduling method based on dynamic resource perception as claimed in claim 1, characterized in that: In S5, mapping management is placed in the ODS library. Each ODS library needs to be configured with its corresponding mapping relationship. Adding, deleting, modifying, and checking mapping relationships all require a database connection. The backend uses JDBC to operate on the transmitted database. The code is: (com.xmjl.xmcloud.datai.component.datacleansing.*).

5. The distributed task intelligent scheduling method based on dynamic resource perception as claimed in claim 1, characterized in that: The aforementioned S6, reporting monitoring, is for relevant departments to check the reporting status of the enterprises under their charge, including reports on unreported, reported, and overdue status. Monitoring is based on report batches, and a report batch consists of the year and month (6 digits) plus the period type (monthly report, quarterly report, annual report). The implementation of the reporting monitoring function depends on the report preset generation function. The report preset generation will generate the report (rp_fina_report_batch) and report batch table (rp_fina_batch) that needs to be filled in in advance based on the overdue date of the period type.

6. The distributed task intelligent scheduling method based on dynamic resource perception as claimed in claim 1, characterized in that: In S7, templates are saved using the create method of ReportTemplateController.

7. A distributed task intelligent scheduling system based on dynamic resource perception as described in claims 1-6, characterized in that: The system specifically includes: Data exchange system, including data integration module, used for data source management, data integration, and data collection; The reporting system includes a report monitoring module and a report management module.

8. The distributed task intelligent scheduling method based on dynamic resource perception as claimed in claim 7, characterized in that: Regarding data source management, the platform only supports database connections in the form of jdbc. When configuring the source type, only four parameters are allowed: jdbcUrl, jdbcDriver, username, and password. The platform only supports two database types: mysql and sqlserver. The platform database connection supports two modes: online and offline. The online database is mainly the platform ods library and the report business database; the offline database is mainly the Kingdee financial library.

9. The distributed task intelligent scheduling method based on dynamic resource perception according to claim 7, characterized in that: The report monitoring module is provided with a one-key reporting configuration function.

10. The distributed task intelligent scheduling method based on dynamic resource perception according to claim 7, characterized in that: The report management module includes report filling, report query, and report deletion.

Images