TWI726431B - A methodology and architecture of task nodes systems - Google Patents

Abstract

This invention relates to a methodology and architecture of task nodes systems, and especially refers to a methodology that the microservices can be standardized through the invention and can be cooperatively operated in a distributed system composed of hierarchical task nodes processes. The invention divides the business logic and the service flow, and disassembles the business logic into various tasks with low complexity and low coupling and bound to the service process with high complexity in the form of work nodes to be processed by the microservices. In the system, a microservice needs to be bound to at least one working node, and the normalized procedure is completed according to the task specification information of the bound working node. When the business logic has an additional change, the present invention only needs to add a micro-service task to the specific working node; when the service process has an additional change, the invention can be adapted by modifying working nodes and processes thereof, thereby achieving the purpose of decreasing the range of changes and reducing complexity, and reducing the time for human development and the time cost of the test.

Description

Task type working node system and method

The present invention relates to a task-based work node system and method, and particularly relates to standardizing microservices to enable collaborative operation in a distributed system designed and composed of a hierarchical task node process.

The development of traditional distributed, 3-tier architecture information system usually uses front-end user interface layer (WEB), back-end data processing middle layer (AP), back-end data processing bottom layer (DB) to construct related services, and front-end uses interface layer (WEB) After receiving the information, the back-end is transferred to the back-end. The back-end data processing middle layer (AP) or the back-end data processing bottom layer (DB) processes the data and stores it, and feeds the processing results back to the front-end user interface layer (WEB) . Once the business logic changes, the usually affected range may include the front-end user interface layer (WEB) must add, modify or change the received information, the back-end data processing middle layer (AP) or the back-end data processing bottom layer (DB) ) Must also be synchronized to make API additions and changes. APIs are usually integrated for maximum performance processing. Therefore, if the new and old logic coexist and increase the complexity of the API, it will undoubtedly increase the cost of manpower development and overall testing time in the API, which will be difficult to respond quickly. As service demand changes, the scalability and scalability of the overall system will be limited by this.

Today's information systems also use IoT technology to try to improve the bottleneck of the above architecture. For example, China's patent number "I465944", the patent name is "a data transmission system and method". The purpose of the invention is to provide a network for cloud computing. In the data transmission environment, the data transmission system and method are to cut and separate the data and the transmission logic. There is no need to consider the content of the transmission data type. Instead, the logic in the transmission rule is used as the basis for the transmission logic, according to different companies Code and/or POS code to complete the exchange of data messages between the cloud and the point of sale POS, to solve the original data transmission structure that requires data and/or records and/or new records, table structure companies Code, POS code and transmission logic are tied together.

Although the invention reduces the coupling between the data structure and the transmission logic, the logic of the data flow between the point of sale POS and the cloud still needs to be controlled through the Policy Center. Therefore, the invention still cannot solve the problem that the increase in business complexity may lead to the logic center. Excessive concentration cannot be effectively handled and may become a system bottleneck problem.

The task-based work node (referred to as work node for short) of the present invention is mainly used to disassemble business logic into hierarchical tasks and bind the identification information of the system process node concept to specific microservices. Each microservice operates according to the binding identification information with the concept of a system process node, and the microservice operates in this method, which is called the microservice standardization process. The transmission basis of the microservice standardization program of the present invention adopts the message queuing telemetry transport (MQTT) technology. MQTT is the M2M communication protocol designed for the Internet of Things. The network bandwidth and hardware requirements are very small, and it is an extremely portable communication protocol. 讯 Agreement. MQTT communication is a one-to-many M2M transmission, using a publish/subscribe message transmission mechanism. This mechanism contains 4 main elements: Publisher, Subscriber, and Topic ), message broker (Broker). Publisher is the source of the message. It sends messages with Topic information to the Broker. Subscribers register the Topic of the message that they want to receive with the Broker. For example, a Publisher publishes a message with a topic of "Dinner", as long as there is a subscription to the Broker. Subscribers whose name is "Dinner" can receive this message. Based on this technical feature, the microservice of the present invention can subscribe the source node information to the Broker on the Internet based on the binding identification information with the concept of the system process node in the normalization process, and after receiving the subscribed source node information If the status or result is the latest and correct, it means that the task processing sub-procedure of the bound identification information is actively executed. After completion, the broker will publish the latest status or result information of the task processing according to the bound identification information; or according to the bound identification information. The identification information with the concept of system process node automatically executes the task processing sub-procedure of the bound identification information at a fixed time, and then releases the latest status or result information of the task processing to the Broker based on the bound identification information after completion; or receives the external system Request the execution of the task processing sub-procedure of the bound identification information, and after completion, release the latest status or result information of the task processing to the Broker based on the bound identification information. The present invention provides the microservices of a distributed system to process various tasks in a standardized and asynchronous manner through the microservice standardization process, simplifies the complexity of the internal business processing of the microservices, and flexibly adjusts the system service process to improve The performance of the overall system and the scalability of the optimization service.

The main purpose of the task-based work node system and method of the present invention is to provide a standardized mechanism for each microservice and enable it to cooperate in a distributed system designed and composed by a hierarchical task node process. The present invention cuts the business logic and service process, and cuts the business logic and service process The service logic is disassembled into low-complexity and low-coupling tasks, and is bound to the high-complexity service process in the form of work nodes to be processed by each microservice. Each microservice can complete the standardized process only by operating a standardized operation process and operating according to the task specification information of the working node. The introduction of the method of the present invention will allow a system to realize the standardization of microservices, provide a standard operating process of microservices to handle simple and simplified internal business operations, and the service process of the overall system can be expanded more flexibly, thereby optimizing the overall system performance, Improve service scalability.

The present invention provides a task-based work node system, which includes: a task setting module for setting each task in the corresponding work phase in the overall service process to generate task specification information for each work node. Tasks are disassembled from business logic; a task data module is used to store the task specification information of each work node; at least one microservice is each including a main program and a task processing subprogram, the at least one microservice At least one is bound to at least one of the working nodes, and at least one of the at least one microservice obtains the task specification information of at least one of the working nodes to which it is bound from the task data module , To operate according to the task specification information of at least one of the working nodes to which it is bound, so as to output the working node status or result information after operation; a working node status result update module for the at least one The work node status or result information output by at least one of the microservices is updated to the task data module; and at least one project message proxy service module supports the Message Queue architecture and is used to provide the at least one microservice Agency service for publishing or subscribing messages between services.

In the task-based work node system, at least one of the at least one microservice obtains from the task data module the task specification information of at least one of the work nodes to which it is bound, including timing standard settings and source information Set for the at least one microservice to be based on the time Sequence standard setting and the source information setting determine whether to execute its task processing subroutine, so that after completing its task processing subroutine, output the work node status or result information for the work node status result update module to output the work node status Or the result information is updated to the task data module, and then the work node title with the work node status or result information is released to the at least one project information agent service module.

The present invention also provides a task-based work node method, which includes the following steps: disassembling business logic into tasks; setting each task in the corresponding work stage in the overall service process to generate task specification information for each work node; Executing the microservice standardization process, including: Process 1: At least one of at least one microservice retrieves at least one of the working nodes to which it is bound according to the main program and task processing subprogram contained in it Task specification information; and enabling at least one of the at least one microservice to operate according to the task specification information of at least one of the working nodes to which it is bound.

The microservice normalization process further includes: Process 2: At least one of the at least one microservice checks the timing standard setting and source information setting in the task specification information of at least one of the working nodes to which it is bound; and 3: If the timing standard is set to a non-null value and the source information is set to a null value, at least one of the at least one microservice will execute its own task processing subroutine according to the timing standard setting; if the timing standard is set to Null value and the source information setting is not null, then at least one of the at least one microservice is based on the source information setting to the task specification information of at least one of the working nodes to which it is bound to the project message agent service Subscribe to the source node title with the source node status result information to query whether the source node status result information of the source node title of the subscription is the latest and correct after receiving the source node title of the subscription. If so, execute The task processing subroutine of its own, if not, it will not Execution; if the timing standard is set to a null value and the source information setting is also a null value, at least one of the at least one microservice receives a request from an external system and passively executes its own task processing subroutine; and if the timing If the standard setting is not a null value and the source information setting is not a null value, at least one of the at least one microservice does not execute its own task processing subroutine.

Then, after at least one of the at least one microservice operates according to the task specification information of at least one of the work nodes to which it is bound, the work node status or result information is output for update, and then the at least one microservice At least one of the services publishes the work node title with the work node status or result information.

In other words, the system of the present invention can be composed of at least a task setting module, a task data module, a working node status result update module, at least one microservice, and at least one project message agent service module. The present invention first disassembles the business analysis into various tasks, and uses the task setting module to set the corresponding working stage of each task in the overall system process to generate the task specification information of the working node and store it in the task data module to provide microservices Binding the basis of specific work nodes for subsequent normalization procedures.

In addition, the microservice referred to in the present invention can be composed of a main program and one or more task processing subprograms. Each task processing subprogram is a different business logic processing unit and needs to be bound to a task specification information of a working node. To operate. A microservice will determine the timing of executing the task based on the source information setting and timing standard setting of the task specification information of the work node: (1) If the source information setting is not a null value and the timing standard is set to a null value, the task will be performed according to the source information setting The project information agent service module of the node subscribes to the source node title with the latest status and result information of the source node. Once the subscription title information is received, the latest status or result information of the source node will be checked and compared with the task data module. If it is confirmed to be correct Actively perform the bound work The task processing subroutine of the node. After completion, the status or result information of the working node will be sent to the working node status result update module according to the status result setting of the bound task specification information of the working node, and the status result update module will send it to the working node status result update module. After updating to the task data module, publish the work node title with the work node status or result information to the project information agent service module of the work node (2) If the source information is set to a null value and the timing standard setting is not a null value , The task processing subroutine of the bound working node is automatically and regularly executed according to the timing standard. After completion, the state or result information of the working node is sent to the working node status result update module according to the bound status result setting, and the working node After the status result update module updates it to the task data module, the project message agent service module of the worker node publishes the job node title with the job node status or result information (3) If the source information is set to a null value and The timing standard setting is also a null value, then the task processing subroutine that receives the external system request to passively execute the bound identification information, after completion, sends the status or result information of the working node to the working node according to the bound status result setting. The result is updated Module: After the work node status result update module updates it to the task data module, the work node title with the work node status or result information is issued to the project information agent service module of the work node.

In order to avoid the defects of crowding or conflict in the setting of multiple service flow programs under the microservices cooperative operation in the overall system, the working node setting of the present invention assigns each task to the service flow sequence of the overall system based on the task level setting principle. The task level setting principle divides the task level into N levels (T1~TN), level T1 is the highest level, and level TN is the lowest level. The level of a work node is determined by the stage of the task in the overall service process, and there is a parent-child relationship between work nodes; the higher-level work node is the parent node of the lower-level work node, and the lower-level work node is The child node of the higher level. A task processing subprogram needs to be bound to a work section Point, the execution of the task processing subroutine can be triggered by the source node of the bound working node. And the source node is divided into child node or parent node; if it is triggered by multiple source nodes, the source node must be the child node of the working node to set it, and if only triggered by a single source node, the source node must be the working node Only the parent node of a node or nodes of the same level can be set. The source node of a working node can only be set to its parent node or child node, and cannot coexist at the same time.

21‧‧‧Task Setting Module

22‧‧‧Task Data Module

231..23n‧‧‧Microservice

241..24n‧‧‧Project Information Agency Service Module

25‧‧‧Work node status result update module

51~55‧‧‧Microservice

61~66‧‧‧Task

MQ1‧‧‧Project Message Agency Service Module

S31~S34‧‧‧Step

S331, S332, S3331~S3338‧‧‧Step

T1~T6‧‧‧Task level

T5-3, T6-4, T3-5, T2-1T4-6, T1-2‧‧‧Working node

Figure 1 is a schematic diagram of the task-based worker node system of the present invention; Figure 2A is a schematic diagram of the task-based worker node method of the present invention; Figure 2B is a schematic diagram of the microservice standardization process of the present invention; Figure 3 is the present invention The working node process example diagram; and Figure 4 is an example diagram of the microservice subscription publishing title of the present invention.

Please refer to the schematic diagram of the task-based worker node system in Figure 1. The task-based worker node system of the present invention includes a task setting module 21, a task data module 22, one or more microservices 231..23n, one or A plurality of project message agent service modules 241..24n, a working node status result update module 25.

The present invention cuts the business logic and service process, and disassembles the business logic into tasks with low complexity and low coupling for the development of microservices. Each microservice of the microservice 231..23n can be made up of a main program and One or more task processing sub-programs, each task processing The sub-programs perform different business logic processing and all need to be bound to the task specification information of a working node for operation. The task specification information of the working node is the execution of each microservice by the manager using the task setting module 21 The task setting has task identification information with the concept of a system process node, and is stored in the task data module 22.

After the microservice 231..23n obtains the task specification information of the individually bound work node from the task data module 22, it determines to use one of the following methods to perform the task operation based on the obtained task specification information: (1) For the individual binding The project message agent service module 241..24n subscribes to the source node and actively executes it after receiving the confirmation subscription header information; (2) actively executes it regularly; and (3) receives external system requests for passive execution. One or more microservices 231..23n publish the processing status results to the individually bound project message agent service modules 241..24n after completing the task execution. In addition, the microservice 231..23n can send the status or result information of the worker node to the worker node status result update module 25 after completing the task execution, and the worker node status result update module 25 transmits the work transferred by each microservice The node status or result information is updated in the task data module 22.

The task setting module 21 can provide the job node task specification information of each task set by the manager according to the task level setting principle. The task level setting principle of the present invention is to divide the task level into N levels (T1~TN), with level T1 being the highest level and level TN being the lowest level. The work node with a higher level is the parent node of the lower level, and the work node with a lower level is the child node of the higher level. A working node can be triggered by its source node. The source node is divided into a child node or a parent node; if it is triggered by multiple source nodes, the source node must be a child node of the work node to be set to it, and if only triggered by a single source node, the source node must be the work node Node's parent node Or nodes of the same level can be set to it. The source node of a working node can only set its parent node or child node, and cannot coexist at the same time.

The task specification information of the working node includes at least: the microservice setting of the working node, the project message proxy setting, the source information setting, the timing standard setting, and the status result setting. The timing standard setting includes at least each unit period, the number of executions per unit period, and the execution start time of each unit. Its composition structure is: {per unit period}TS{execution times per unit period}-{execution start time per unit}. Each unit period is hour/day/month/week/year as a unit period, and its codes are _byHourly/_byDaily/_byMonthly/_byWeekly/_byYearly; the number of execution times per unit period is 2 codes; the execution start time of each unit is at least 5 code. If month is the unit time period, you can set first as the execution day of the first day of the month, last as the execution day of the last day of the month, and the specified date of the month as the execution day, separated by "|"; if With week as the unit time period, you can set first as the execution day of the first day of the week, last as the execution day of the last day of the week, and the specified date of the week as the execution day, separated by "|"; Year is per unit time period, you can set 4 digits as the designated execution day and separated by "|". E.g:

(1) The time period is per hour, and the execution time unit is calculated in minutes

˙_byHourlyTS01-00:00 (60/1=60: Calculate the number of times in hours, execute once per hour, and execute once every 60 minutes from 00:00)

˙_byHourlyTS12-00:00 (60/12=5: Calculate the number of times in hours, execute 12 times per hour, and execute once every 5 minutes from 00:00)

˙_byHourlyTS06-00:00 (60/6=10: Calculate the number of times in hours, execute 6 times per hour, and execute once every 10 minutes from 00:00)

(2) Take daily as the unit time period, and the execution time unit is calculated in hours

˙_byDailyTS01-00:00 (24/1=24: Calculate the number of times per day, execute once a day, and execute once every 24 hours from 00:00)

˙_byDailyTS12-00: 00 (24/12=2: Calculate the number of times per day, execute 12 times a day, and execute once every 2 hours from 00:00)

˙_byDailyTS24-00: 00 (24/24=1: Calculate the number of times per day, execute 24 times a day, and execute once every 1 hour from 00:00)

(3) Take monthly as the unit time period, and the execution time unit is calculated by day

˙_byMonthlyTS01-first-00:00 (calculated by the number of months, executed once a month, executed at 00:00 on the first day of each month)

˙_byMonthlyTS01-last-12: 00 (calculated by the number of months, executed once a month, executed at 12:00 on the last day of each month)

˙_byMonthlyTS02-01|15-00:00 (calculated by the number of months, executed twice a month, executed at 00:00 on the 1st and 15th of each month)

(4) Use weekly as the unit time period, and the execution time unit is calculated by the day

˙_byWeeklyTS01-first-00:00 (calculated by the number of weeks, executed once a week, executed at 00:00 on the first day of the week)

˙_byWeeklyTS01-last-12: 00 (calculated by weeks, executed once a week, executed at 12:00 on the last day of the week)

˙_byWeeklyTS02-01|04-00:00 (calculated on a weekly basis, executed twice a week, executed every Monday and Thursday at 00:00)

(5) The unit of time is per year, and the unit of execution time is calculated by month

˙_byYearlyTS01-0101-00: 00 (calculated by the number of years, executed once a year, (Executed at 00:00 on the first day of each year)

˙_byYearlyTS01-1231-12: 00 (calculated by the number of years, executed once a year, executed at 12:00 on the last day of the year)

˙_byYearlyTS02-0101|0601-00:00 (calculated by the number of years, executed twice a year, executed at 00:00 on 1/1, 6/1 each year)

Please refer to the schematic diagram of the task-based worker node method in Figure 2A and the microservice standardization process diagram in Figure 2B.

In step S31, the business logic is disassembled into tasks, and then in step S32, each task is set in the corresponding work stage in the overall service process to generate task specification information for each work node, and then in step S33, Carry out the microservice standardization process. The microservice standardization process includes step S331: at least one of the one or more microservices retrieves the task specifications of the working node bound to the matching microservice settings from the task data module according to its own main program and task processing subprogram Information, and then go to step S332. Step S332: Check the timing standard setting and source information setting in the task specification information of the working node; if the timing standard setting shown in step S3331 is not a null value and the source information is set to a null value, proceed to step S3332: the at least one The microservice executes its own task processing subroutines according to the timing standard settings; if the timing standard is set to a null value and the source information is set to a non-null value as shown in step S3333, proceed to step S3334: the at least one microservice is based on the source The information setting subscribes the source node title with the source node status result information to the project message agent service module of the working node, and after receiving the subscribed source node title, it queries the task data module and compares what the source node title contains If the status result information is up-to-date and correct, if it is, it will execute its own task processing subroutine; if such If the timing criterion shown in step S3335 is set to a null value and the source information setting is also a null value, proceed to step S3336: the at least one microservice receives an external system request to passively execute its own task processing subroutine; if as shown in step S3337 If the timing standard setting is not a null value and the source information setting is not a null value, then proceed to step S3338: the at least one microservice does not execute its own task processing subroutine.

After completing steps S3331~S3338, step S34 is executed: the at least one microservice transmits the status or result information of the worker node to the status result update module according to the status result setting, and the status result update module updates it After reaching the task data module, the microservice then publishes the work node title with the work node status or result information to the project information agent service module of the work node.

The specific description is as shown in the example diagram of the working node process in Figure 3. The example cuts the business logic and service process, and disassembles the business logic into raw material purchase, product production, raw material inventory statistics, product sales, product sales analysis, product inventory statistics, etc. 6 tasks 61~66, the task level is divided into task levels T1~T6, and the service process is regular task 61: raw material purchase (work node: T5-3) and then task 62: product production (work node: T6-4) ) Starts to execute, and task 63: raw material inventory statistics (work node: T3-5) must be based on the immediate response inventory statistics of in and out of the warehouse; each task 64: product sales (work node: T2-1) must be performed immediately after task 65 : Product sales analysis (work node: T4-6); task 62: product production (work node: T6-4) completion and task 64: product sales (work node: T2-1) completion must also be synchronized to task 66: products Inventory statistics (work node: T1-2). The following describes the task specification information and operating procedures of the work nodes of the microservices bound to each task 61~66:

Task 61: Purchase of raw materials. The microservice retrieves the task specification information of the working node T5-3 bound to the matching microservice settings from the task data module according to its own main program AP2 and task processing subprogram AP2SP1 for operation. The timing standard of working node T5-3 is set to "_byDailyTS02-00:30", the source information is set to null, the status result is set to "rB31/T5-3", and the project message agent is set to "MQ1", according to the microservice specifications The process determines that because the timing standard setting of the working node T5-3 is not a null value and the source information is set to a null value, the microservice main program AP2 will set the daily timed task processing subroutine AP2SP1 twice according to the timing standard. It will be executed every 12 hours starting at 00:30 on the day. After the execution is completed, the status or result information of the worker node is sent to the worker node status result update module according to the status result setting. After the worker node status result update module updates it to the task data module, it is then sent to the project according to the project message agent setting The message broker service module (MQ1) publishes the title information rB31/T5-3 containing the status result setting information.

Task 62: Product production. The microservice retrieves the task specification information of the working node T6-4 bound to the matching microservice settings from the task data module according to its own main program AP1 and task processing subprogram AP1SP1 for operation. The timing standard of the working node T6-4 is set to null, the source information is set to "rB31/T5-3", the status result is set to "rB32/T6-4", and the project message agent is set to "MQ1", according to the microservice specifications The process determines that because the timing standard of the working node T6-4 is set to a null value and the source information is set to a non-null value, the microservice main program AP1 will send the project message agent service module (MQ1 ) Subscribe to the source node title information rB31/T5-3 that contains the status result information, and after receiving the subscription title rB31/T5-3 information, query the task data module to check whether the information contained in the rB31/T5-3 title is The latest and correct, if yes, execute the task processing sub-program AP1SP1. After completion The status result setting sends the status or result information of the worker node to the status result update module of the worker node. After the status result update module of the worker node updates it to the task data module, it is then sent to the project information agent service module according to the project information agent settings. The group (MQ1) publishes the title information rB32/T6-4 containing the status result setting information.

Task 63: Raw Material Inventory Statistics and Task 66: Product Inventory Statistics, by the microservice according to its own main program AP3 and task processing subprograms AP3SP1 and AP3SP2 to retrieve the working node T3 bound by the matching microservice settings from the task data module -5. The task specification information of T1-2 is operated. The timing standard of the work node T3-5 is set to null, the source information is set to "rB31/T5-3, rB32/T6-4", the status result is set to "rB32/T3-5", and the project message agent is set to "MQ1" According to the microservice standardization process, it is determined that the timing standard of the working node T3-5 is set to a null value and the source information setting is not a null value. Therefore, the main microservice program AP3 will send the message to the project based on the source information setting and the project message agent setting. The agent service module (MQ1) subscribes to the source node title information rB31/T5-3 and rB32/T6-4 containing the status result information, and forwards the task after receiving the subscription title rB31/T5-3 or rB32/T6-4 information The data module queries whether the information contained in the rB31/T5-3 or rB32/T6-4 header is the latest and correct. If so, execute the task processing sub-program AP3SP1. After completion, the status or result information of the worker node is sent to the worker node status result update module according to the status result setting. After the worker node status result update module updates it to the task data module, it then sends the project message according to the project message agent setting The agent service module (MQ1) publishes the title information rB32/T3-5 containing the status result setting information. The timing standard of the work node T1-2 is set to null, the source information is set to "rB36/T2-1, rB32/T6-4", the status result is set to "rB32/T1-2", and the project message agent is set to "MQ1" ", according to the microservice normalization process, it is determined that the timing standard of the working node T1-2 is set to a null value and the source information is set to a non- Null value, so the microservice main program AP3 will subscribe to the project message broker service module (MQ1) the source node header information rB36/T2-1 and rB32/T6-4 containing the status result information according to the source information setting and the project message broker setting , And after receiving the subscribed title rB36/T2-1 or rB32/T6-4 information, check the task data module to see if the information contained in the rB36/T2-1 or rB32/T6-4 title is the latest and correct Information, if yes, execute the task processing sub-program AP3SP2. After completion, according to the status result setting, the worker node status or result information will be sent to the worker node status result update module, and the worker node status result update module will update it to the task data module Then, according to the project message proxy setting, the title information rB32/T1-2 containing the status result setting information is released to the project message proxy service module (MQ1).

Task 64: Product sales. The microservice retrieves the task specification information of the working node T2-1 bound to the matching microservice settings from the task data module according to its own main program WS1 and task processing subprogram WS1SP1. The timing standard of the work node T2-1 is set to null value, the source information is set to null value, the status result is set to "rB36/T2-1", and the project information agent is set to "MQ1". The timing standard setting and source information setting of node T2-1 are all empty values, so the main microservice program WS1 receives the request from the external system to passively execute the task processing subprogram WS1SP1, and after the execution is completed, the status or result of the working node is set according to the status result. The information is sent to the worker node status result update module, and the worker node status result update module updates it to the task data module, and then publishes the status result setting to the project message agent service module (MQ1) according to the project message service setting Information title information rB36/T2-1.

Task 65: Product sales analysis is bound by the microservice according to its own main program AP4 and task processing subprogram AP4SP1 to retrieve the matching microservice settings from the task data module The task specification information of working node T4-6 is operated. The timing standard of the working node T4-6 is set to null, the source information is set to "rB36/T2-1", the status result is set to "rB32/T4-6", and the project message agent is set to "MQ1", according to the microservice specifications The process determines that because the timing standard of the working node T4-6 is set to a null value and the source information is set to a non-null value, the microservice main program AP4 will send the project message agent service module (MQ1 ) Subscribe to the source node title information rB36/T2-1 that contains the status result information, and after receiving the subscription title rB36/T2-1 information, query the task data module to check whether the information contained in the rB36/T2-1 title is The latest and correct information. If yes, execute the task processing sub-program AP4SP1. After completion, according to the status result setting, the worker node status or result information will be sent to the worker node status result update module, and the worker node status result update module will update it to After the task data module, the title information rB32/T4-6 containing the status result setting information is released to the project message proxy service module (MQ1) according to the project message proxy setting.

Please refer to the example of the microservice subscription publishing title in Figure 4. The two dotted lines in the project message broker service module MQ1 represent the sending source and receiving destination of a title information respectively. It is divided by thickness. If the thickness of the solid line and the dashed line are the same, the microservice is the publishing end of the title information, otherwise, the other end is the subscription receiving end of the title information. The main program of each microservice 51-55 respectively subscribes the source node title to the project message agent service module MQ1 according to the task specification information of the bound working node and publishes the title information containing the status result setting information. By using the one-to-many transmission mechanism of MQTT communication through the project message broker service module MQ1, all subscribers can receive the published title information at the same time. As shown in Figure 4, the main program AP2 of the microservice 51, according to the task specification information of the bound working node T5-3, completes the task processing sub-program AP2SP1 according to the status result. Set to send worker node status or result information to the worker node status result update module, after the worker node status result update module updates it to the task data module, then publish the status result setting to the project message agent service module MQ1 The title information of the information rB31/T5-3, and the microservice 52 main program AP1 and the microservice 53 main program AP3 also subscribe to the source node title information containing status result information according to the bound work node task specification information rB31/T5- 3. After receiving the subscribed title information rB31/T5-3, query the task data module to see if the information contained in the title is the latest and correct information. If so, the microservice 52 main program AP1 and the microservice 53 main program AP3 executes the task processing sub-programs AP1SP1 and AP3SP1 respectively. After completion, the status or result information of the working node is sent to the working node status result update module according to the status result setting, and the working node status result update module updates it to the task data module. After grouping, the title information rB32/T6-4, rB32/T3-5 containing the status result setting information is released to the project message proxy service module MQ1 according to the project message proxy service settings.

」代表標題資訊rB31/T5-3來說，微服務51主程序AP2即為標題發布端、而微服務52主程序AP1及微服務53主程序AP3即為標題訂閱接收端；以粗虛線「

」代表標題資訊rB32/T6-4來說，微服務52主程序AP1即為標題發布端、而微服務53主程序AP3即為標題訂閱接收端；以虛線「

」代表標題資訊rB36/T2-1來說，微服務54主程序WS1即為標題發布端、而微服務53主程序AP3及服務55主程序AP4即為標題訂閱接收端；其他如標題資訊rB32/T4-6及rB32/T3-5、rB32/T1-2因無標題訂閱接收端因此無虛線代表，以此說明。 With a thin dashed line "

"For the title information rB31/T5-3, the main program AP2 of microservice 51 is the title publishing end, while the main program AP1 of microservice 52 and the main program AP3 of microservice 53 are the title subscription receiving end;

"For the title information rB32/T6-4, the main program AP1 of the microservice 52 is the title publishing end, and the main program AP3 of the microservice 53 is the title subscription receiving end; the dotted line "

"For the title information rB36/T2-1, the main program WS1 of the microservice 54 is the title publishing end, and the main program AP3 of the microservice 53 and the main program AP4 of the service 55 are the title subscription receiver; others such as the title information rB32/ T4-6 and rB32/T3-5, rB32/T1-2 have no dashed line because they have no title to subscribe to the receiver, so this is the description.

To sum up, through the task-based work node system and method of the present invention, it especially refers to the division of business logic and service flow, and the disassembly of business logic into low complexity In addition, various tasks with low coupling are bound to high-complexity service processes in the form of work nodes and are processed by microservices. The introduction of the method of the present invention will allow a system to realize the standardization of microservices, provide a standard operation process of microservices to handle simple and simplified internal business operations, and enable it to operate collaboratively in a decentralized design composed of hierarchical task node processes. system. Reduce the complexity of development by narrowing the scope of business logic changes in internal microservices, and set the complexity of service processes externally to optimize overall system performance and improve service scalability.

21‧‧‧Task Setting Module

22‧‧‧Task Data Module

231..23n‧‧‧Microservice

241..24n‧‧‧Project Information Agency Service Module

25‧‧‧Work node status result update module

Claims (7)

A task-based work node system includes: a task setting module for setting the corresponding work stages of each task in the overall service process to generate task specification information of each work node, wherein each task is set by The business logic is disassembled; a task data module for storing task specification information of each working node; at least one microservice, each including a main program and a task processing subprogram, at least one of the at least one microservice Is bound to at least one of the working nodes, and at least one of the at least one microservice obtains the task specification information of at least one of the working nodes to which it is bound from the task data module, according to The task specification information of at least one of the working nodes to which it is bound is operated so as to output the working node status or result information after operation; a working node status result update module is used for the at least one microservice The output of at least one of the worker node status or result information is updated to the task data module; and at least one project message agent service module that supports the message queue structure and is used to provide message publishing or subscription between the at least one microservice The proxy service, wherein at least one of the at least one microservice obtains from the task data module the task specification information of at least one of the working nodes to which it is bound, including timing standard settings and source information settings, for The at least one microservice determines whether to execute its task processing subroutine according to the timing standard setting and the source information setting, so that after completing its task processing subroutine, it outputs the status or result information of the work node for the update model of the status result of the work node. The group updates the output work node status or result information to the task data module, and then publishes the work node title with the work node status or result information to the at least one project information agent service module. A task-based work node system includes: a task setting module for setting the corresponding work stages of each task in the overall service process to generate task specification information of each work node, wherein each task is set by The business logic is disassembled; a task data module for storing task specification information of each working node; at least one microservice, each including a main program and a task processing subprogram, at least one of the at least one microservice Is bound to at least one of the working nodes, and at least one of the at least one microservice obtains the task specification information of at least one of the working nodes to which it is bound from the task data module, according to The task specification information of at least one of the working nodes to which it is bound is operated so as to output the working node status or result information after operation; a working node status result update module is used for the at least one microservice The output of at least one of the worker node status or result information is updated to the task data module; and at least one project message agent service module that supports the message queue structure and is used to provide message publishing or subscription between the at least one microservice In the agency service of, the task setting module specifies the order of each task in the overall service process according to the task level setting principle, and the task level setting principle divides the task level from the highest level to the lowest level into N levels, Wherein, the level of each work node is determined according to the stage of each task in the overall service process. A work node with a higher level is a parent node of a lower level, and a work node with a lower level is a child node of a higher level. The task-based worker node system described in item 2 of the scope of patent application, wherein the task processing subprogram of the at least one microservice is bound to the at least one worker node, and the execution of the task processing subprogram is bound by it The source node of the working node is triggered, and if it is triggered by more If one of the source nodes is triggered, the multiple source nodes are the child nodes of the working node. If only triggered by a single source node, then the single source node is the parent node of the working node or nodes of the same level, and each The source node of a working node can only be set to its parent node or child node. For example, the task-based work node system described in item 2 of the scope of patent application, wherein the task specification information of each work node includes: the microservice settings of each work node, including the main program information and task processing times of each microservice Program information; the project message proxy setting of each working node is the communication proxy server host information required by each microservice to subscribe or publish the relevant title of each working node; the source information setting of each working node is the The source node title information of the working node; the timing standard setting of each working node is the timing processing timing information of each working node; the status result setting of each working node is the working node title with the working node status or result information News. A task-based method for working nodes includes the following steps: disassembling business logic into tasks; setting each task in the corresponding work phase in the overall service process to generate task specification information for each working node; executing microservice specifications The transformation process includes: Procedure 1: At least one of at least one microservice retrieves the task specification information of at least one of the working nodes to which it is bound based on the main procedure and task processing subroutine contained in it; Procedure 2: Make at least one of the at least one microservice check the timing standard setting and source information setting in the task specification information of at least one of the working nodes to which it is bound; Procedure 3: If the timing standard setting is not Is a null value and the source information is set to a null value, then at least one of the at least one microservice executes its own task processing subroutine according to the timing standard setting; if the timing standard is set to a null value and the source information setting If it is not a null value, at least one of the at least one microservice is set to have a source node in the task specification information of at least one of the working nodes to which it is bound according to the source information setting The source node title of the status result information is used to query whether the source node status result information of the source node title of the subscription is the latest and correct after receiving the source node title of the subscription. If it is, it will execute its own task processing If not, it will not be executed; if the timing standard is set to a null value and the source information setting is also a null value, at least one of the at least one microservice will receive a request from an external system and passively execute its own The task processing subroutine; and if the timing standard setting is not a null value and the source information setting is also not a null value, at least one of the at least one microservice does not execute its own task processing subroutine; and At least one of the at least one microservice operates according to the task specification information of at least one of the working nodes to which it is bound. For example, the task-based work node method described in item 5 of the scope of patent application includes: After enabling at least one of the at least one microservice to operate according to the task specification information of at least one of the working nodes to which it is bound, the status or result information of the working node is output for update, and then the at least one microservice At least one of the services publishes the work node title with the work node status or result information. For example, the task-based work node method described in item 5 of the scope of patent application, wherein the timing standard setting of each work node at least includes the execution times per unit period, the number of executions per unit period, the execution start time per unit, and each unit period is Hours, days, months, weeks, and years are a unit period.

Images