CN113709229A

CN113709229A - Data-driven intelligent Internet of things platform workflow implementation system and method

Info

Publication number: CN113709229A
Application number: CN202110974302.9A
Authority: CN
Inventors: 李强; 冯元浩; 王震; 周颢; 郭星; 李向阳
Original assignee: Deqing Alpha Innovation Research Institute
Current assignee: Deqing Alpha Innovation Research Institute
Priority date: 2021-08-24
Filing date: 2021-08-24
Publication date: 2021-11-26

Abstract

A data-driven intelligent Internet of things platform workflow realization system comprises three modules including a function library, a message repeater and a scheduling controller, wherein the function library is used for developing and storing various function functions required by a user, software and hardware are virtualized into data, unified processing is carried out, common program languages are supported, developed templates are provided, uploading and downloading are supported, containerization is carried out, and the mobility is good; the message repeater is used for communication among the unified functions, each function in the function library possibly needs to receive messages of other functions, and the method has the advantages of simple data driving, reusability, easiness in expansion and the like.

Description

Data-driven intelligent Internet of things platform workflow implementation system and method

Technical Field

The invention belongs to the field of intelligent Internet of things systems, and particularly relates to a data-driven intelligent Internet of things platform workflow implementation system and method.

Background

In recent years, the industry of the internet of things is developing vigorously, and the prospect is wide. The related industries include industry, agriculture, animal husbandry, fishery, medical treatment, education and the like, hundreds of millions of devices are connected together to generate massive Internet of things data. In order to manage these devices and the data generated by them conveniently, various internet of things platforms are created, such as a chinese telecom CTWing internet of things platform, a chinese mobile OneNet internet of things platform, an ariloc internet of things platform, a huachen internet of things platform, and the like.

The functionality of these platforms involves two aspects, one being the connection: the platform is connected with various devices (audio and video devices, various sensors, edge gateways and the like) so that the generated data can be transmitted to the platform; the second is application: the platform collects, analyzes and processes the transmitted data, and controls the devices according to requirements to form the workflow of each color. The Internet of things platforms enable various industries and have corresponding intelligent factories, intelligent agriculture, intelligent medical treatment, intelligent education and the like.

The internet of things platform controls the equipment through commands or written functions to complete functions desired by a user, and the functions are called as workflows. Although domestic internet of things platforms have various styles, the internet of things platform has the following common points: the workflow is event driven for orchestrating functions into coordinated microservice applications. Each function in the workflow may be driven by events from various sources. The workflow groups functions and trigger events into a coherent unit and describes the execution of the functions and the information passed in a prescribed manner. A workflow may be viewed as a collection of states and transitions and branches between those states, and each state may have associated events and/or functions. The workflow may be invoked from a CLI command or may be triggered dynamically when an event arrives from an event source. Events from event sources may also be associated with particular states in the function workflow. These states in the workflow will wait for one or more events from one or more event sources to arrive before performing their associated operations and entering the next state.

Event-driven workflow-based workflow has the following disadvantages:

1. poor reusability and automation: control and data coupling reduce the software reuse rate and the code automatic generation rate;

2. poor expanding adaptability: for an event, different conditions can trigger different behaviors, so that branches are generated necessarily, and after a complex event is generated or the scale of the event is increased, the branches become very large and are difficult to maintain and debug;

without a uniform message channel, it is very inconvenient to transfer messages between functions.

Disclosure of Invention

The invention overcomes the defects of the prior art, and provides a data-driven intelligent Internet of things platform workflow realization method and a system thereof based on the problems of the prior platform, and the workflow can be constructed by taking data as a core.

The invention aims to: a data-driven intelligent Internet of things platform workflow realization system comprises three modules including a function library, a message repeater and a scheduling controller, wherein the function library is used for developing and storing various function functions required by a user, software and hardware are virtualized into data, unified processing is carried out, common program languages are supported, developed templates are provided, uploading and downloading are supported, containerization is carried out, and the mobility is good; the message repeater is used for communication among the unified functions, each function in the function library may need to receive messages of other functions and may also transmit messages to other functions, even the two functions occur simultaneously, the component is used for constructing channels for transmitting messages among the functions and is uniformly formatted, therefore, the messages among the functions can be transmitted in a uniform format, the user does not need to care how the message is transmitted, the arrangement controller is used for arranging functions into workflow, the functions completed by each function are limited, a plurality of functions are required to be connected and arranged into the workflow to realize the complex functions required by the user, the component defines an inlet and an outlet for the functions, the two functions can be connected end to realize the transmission of data flow and support multiple inlets and multiple outlets, the function library, the message repeater and the scheduling controller jointly complete the realization of the data-driven intelligent Internet of things platform function workflow.

A use method of a data-driven intelligent Internet of things platform workflow implementation system comprises the following steps: the method comprises the following steps: 1) Searching/writing functions forming the workflow, and for a user, only finding out a function required by the user from a function library of the platform and dragging the function into a working space; for developers, to fulfill the requirements of users, the corresponding functions need to be written by themselves and stored in a function library, in the development process, a function template given by a platform takes data as a core, the functions are constructed through the change of the data in a logic workflow, in a popular way, input, intermediate values and output are received or output in a data form, and the data only care about the change (determined by the requirements) of the data in the processing process and do not depend on event change, so that the number of branches is greatly reduced, namely, complex functions are fulfilled in a linear organization form;

2) the workflow is constructed, the functions are connected by lines, a user connects the functions in the working space by the lines according to certain service logic, namely, the input and the output of data are connected to form a complete workflow, the lines are actually the representation of a message repeater, the data processed by the previous function flows into the next function in a pipeline mode, the design is very common, the message repeater is arranged on the basis of event-driven workflow or data-driven workflow, but the function can be reused on the basis of data driving, as long as the input or the output of the function meets the format requirement, the function can be connected to be used as a ring in the workflow, and as long as the event does not meet the requirement, the function cannot be used for the workflow, and is equivalent to the condition of exclusive use, the whole operation is based on the module of the scheduling controller, which is also the core module of the platform.

The workflow of the invention is connected with three functions, has the functions of collecting, filtering and storing the data of the temperature and humidity sensor into the database, and has the advantages of simple driving, reusability, easy expansion and the like.

Drawings

FIG. 1 is a schematic diagram of three modules of a system and implementation;

FIG. 2 is a schematic diagram of the differences in two platform build workflows;

FIG. 3 is a schematic diagram of an example of a platform's workflow.

Detailed Description

The invention will be described in detail below with reference to the following figures: as shown in fig. 1 to 3, the system for implementing workflow of a data-driven intelligent internet of things platform according to the present invention includes three modules, namely, a function library, a message repeater, and a scheduling controller, wherein the function library is used for developing and storing various function functions required by a user, software and hardware are virtualized into data, processed uniformly, support common programming languages, have developed templates, support uploading and downloading, and are containerized, and have good mobility; the message repeater is used for communication among the unified functions, each function in the function library may need to receive messages of other functions and may also transmit messages to other functions, even the two functions occur simultaneously, the component is used for constructing channels for transmitting messages among the functions and is uniformly formatted, therefore, the messages among the functions can be transmitted in a uniform format, the user does not need to care how the message is transmitted, the arrangement controller is used for arranging functions into workflow, the functions completed by each function are limited, a plurality of functions are required to be connected and arranged into the workflow to realize the complex functions required by the user, the component defines an inlet and an outlet for the functions, the two functions can be connected end to realize the transmission of data flow and support multiple inlets and multiple outlets, the function library, the message repeater and the scheduling controller jointly complete the realization of the data-driven intelligent Internet of things platform function workflow.

Claims

1. a data-driven intelligent Internet of Things platform workflow implementation system, characterized in that the system includes three modules containing a function function library, a message repeater, and an orchestration controller, wherein the function function library is used to develop and store users All required functions, software and hardware are virtualized into data, processed uniformly, support common programming languages, have developed templates, support upload and download, and are containerized, with good portability; message The repeater is used to communicate between functions. Each function in the function library may need to receive messages from other functions, and may also transmit messages to other functions, or even both occur at the same time. This component is used to build functions to pass between The channel of the message is formatted in a unified manner. Therefore, messages can be passed between functions in a unified format, and the user does not need to care about how the message is passed. The orchestration controller is used to orchestrate functions to become workflows, and each function completes The functions are limited. To realize the complex functions required by the user, it is necessary to connect and arrange multiple functions to become a workflow. This component defines the entrance and exit for the functions, and the two functions can be connected end to end to realize the data flow. It supports multi-entry and multi-exit, and the functional library, message repeater, and orchestration controller jointly complete the realization of the data-driven intelligent IoT platform function workflow.

2. The use method of the data-driven intelligent IoT platform workflow implementation system according to claim 1: characterized in that the method comprises the following steps: 1) Find/write the functions that constitute the workflow, for the user , you only need to find the function you need in the function library of the platform and drag it into the workspace; for developers, to meet the needs of users, they need to write the corresponding functions and save them. Enter the function function library. During the development process, the function template given by the platform is based on data, and the function is constructed through the change of data in the logical workflow. Form reception or output, the data only cares about its own changes (determined by requirements) in the processing process and does not depend on event changes, which greatly reduces the number of branches, that is, completes complex functions in a linear organizational form;

2) Build a workflow, connect functions with "lines", and users connect the functions in the workspace with "lines" according to certain business logic, that is, connect the input and output of data to organize a complete work Flow, "line" is actually the performance of the message repeater, allowing the data processed by the previous function to flow into the next function through the form of a pipeline. This is a very common design, whether based on event-driven workflow or based on data-driven work. Streams have this message repeater, but the advantage of being data-driven is that functions can be reused. As long as the input or output of this function meets the format requirements, it can be connected as a part of the workflow, and based on event-driven function, as long as the event does not meet the requirements, it cannot be used in the workflow, which is equivalent to only "special use". The entire operation above is based on the orchestration controller module, which is also the core module of the platform.