An edge algorithm network host networking interaction system and a data real-time interaction method

Abstract

The application provides an edge computing host networking interaction system and a data real-time interaction method, which can uniformly and finely manage the real-time distribution and reception of data, events and commands between various application programs of the edge computing host, and simplify the host application network programming problem. The edge computing host networking interaction system comprises: an edge computing host, wherein the edge computing host is in communication connection with a terminal, the edge computing host is configured to collect data of the terminal, and the edge computing hosts are in data interaction through a data distribution service; a cloud control platform, wherein the cloud control platform is in communication connection with the edge computing host, the cloud control platform and the edge computing host are in data interaction through the data distribution service; and a network middleware, wherein the network middleware is implanted in the edge computing host and the cloud control platform, and the network middleware interacts with a system of the data distribution service.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an edge computing network host networking interaction system and a data real-time interaction method. BACKGROUND

[0002] Under the background of digital transformation, driven by the demand for business cloudization and the development of technology such as the Internet of Things, 5G, and AI, edge computing has realized the sinking of resources and services to the edge location, providing edge intelligent services nearby to meet the key needs of industry digitization in agile connection, real-time business, data optimization, application intelligence, security and privacy protection, etc. The edge computing network host is deployed on the network edge close to the object or data source, providing a distributed open platform that integrates network, computing, storage, application and other core capabilities, and providing edge intelligent services nearby. However, the edge computing network host needs to deploy related computing application programs and interact with other network nodes or associated devices, that is, if the traditional implementation method is used, data communication between each node and each application program is needed, which will increase the cost of application program development and may cause certain security risks to the device. SUMMARY

[0003] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide an edge computing network host networking interaction system and a data real-time interaction method, which can uniformly and finely manage the real-time distribution and reception of data, events and commands between each application program of the edge computing network host, and simplify the host application network programming problem.

[0004] According to one aspect of the present application, an edge computing network host networking interaction system is provided, comprising: an edge computing network host, wherein the edge computing network host is in communication connection with a terminal, the edge computing network host is configured to collect data of the terminal, and the edge computing network hosts are in data interaction through a data distribution service; a cloud control platform, the cloud control platform is in communication connection with the edge computing network host, and the cloud control platform and the edge computing network host are in data interaction through the data distribution service; a network middleware, the network middleware is implanted in the edge computing network host and the cloud control platform, the network middleware interacts with a system of the data distribution service, the network middleware is used to provide an application programming interface to create a publisher and a subscriber of the data distribution service, set a quality of service of data transmission, and process received data; wherein the edge computing network host and the edge computing network host, and the edge computing network host and the cloud control platform realize data interaction of the data distribution service through the application programming interface.

[0005] In an embodiment, the network middleware comprises a data distribution service shared domain, which is arranged between the edge computing network host and the cloud control platform, and comprises a publish-subscribe model, and is configured to isolate data streams of different parts between the edge computing network host and the cloud control platform.

[0006] In an embodiment, the network middleware further comprises a first application programming interface, which is implanted in the edge computing network host, and is used for instant interaction of application data between edge computing network host nodes, and decouples the application layer and the network layer; and a second application programming interface, which is implanted in the edge computing network host, and is used for interaction of application data between edge computing network host nodes, and provides a unified data bus, and decouples network interface calls between applications.

[0007] In an embodiment, the network middleware further comprises a third application programming interface, which is implanted in the edge computing network host and the cloud control platform, and is used for instant interaction of data between edge computing network host nodes, and is configured to have a large number of fine control real-time Qos parameters, control interaction data, and perform flow control to limit the amount of data sent under preset conditions; and a fourth application programming interface, which is implanted in the edge computing network host and the cloud control platform, and is configured to, while the edge computing network host nodes are interacting, synchronously distribute data to the cloud control platform for recording, storage and analysis.

[0008] According to another aspect of the present application, an edge computing network host application data real-time interaction method is provided, which is applied to the edge computing network host networking interaction system of any of the above embodiments, and comprises the following steps: based on the data distribution service specification configured by the network middleware, when the edge computing network host sends data to other edge computing network hosts and the cloud control platform, a publisher in the data distribution service is called to set the behavior and transmission quality of published data; and after the behavior and transmission quality of published data are set, the edge computing network host sends real-time data to other edge computing network hosts and the cloud control platform.

[0009] In an embodiment, the edge computing network host application data real-time interaction method comprises: based on the data distribution service specification configured by the network middleware, when the edge computing network host receives data of other edge computing network hosts and the cloud control platform, a subscriber in the data distribution service is called to set the behavior and receiving quality of the subscribed data; after the behavior and receiving quality of the subscribed data are set, the edge computing network host receives data of other edge computing network hosts and the cloud control platform in real time.

[0010] In an embodiment, the edge computing network host networking interaction system further comprises: a data distribution service shared domain, and the edge computing network host application data real-time interaction method further comprises: obtaining a publish-subscribe communication strategy provided by a publish-subscribe model in the data distribution service shared domain; wherein, based on the data distribution service specification configured by the network middleware, when the edge computing network host sends data to other edge computing network hosts and the cloud control platform, a publisher in the data distribution service is called to set the behavior and transmission quality of the published data, including: based on the publish-subscribe communication strategy, when the edge computing network host sends data to other edge computing network hosts and the cloud control platform, a publisher in the data distribution service is called to set the behavior and transmission quality of the published data.

[0011] In an embodiment, the edge computing network host application data real-time interaction method further comprises: based on the data distribution service specification configured by the network middleware, the edge computing network host subscribes data in the cloud control platform; after the data subscription in the cloud control platform is completed, data storage and data analysis are performed according to a preset strategy.

[0012] In an embodiment, the network middleware further comprises: a first application programming interface and a second application programming interface; the edge computing network host application data real-time interaction method further comprises: based on the first application programming interface, data of each application between edge computing network host nodes is interacted in real time, and the application layer and the network layer are decoupled; based on the second application programming interface, each application program data of the edge computing network host is interacted, and based on a unified data bus of the second application programming interface, the edge computing network host performs network interface call decoupling between each application program.

[0013] In an embodiment, the network middleware further comprises a third application programming interface and a fourth application programming interface; the edge algorithm network host application data real-time interaction method further comprises: based on the third application programming interface, each data of each node of each edge algorithm network host is interacted in real time, and based on a large number of real-time Qos parameters with fine control configured by the third application programming interface, each edge algorithm network host controls each interaction data and traffic, and limits the data amount sent under a preset condition; based on the fourth application programming interface, each node of each edge algorithm network host is interacted, and data is synchronously distributed to a cloud control platform for record storage and analysis processing.

[0014] The edge algorithm network host networking interaction system and the data real-time interaction method provided by the application can effectively reduce the huge workload brought by application development, save manpower investment, capital investment, operation and maintenance, and improve the quality and security of network communication to a certain extent. BRIEF DESCRIPTION OF DRAWINGS

[0015] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which: The accompanying drawings are included to provide a further understanding of the embodiments of the application and constitute a part of the specification. The embodiments of the application are explained together with the application and do not constitute a limitation on the application. In the drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 FIG. 1 is a structural schematic diagram of an edge algorithm network host networking interaction system provided by an exemplary embodiment of the application.

[0017] Figure 2 FIG. 3 is a network topology structure diagram to which the application is applicable.

[0018] Figure 3 FIG. 4 is a configuration structure schematic diagram of an edge algorithm network host network middleware provided by an exemplary embodiment of the application.

[0019] Figure 4 FIG. 5 is a flow schematic diagram of an edge algorithm network host application data real-time interaction method provided by an exemplary embodiment of the application.

[0020] Figure 5 FIG. 6 is a principle schematic diagram of an edge algorithm network host application data real-time interaction provided by an exemplary embodiment of the application. DETAILED DESCRIPTION

[0021] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and are not all the embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.

[0022] Edge Computing is a distributed computing model that places computing resources and data storage at the edge of the network, close to the data source, rather than in remote data centers. Edge computing aims to reduce data transmission delay, improve service response speed, and reduce the burden on central data centers. In the edge computing environment, edge devices or edge nodes play an important role. Edge devices or edge nodes can be various physical devices, such as smartphones, sensors, routers, industrial controllers, etc., which are located at the edge of the network and are responsible for processing data, performing computing tasks and providing services. These devices are usually connected to the edge computing network and implement data processing and service provision through the architecture of edge computing. In the edge computing network, edge devices can play the role of hosts, i.e. devices responsible for performing computing tasks, storing data and providing services in the edge computing network. These edge hosts can undertake different tasks as needed, such as data processing, real-time analysis, content distribution, etc., to provide faster and more reliable services at the edge. Edge hosts play a key role in the edge computing network, helping to improve system response speed, reduce network latency, and provide better user experience. By pushing computing resources and services to the edge of the network, edge hosts can better meet the needs of various application scenarios, including the Internet of Things, smart cities, industrial automation, etc.

[0023] Since edge computing hosts can process application data, i.e. unlike traditional network devices, some network traffic data will not only be forwarded through the protocol stack, but also need to be processed by the application before the processed result data is sent out, or the received data is processed by the application before it is sent to the next device. Therefore, the present method proposes an edge computing host networking interaction system and a data real-time interaction method, which can enable the data information processed by the application in the edge computing host to be sent and received through a unified mechanism. The advantage of this system and method is to reduce the programming and configuration problems of each application in processing network data sending and receiving, facilitate the unified and fine management and control of application data processing, reduce network monitoring ports, reduce network security risks, realize the plug-and-play of edge computing hosts and improve the flexibility of their networking, and provide a standard communication layer that shields a large amount of underlying complexity and simplifies the development of distributed systems.

[0024] The method proposes an edge algorithm network host networking interaction system and a data real-time interaction method based on DDS (Data Distribution Service). DDS is a communication protocol and software architecture for real-time data exchange, which adopts a publish / subscribe architecture, emphasizes data-centric, provides rich QoS service quality policies to guarantee real-time, efficient and flexible data distribution, and can meet various distributed real-time communication application requirements. In DDS, Publisher and Subscriber are two important concepts for describing the data publishing and subscribing process. In DDS, the Publisher is responsible for publishing data, which sends data to the data sharing domain (Domain) in DDS so that the Subscriber can receive and process the data. The Publisher usually contains data producers that generate data and send it to DDS for the Subscriber to subscribe and consume. In DDS, the Publisher and the Subscriber communicate through the DDS middleware to realize the publishing and subscribing of real-time data. By calling the Publisher, the data publishing function can be realized, so that the data can be obtained and processed by the Subscriber in time, thereby realizing the function of real-time data exchange in the distributed system. By calling the Subscriber, the application program can realize the subscription and reception of real-time data, thereby realizing the sharing and transmission of data in the real-time system. The Subscriber and the Publisher jointly constitute the publish-subscribe model of DDS, providing an efficient data communication and sharing mechanism for real-time systems.

[0025] Figure 1 is a structural schematic diagram of an edge algorithm network host networking interaction system provided by an exemplary embodiment of the present application, like Figure 1As shown, according to one aspect of the present application, an edge computing host networking interaction system 100 is provided, which comprises: an edge computing host 10, wherein the edge computing host 10 is in communication connection with a terminal, the edge computing host 10 is configured to collect data of the terminal, and the edge computing hosts 10 interact with each other through a data distribution service; a cloud control platform 30, which is in communication connection with the edge computing host 10, and the cloud control platform 30 and the edge computing host 10 interact with each other through the data distribution service; a network middleware 20, which is implanted in the edge computing host 10 and the cloud control platform 30, the network middleware 20 interacts with the system of the data distribution service, the network middleware 20 is used to provide an application programming interface to create a publisher and a subscriber of the data distribution service, set a quality of service of data transmission and process received data; wherein the edge computing host 10 and the edge computing host 10, and the edge computing host 10 and the cloud control platform 30 realize data interaction of the data distribution service through the application programming interface.

[0026] Figure 2 The network topology structure diagram to which the present application is applicable is shown in Figure 1. Figure 2 As shown, the edge computing host can comprise a first edge computing host 11 and a second edge computing host 12, the first edge computing host 11 and the second edge computing host 12 can collect data of a terminal 40 or data of a sensor 50, the first edge computing host 11 and the second edge computing host 12 can interact with each other, the first edge computing host 11 and the second edge computing host 12 are any two edge computing hosts in a networking topology, the application program 1 and the application program n in the first edge computing host 11 and the second edge computing host 12 can interact with each other, the application program 1 and the application program n are any two application programs in the edge computing host, and the first edge computing host 11 and the second edge computing host 12 can interact with the cloud control platform 30. The network middleware is a package of DDS (Data Distribution Service), which realizes data interaction of application programs in the edge computing host through DDS, can effectively reduce a huge workload brought by application program development, and improve the quality of network communication. When using DDS, a developer usually uses a specific application programming interface (API) to interact with the DDS system to publish data, subscribe data and control data transmission behavior, and the edge computing host and the cloud control platform access DDS through the API to make DDS play a role therein.

[0027] In an embodiment, the network middleware comprises a data distribution service shared domain, the data distribution service shared domain is arranged between the edge computing network host and the cloud control platform, the data distribution service shared domain comprises a publish-subscribe model, and the data distribution service shared domain is configured to isolate data streams of different parts between the edge computing network host and the cloud control platform.

[0028] In a DDS (Data Distribution Service), a DDSDomain is used to isolate different DDS instances and data streams. A DDSDomain can be understood as a logical container that contains a group of Publishers and Subscribers, which share the same communication rules and QoS (Quality of Service) settings. In a DDS system, multiple different Domains can be defined, each with a unique identifier (Domain ID). By assigning different entities (such as Publishers and Subscribers) to different Domains, data isolation and management can be achieved. The concept of DDSDomain helps to organize and manage complex real-time systems, allowing data streams between different parts to run and interact independently while maintaining data security and reliability. By configuring different QoS settings in different Domains, data transmission can be customized according to system requirements to meet the specific requirements of real-time systems. The Subscriber and Publisher together form the Publish-Subscribe model of DDS, which is a common communication pattern used to decouple and communicate between components in a distributed system, providing an efficient data communication and sharing mechanism for real-time systems. In the Publish-Subscribe model, the Publisher and Subscriber are decoupled and do not communicate directly, but through a central component to deliver messages.

[0029] In an embodiment, the network middleware further comprises a first application programming interface implanted in the edge computing network host, the first application programming interface is used for real-time interaction of data of each application between each edge computing network host node, and decoupling of the application layer and the network layer; and a second application programming interface implanted in the edge computing network host, the second application programming interface is used for interaction between data of each application program of the edge computing network host, and provides a unified data bus to decouple network interface calls between each application program.

[0030] In the edge computing network host or cloud control platform, a network middleware is implanted, a first application programming interface (API) is provided, and data of each application between each host node is interacted in real time, the development and deployment difficulty of the computing power application is simplified, and the application layer and the network layer are decoupled. In the edge computing network host or cloud control platform, a network middleware is implanted, a second application programming interface is provided, the second application programming interface is used for interaction between each application program data of the host, the development and deployment difficulty of the computing power application is simplified, a unified data bus is provided, and the network interface calling between each application program is decoupled.

[0031] In an embodiment, the network middleware further comprises: a third application programming interface implanted in the edge computing network host and the cloud control platform, the third application programming interface is used for real-time interaction of each data of each node of each edge computing network host, the third application programming interface is configured to have a large number of fine control real-time Qos parameters, control each interaction data, and perform flow control to limit the data amount sent under a preset condition; and a fourth application programming interface implanted in the edge computing network host and the cloud control platform, the fourth application programming interface is configured to synchronize and distribute data to the cloud control platform for recording, storage and analysis processing while the nodes of each edge computing network host are interacting.

[0032] In the edge computing network host or cloud control platform, a network middleware is implanted, a third application programming interface is provided, and each data of each host node is interacted in real time, which has a large number of fine control real-time Qos parameters, fine control of each interaction data, flow control, and limitation of the data amount sent under a specific condition, so as to improve the quality of network communication. In the edge computing network host or cloud control platform, a network middleware is implanted, a fourth application programming interface is provided, and data can be synchronized and distributed to the cloud control platform for recording, storage and analysis processing while the nodes of each host are interacting.

[0033] Therefore, through each API provided by the network middleware, real-time data interaction is realized, the port of each application network communication monitoring is reduced, the security risk of device network communication is reduced, the related problems of point-to-point system integration and network node expansion (realizing plug and play of the edge intelligent gateway node, and not needing configuration when joining or leaving the network) are simplified, and the interoperability and architecture evolution ability are improved. Moreover, configurable secure communication is realized, including identity authentication of a remote participant, access control of an entity and data encryption, and the like, and high portability is realized, the DDS specification and IDL definition are followed, and an application program can be switched between different DDS implementations only by recompilation.

[0034] Figure 3 FIG. 1 is a configuration structure schematic diagram of an edge computing network host network middleware provided by an exemplary embodiment of the present application,Figure 3 As shown, the edge computing host can deploy related computing power application layers, and the computing power application layer is a user application program that uses Fast DDS API to realize communication in a distributed system, namely an Application module. The computing power application layer of the edge computing host interacts with the network middleware through an API (application programming interface) to exchange data.

[0035] Network middleware: contains DDS layer and RTPS layer, which is the implementation of DDS communication middleware. It allows the deployment of one or more DDS domains, where domain participants within the same domain exchange messages by publishing / subscribing under domain topics. The RTPS layer implements the Real-Time Publish-Subscribe (RTPS) protocol to support interoperability with DDS applications, and this layer serves as an abstraction layer for the transport layer. API (application programming interface) defines the way of interaction and communication protocol between software components. DDS Domain can be understood as a logical container that contains a group of publishers and subscribers, which share the same communication rules and QoS (Quality of Service) settings. DDS Participant is one of the core concepts of DDS, representing a participant or entity in the DDS system. DDS Participant is the top-level entity in the DDS system, representing a participant in the entire DDS domain. DDS Participant usually contains the following important elements: 1. DomainID (domain identification): each DDS Participant belongs to a specific domain, and DomainID is used to distinguish different DDS domains. 2. QoS (Quality of Service): DDS Participant can configure various quality of service parameters to control data transmission and interaction behavior, such as reliability, delay, bandwidth, etc. 3. Topic (topic): DDS Participant can create publishers and subscribers to publish and subscribe to specific topics, realizing data publishing and subscribing. Through DDS Participant, applications can establish connections with the DDS system, configure parameters of DDS instances, and create publishers and subscribers to realize data publishing and subscribing. DDS Participant is an important component in the DDS system, providing an entry point and context environment for applications to interact with the DDS system. In DDS, a DDS Participant instance usually represents an application instance that communicates with the DDS system and participates in data publishing and subscribing. DDS Participant is responsible for managing connections with the DDS system, configuring parameters of DDS instances, and providing a context environment for applications to interact with the DDS system.

[0036] DataReader is an entity for subscribing to data, and it is a component in DDS for subscribing to data. It is responsible for receiving data from the data sharing domain in DDS and providing data to the application for processing and consumption. DataWriter is a component in DDS for publishing data. It is responsible for sending data to the data sharing domain in DDS so that subscribers can receive and process these data.

[0037] DDSWriterHistory: DDSWriterHistory is used to track historical data sent by data writers (Writers). When a writer publishes data, these data will be stored in DDSWriterHistory so that subscribers can access previously published data. Through DDSWriterHistory, subscribers can get historical data, not just the latest data. DDSReaderHistory: DDSReaderHistory is used to track historical data received by data readers (Readers). When a reader subscribes to data, it can choose to store a certain amount of historical data in DDSReaderHistory so that it can access previously received data when needed. Through DDSReaderHistory, subscribers can get previously received data, not just the latest data. The management of these historical records is very important for the processing and distribution of real-time data. By using DDSWriterHistory and DDSReaderHistory, subscribers can access previous data for backtracking, analysis or reprocessing when needed. This function is very useful for many application scenarios, such as monitoring systems, log recording, data analysis, etc.

[0038] DDSI-RTPS: RTPS (Real Time Publish Subscribe Protocol) protocol realizes the maximum effort reliable publish-subscribe through TCP / UDP, etc. This specification defines the interoperability protocol of DDS, whose purpose and scope is to ensure that applications based on different vendor's DDS implementations can interoperate.

[0039] In DDS (Data Distribution Service), RTPS (Real-Time Publish-Subscribe) is a protocol used to implement real-time data publishing and subscribing. RTPS defines the way data is transmitted and the interaction rules in the network. In RTPS, there are two key concepts: RTPSDomain and RTPSParticipant. RTPSDomain: RTPSDomain represents a logical data communication domain, which is a collection of participants (RTPSParticipant). In an RTPSDomain, different RTPSParticipant can communicate and exchange data with each other. RTPSDomain provides a logically isolated unit to help organize and manage communication between participants. RTPSParticipant: RTPSParticipant is an entity participating in RTPS communication, representing an actual communication node or process. Each RTPSParticipant has a unique identifier to uniquely identify the participant in the RTPSDomain. RTPSParticipant is responsible for communicating with other participants, publishing data, and subscribing to data. Through the concepts of RTPSDomain and RTPSParticipant, the RTPS protocol realizes real-time data communication and exchange in distributed systems. RTPSDomain provides logical isolation to help manage and organize communication relationships between participants, while RTPSParticipant represents the actual entity participating in communication. In practical applications, developers can create and manage RTPSDomain and RTPSParticipant to build real-time data communication systems, implement data publishing and subscribing, and realize real-time data exchange and processing.

[0040] RTPSWriter: RTPSWriter is a data publisher, responsible for publishing data into the network for other subscribers to subscribe. When a data publisher publishes data using the RTPS protocol, it creates an RTPSWriter instance and sends data to the network through this instance. RTPSWriter is responsible for transmitting data to the network and establishing data communication with subscribers. RTPSReader: RTPSReader is a data subscriber, responsible for subscribing and receiving data published in the network. When a data subscriber subscribes to data using the RTPS protocol, it creates an RTPSReader instance and receives data sent from the network through this instance. RTPSReader is responsible for receiving and processing data from RTPSWriter and providing data to subscribers for processing. Through the pairing of RTPSWriter and RTPSReader, the data publishing and subscription mechanism is realized. RTPSWriter is responsible for publishing data, while RTPSReader is responsible for subscribing data, thereby realizing real-time transmission and exchange of data. This publish-subscribe mode allows different system components to independently publish and subscribe to data, achieving decoupling and flexibility between systems. In practical applications, developers can use RTPSWriter and RTPSReader to implement real-time data publishing and subscription, build real-time data communication systems, and realize real-time data transmission and processing.

[0041] Network transport layer: Fast DDS can be used with various transport protocols, such as UDPv4, UDPv6, TCPv4, TCPv6, or SHM (Shared Memory Transport Protocol). The RTPSWriter of the network middleware sends data to other device applications through TCP / UDP, and the application data of other devices is transmitted to the RTPSReader through TCP / UDP. IP (Internet Protocol) is an important protocol in the network transport layer, responsible for transmitting data packets in the network. The IP protocol defines the routing and delivery rules of data packets in the network, ensuring that data can be transmitted from the source host to the target host.

[0042] Figure 4 is the flowchart of the edge algorithm network host application data real-time interaction method provided by an exemplary embodiment of the present application, as shown in Figure 4 The edge algorithm network host application data real-time interaction method is applied to the edge algorithm network host networking interaction system of any one of the above embodiments, and the edge algorithm network host application data real-time interaction method comprises:

[0043] Step 100: Based on the data distribution service specification configured by the network middleware, when the edge computing host sends data to other edge computing hosts and the cloud control platform, it calls the publisher in the data distribution service to set the behavior and transmission quality of publishing data.

[0044] The DDS specification (Data Distribution Service Specification) describes a data-centric publish-subscribe (DCPS) model for distributed application communication and integration. This specification defines application programming interfaces (APIs) and communication semantics (behavior and quality of service) that enable efficient communication of information from information producers to matching consumers. The edge computing host or cloud control platform performs data processing and sending, calls the Publisher in DDS, and performs DataWriter and Qos control. In DDS, the Publisher is responsible for publishing data and sending it to the data sharing domain in DDS so that subscribers can receive and process the data. The Publisher usually contains data producers that generate and send data to DDS for subscribers to subscribe and consume. Calling the Publisher in DDS means using code or APIs in the DDS system to create a Publisher object and using the object to publish data. By calling the Publisher, you can specify the data content, data type, publishing frequency, and other information to be published so that other subscribers can subscribe and receive the data. In the data distribution service (DDS), DataWriter is an entity for publishing data, and QoS (Quality of Service) control is used to configure data publishing and transmission parameters. DataWriter and QoS control means customizing the behavior and transmission quality of publishing data. DataWriter is a component in DDS for publishing data. It is responsible for sending data to the data sharing domain in DDS so that subscribers can receive and process the data. Through DataWriter, you can control the content, frequency, and transmission method of publishing data to meet system requirements. QoS control refers to configuring a series of parameters to control the quality and behavior of data publishing and transmission, including but not limited to data transmission reliability, delay, bandwidth, persistence, etc. Through QoS control, you can customize the behavior of data transmission to meet system requirements for data transmission quality and performance.

[0045] Step 200: After setting the behavior and transmission quality of publishing data, the edge computing host sends real-time data to other edge computing hosts and the cloud control platform.

[0046] After the behavior of publishing data and the setting of transmission quality are completed, data interaction can be performed between the edge computing host 1 and the edge computing host 2 based on the DDS, the edge computing host 1 and the edge computing host 2 being any two edge computing hosts in the networking topology, data interaction can be performed between the application program 1 and the application program n in the edge computing host, the application program 1 and the application program n being any two application programs in the edge computing host, and the edge computing host 1 and the edge computing host 2 can perform data interaction with the cloud control platform.

[0047] In an embodiment, the edge computing host application data real-time interaction method comprises: based on the data distribution service specification configured by the network middleware, when the edge computing host receives data of other edge computing hosts and the cloud control platform, a subscriber in the data distribution service is invoked, and the behavior of subscribing data and the receiving quality are set; after the behavior of subscribing data and the receiving quality are completed, the edge computing host receives data of other edge computing hosts and the cloud control platform in real time.

[0048] The edge computing host or cloud control platform receives and processes data, calls the Subscriber in DDS, and controls the DataReader and QoS. In DDS (Data Distribution Service), calling the Subscriber refers to creating a Subscriber entity in the application program to receive data of a specific topic. Subscriber is an important concept in DDS, used to subscribe to data published by Publisher and receive data updates from data streams. When an application program needs to receive data of a specific topic, it creates a Subscriber object and subscribes to the topic of interest through the Subscriber object. Once the subscription is successful, the Subscriber will start receiving data related to the topic and pass it to the application program for processing. By calling the Subscriber, the application program can achieve real-time data subscription and reception, thereby realizing data sharing and transmission in real-time systems. In addition, in the data distribution service (DDS), DataReader is an entity for subscribing to data, and QoS (Quality of Service) control is used to configure data subscription and reception parameters. Controlling the DataReader and QoS means customizing the behavior and quality of receiving data. DataReader is a component in DDS for subscribing to data. It is responsible for receiving data from the data sharing domain in DDS and providing data to the application program for processing and consumption. DataReader is usually associated with a specific data type and is responsible for subscribing to data of that type and passing it to the application program for processing. Therefore, through DataReader, the behavior of subscribing to data can be controlled, such as the type of data subscribed to, the frequency of subscription, etc. QoS control is a series of parameters configured to control the quality and behavior of data subscription and reception. These parameters include but are not limited to data transmission reliability, delay, bandwidth, persistence, etc. Through QoS control, the behavior of data reception can be customized to meet the system's requirements for data reception quality and performance. After completing the customization of the behavior of data subscription and the quality of transmission, based on DDS, edge computing host 1 and edge computing host 2 can interact with each other, and edge computing host 1 and edge computing host 2 are any two edge computing hosts in the networking topology. The application program 1 and the application program n in the edge computing host can interact with each other, and the application program 1 and the application program n are any two application programs in the edge computing host. Edge computing host 1 and edge computing host 2 can interact with the cloud control platform.

[0049] Figure 5is a schematic diagram of the principle of edge algorithm network host application data real-time interaction provided by an exemplary embodiment of the present application, as shown in Figure 5 The first edge algorithm network host 11 and the cloud control platform 30 can both send data through the Publisher in the DDS Domain, and the Publisher is responsible for publishing data, which sends data to the data sharing domain (Domain) in the DDS, so that the subscribers can receive and process the data. The second edge algorithm network host 12 and the cloud control platform 30 can both subscribe to the data published by the Publisher in the DDS Domain, and receive data updates from the data stream, such as receiving data of a specific topic (Topic). In addition, the first edge algorithm network host 11 and the second edge algorithm network host 12 can interact with each other through the DDS Domain, and the first edge algorithm network host 11, the second edge algorithm network host 12 and the cloud control platform 30 can also interact with each other through the DDS Domain, and the first edge algorithm network host 11 and the second edge algorithm network host 12 are any two edge algorithm network hosts in the networking.

[0050] In an embodiment, the edge algorithm network host networking interaction system further comprises a data distribution service sharing domain, and the edge algorithm network host application data real-time interaction method further comprises: obtaining the publish-subscribe communication strategy provided by the publish-subscribe model in the data distribution service sharing domain; wherein, based on the data distribution service specification configured by the network middleware, when the edge algorithm network host sends data to other edge algorithm network hosts and the cloud control platform, the publisher in the data distribution service is called to set the behavior and transmission quality of the published data, including: based on the publish-subscribe communication strategy, when the edge algorithm network host sends data to other edge algorithm network hosts and the cloud control platform, the publisher in the data distribution service is called to set the behavior and transmission quality of the published data.

[0051] The data distribution service shared domain provides a configurable real-time and reliable publish-subscribe communication strategy, supporting synchronous and asynchronous data publishing modes. The built-in dynamic discovery mechanism for publishers and subscribers supports plug-and-play connection, and joining or leaving the network without configuration. Configurable throughput control is supported, which can limit the amount of data sent under certain conditions; UDPv4 / UDPv6 / TCPv4 / TCPv6 / SHM (shared memory) transmission is supported; configurable secure communication is provided, including remote participant authentication, entity access control, and data encryption; pre-allocated resources are allowed, reducing dynamic resource allocation, avoiding unlimited use of resources, and minimizing the need for data replication. The publish-subscribe model in the data distribution service shared domain is a common communication pattern used to decouple and communicate between components in a distributed system. In this model, components in the system are divided into two categories: publishers and subscribers. Publishers are responsible for publishing messages or data, and they send messages to a central component (such as a message broker or message bus) without needing to know directly which components will receive these messages. Subscribers express interest in specific types of messages or data and subscribe to these messages from the central component. Once new messages are published, subscribers will receive them. In the publish-subscribe model, publishers and subscribers are decoupled and do not communicate directly, but use a central component to deliver messages. Therefore, using the publish / subscribe architecture for point-to-point, distributed communication, the data interaction between the edge computing network host nodes and the cloud control platform is realized, and the real-time, reliability, bandwidth control, security, and stability of DDS are used to simplify the data communication of edge computing network host application programs, provide a data bus for application program communication, reduce the coupling degree of the application layer and the network layer, and perform unified management of host application program data transmission, thereby reducing interaction latency, reducing network load, enriching business types, and optimizing processing modes.

[0052] In an embodiment, the edge computing network host application data real-time interaction method further includes: based on the data distribution service specification configured by the network middleware, the edge computing network host performs data subscription in the cloud control platform; after completing the data subscription in the cloud control platform, performing data storage and data analysis according to a preset strategy.

[0053] In addition to program data interaction between edge computing network hosts, data subscription can also be performed in the cloud control platform, and data storage, analysis, and other processing can be performed according to the strategy.

[0054] In an embodiment, the network middleware further comprises: a first application programming interface and a second application programming interface; the edge algorithm network host application data real-time interaction method further comprises: based on the first application programming interface, the data of each application between each edge algorithm network host node is interacted in real time, and the decoupling of the application layer and the network layer is performed; based on the second application programming interface, the data of each application program of the edge algorithm network host is interacted, and based on the unified data bus of the second application programming interface, the edge algorithm network host performs the decoupling of the network interface call between each application program.

[0055] A network middleware is implanted in the edge algorithm network host or the cloud control platform, a first application programming interface (API) is provided, and the data of each application between each host node is interacted in real time, so that the development and deployment difficulty of the computing power application program is simplified, and the decoupling of the application layer and the network layer is performed. A network middleware is implanted in the edge algorithm network host or the cloud control platform, a second application programming interface is provided, the second application programming interface is used for the interaction between the data of each application program of the host, the development and deployment difficulty of the computing power application program is simplified, a unified data bus is provided, and the decoupling of the network interface call between each application program is performed.

[0056] In an embodiment, the network middleware further comprises: a third application programming interface and a fourth application programming interface; the edge algorithm network host application data real-time interaction method further comprises: based on the third application programming interface, each data of each edge algorithm network host node is interacted in real time, and based on the third application programming interface configured with a large number of real-time Qos parameters that can be finely controlled, each edge algorithm network host controls each interaction data and flow, and limits the amount of data sent under a preset condition; based on the fourth application programming interface, while each edge algorithm network host node is interacted, data is synchronously distributed to the cloud control platform for record storage and analysis processing.

[0057] A network middleware is implanted in the edge algorithm network host or the cloud control platform, a third application programming interface is provided, and each data of each host node is interacted in real time, which has a large number of real-time Qos parameters that can be finely controlled, finely controls each interaction data, controls flow, and can limit the amount of data sent under a specific condition, thereby improving the quality of network communication. A network middleware is implanted in the edge algorithm network host or the cloud control platform, a fourth application programming interface is provided, and while each host node is interacted, data can be synchronously distributed to the cloud control platform for record storage and analysis processing.

[0058] Therefore, through the APIs provided by the network middleware, instant data interaction is realized, the ports for network communication monitoring of each application program are reduced, the security risks of device network communication are reduced, the related problems of point-to-point system integration and the expansion of network nodes (realizing plug-and-play of edge intelligent gateway nodes, without configuration when joining or leaving the network) are simplified, the interoperability and architecture evolution capability are improved. Moreover, configurable secure communication is realized, including identity authentication of remote participants, access control of entities and data encryption, etc., with high portability, compliance with DDS specification, IDL definition, etc., and the application program can be switched between different DDS implementations only by recompilation.

[0059] The embodiment of the present application provides a kind of edge algorithm network host networking interaction system. System embodiment can be realized by software, it can also be realized by hardware or software and hardware combined mode.From the hardware layer, in addition to CPU, memory, network interface and non-volatile memory, the device of the system in embodiment can generally include other hardware, such as the forwarding chip responsible for processing message etc.Example for software realization, as a logical system, it is formed by CPU of its device reading corresponding computer program instruction in non-volatile memory to memory and running.

[0060] According to another aspect of the present application, a computer-readable storage medium is provided, which stores a computer program for executing the edge algorithm network host application data real-time interaction method of any of the above embodiments.

[0061] In addition to the above method and device, the embodiment of the present application can also be a computer program product, which includes computer program instructions that, when executed by a processor, cause the processor to perform the steps of the edge algorithm network host application data real-time interaction method according to various embodiments of the present application described in the above "Exemplary Methods" section of this specification.

[0062] According to another aspect of the present application, an electronic device is provided, which includes: a processor; a memory for storing processor-executable instructions; and the processor for executing the edge algorithm network host application data real-time interaction method of any of the above embodiments.

[0063] In addition, the embodiment of the present application can also be a computer-readable storage medium, which stores computer program instructions, and the computer program instructions, when executed by a processor, cause the processor to perform the steps of the edge algorithm network host application data real-time interaction method according to various embodiments of the present application described in the above "Exemplary Methods" section of this specification.

[0064] The above merely provides the preferred embodiment of the present application, and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. An edge computing host networking interaction system, characterized in that, The edge computing host is in communication connection with the terminal, and is configured to collect data of the terminal, and the edge computing hosts are in data interaction through a data distribution service; The cloud control platform is in communication connection with the edge computing hosts, and is in data interaction with the edge computing hosts through the data distribution service; The network middleware is implanted in the edge computing hosts and the cloud control platform, interacts with a system of the data distribution service, and is used for providing an application programming interface to create publishers and subscribers of the data distribution service, set quality of service of data transmission, and process received data; The edge computing hosts and the cloud control platform are in data interaction of the data distribution service through the application programming interface; The network middleware further comprises: A first application programming interface is implanted in the edge computing host, and is used for instant interaction of data of each application between edge computing host nodes, and decoupling of an application layer and a network layer; A second application programming interface is implanted in the edge computing host, and is used for interaction between data of each application program of the edge computing host, and provides a unified data bus to decouple network interface calls between each application program. The network middleware comprises:

2. The edge computing host networking interactive system according to claim 1, wherein, A data distribution service shared domain is arranged between the edge computing hosts and the cloud control platform, and comprises a publish-subscribe model, and is configured to isolate data streams of different parts between the edge computing hosts and the cloud control platform. The network middleware further comprises:

3. The edge computing host networking interactive system of claim 1, wherein, A third application programming interface is implanted in the edge computing host and the cloud control platform, and is used for instant interaction of data of each node of each edge computing host, has a fine-controllable real-time Qos parameter, controls each interaction data, and performs flow control to limit a data amount sent under a preset condition; A fourth application programming interface is implanted in the edge computing host and the cloud control platform, and is configured to, while each node of each edge computing host is interacting, synchronously distribute data to the cloud control platform for record storage and analysis processing. The edge computing host networking interaction system and the edge computing host application data real-time interaction method are applied to any one of the edge computing host networking interaction systems in claims 1-3.

4. An edge computing host application data real-time interaction method, characterized in that, ​ Based on the network middleware configuration of the data distribution service specification, when the edge computing host sends data to other edge computing hosts and the cloud control platform, the publisher in the data distribution service is called to set the behavior and transmission quality of publishing data; After setting the behavior and transmission quality of publishing data, the edge computing host sends real-time data to other edge computing hosts and the cloud control platform.

5. The edge computing host application data real-time interaction method according to claim 4, characterized in that, The edge computing host application data real-time interaction method comprises: Based on the network middleware configuration of the data distribution service specification, when the edge computing host receives data from other edge computing hosts and the cloud control platform, the subscriber in the data distribution service is called to set the behavior and reception quality of subscribing data; After setting the behavior and reception quality of subscribing data, the edge computing host receives real-time data from other edge computing hosts and the cloud control platform.

6. The edge computing host application data real-time interaction method according to claim 4, characterized in that, The edge computing host networking interaction system further comprises a data distribution service shared domain, and the edge computing host application data real-time interaction method further comprises: Obtaining the publish-subscribe communication strategy provided by the publish-subscribe model in the data distribution service shared domain; Based on the network middleware configuration of the data distribution service specification, when the edge computing host sends data to other edge computing hosts and the cloud control platform, the publisher in the data distribution service is called to set the behavior and transmission quality of publishing data, comprising: Based on the publish-subscribe communication strategy, when the edge computing host sends data to other edge computing hosts and the cloud control platform, the publisher in the data distribution service is called to set the behavior and transmission quality of publishing data.

7. The edge computing host application data real-time interaction method according to claim 4, characterized in that, The edge computing host application data real-time interaction method further comprises: Based on the network middleware configuration of the data distribution service specification, the edge computing host subscribes data in the cloud control platform; After subscribing data in the cloud control platform, the edge computing host stores and analyzes data according to a preset strategy.

8. The edge computing host application data real-time interaction method according to claim 4, characterized in that, The network middleware further comprises a first application programming interface and a second application programming interface; the edge computing host application data real-time interaction method further comprises: Based on the first application programming interface, the data of each application between the edge computing host nodes is interacted in real time, and the application layer and the network layer are decoupled; Based on the second application programming interface, the application data of the edge computing host is interacted, and based on the unified data bus of the second application programming interface, the edge computing host is decoupled in network interface calling between each application.

9. The edge computing host application data real-time interaction method according to claim 4, characterized in that, The network middleware further comprises a third application programming interface and a fourth application programming interface; the edge computing host application data real-time interaction method further comprises: Based on the third application programming interface, each data of each node of the edge computing host is interacted in real time, and based on the real-time Qos parameters with fine control configured by the third application programming interface, each edge computing host controls each interaction data and flow, and limits the data amount sent under a preset condition; Based on the fourth application programming interface, while each node of the edge computing host is interacted, data is synchronously distributed to the cloud control platform for record storage and analysis processing.

Images