A method for achieving intelligent intent setting in heterogeneous networks based on a network programming language.

Abstract

To provide a method of realizing intentional intelligent setting in a heterogeneous network on the basis of a network programming language.SOLUTION: A method includes the steps of: generating a network policy adapted to Frenetic network programming language specifications by combining a network resource and a state by a service orchestrator on the basis of an intentional keyword; selecting a corresponding domain controller in accordance with a service requirement by a global network controller and distributing the network policy; and compiling the network policy as an OpenFlow flow table by a domain controller that supports the OpenFlow protocol and converting the network policy into a YANG data model by a policy translator via a domain controller that supports NETCONF protocol.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention belongs to the technical field of communication networks, and specifically relates to a method for realizing intelligent setting of user intentions in a heterogeneous network environment based on a network programming language.

Background Art

[0002] With the rapid development of information technology, the network has become one of the important infrastructures of modern society. Currently, the network is not only a tool for connecting people and people, people and information, but also an important driving force for promoting industrial innovation and improving work efficiency. With the rapid expansion and diversification of the network scale, the complexity and heterogeneity of the network structure are increasing day by day, and the requirements of users for network services are also diversified and dynamic.

[0003] Thanks to the proposal of new network technologies such as Software Defined Networking (SDN) and ForCES (Forwarding and Control Element Separation), the conventional network architecture is more flexible and stable than the traditional network architecture. However, the maintenance and management of network lower-level devices still highly depend on the manual settings of network developers. With the expansion of the network scale, the manual setting of the network becomes difficult, inefficient, and the stability and accuracy of the entire network are restricted by the risks caused by inevitable human errors, which also leads to slow service provision and response speed.

[0004] The OpenFlow and NETCONF protocols are the two most common controller-device communication protocols in current network architectures. While both protocols offer powerful tools for automated and centralized network management, they focus on different aspects and strengths. For example, the OpenFlow protocol allows controllers to directly control the flow tables of network devices, enabling flexible control and management of network traffic, making it more suitable for specific scenarios such as data centers or research environments. The NETCONF protocol, on the other hand, allows network administrators to securely access various functions for remotely configuring network devices, supporting features such as configuration verification and transaction management, resulting in superior performance in traditional enterprise networks. Heterogeneous networks built by combining the OpenFlow and NETCONF protocols fully leverage the strengths of each protocol, providing flexible network management and control capabilities to meet diverse service requirements. However, the deployment and maintenance of heterogeneous networks are relatively complex, leading to demands for integrated management and intelligent organization of services in heterogeneous network environments.

[0005] With the rapid pace of digital transformation, user requirements for network services are becoming more diverse and dynamic. Traditional network configuration and management methods struggle to capture and understand user service requirements, and are unable to customize network services to meet the diverse needs of individual network users. Intent-based networking (IBN) is an emerging technological concept as a brand-new network management and control pattern, aiming to apply a higher level of intelligence and anticipated states instead of the manual process of configuring the network. By using artificial intelligence and machine learning methods, the network software can intelligently determine how to achieve the desired outcome or service objectives, simply by defining the desired results. By introducing the concept of "intent" into traditional network architectures, such high-level service objectives can be translated into specific settings for lower-level network element devices, significantly improving the flexibility and adaptability of traditional network architectures and better meeting the diverse service requirements of users.

[0006] Network programming languages ​​(DSLs) are domain-specific languages ​​dedicated to network programming and are used to configure and manage network devices. Frenetic is one of the most mature representatives in this field, enabling the description of network behavior and policies in a manner approaching natural language. By leveraging Frenetic's characteristics of flexibly describing service capabilities and describing network policies with Python code, and combining this with the code generation capabilities of a Large Language Model (LLM), the programming and configuration process for network devices can be significantly simplified and accelerated. However, until now, the Frenetic network programming language has only provided support for OpenFlow network devices, which means that the Frenetic network programming language is not very effective as an efficient and flexible network management tool in heterogeneous network environments.

[0007] In summary, while conventional network architectures offer a dramatic improvement in scale and performance compared to traditional network architectures, they still face challenges such as difficulty in deploying and maintaining heterogeneous networks, a low degree of intelligence in their organization, and an inability to meet users' service requirements in a timely manner. [Overview of the Initiative]

[0008] To address the challenges inherent in conventional network architectures, this invention provides a method for setting intentions in heterogeneous networks based on a network programming language. By combining natural language processing, network programming languages, and large-scale language modeling technologies, flexible design of services targeting each user intention is achieved, guaranteeing service delivery and response speed while simultaneously improving the diversity of network services and the flexibility of the network. Utilizing automaton theory, a policy translator is modularly designed to handle variations between OpenFlow and NETCONF protocols. Within the NETCONF domain, network policies are extracted and transformed, and configuration files for corresponding network element devices are generated. This enables intelligent organization, integrated management, and maintenance of user services in heterogeneous network environments. Furthermore, based on the concept of closed-loop control, deployed services are monitored and analyzed to design a dynamic service adjustment process, thereby guaranteeing QoS for network services.

[0009] A method for realizing intents in heterogeneous networks based on a network programming language, which combines intent realization with network programming language and policy translation technology, The global SDN controller acquires heterogeneous network resources and status information, The steps include: the user describing network service requirements in voice or text format, The intent analysis engine analyzes the service requirements and extracts intent keywords. The service orchestrator combines network resources and states based on intent keywords to generate a network policy that conforms to the Frenetic network programming language specification. The global network controller selects the corresponding domain controller based on service requirements and distributes network policies. The steps include: a domain controller supporting the OpenFlow protocol compiles the network policy as an OpenFlow flow table, and a domain controller supporting the NETCONF protocol converts the network policy into a YANG data model using a policy translator; The process includes the steps of an OpenFlow controller distributing a flow table to lower-level network element devices that support the OpenFlow protocol, and a NETCONF controller distributing a YANG data model to corresponding NETCONF network element devices, thereby enabling service placement and initiating closed-loop control.

[0010] The aforementioned method, 1) Each domain controller distributes query commands or requests to lower network element devices, analyzes the response data returned by the lower network element devices, extracts necessary information including device topology information, port status information, fault and alarm information, network function information and device configuration, and after completing resource state synchronization, the OpenFlow domain controller and NETCONF domain controller integrate the data into a heterogeneous network global SDN controller. 2) A step in which the user inputs intent in the form of text or voice, wherein the intent includes the user's network service requirements and a high-level description of the network, and represents the network service goals that the user wants to achieve across heterogeneous networks, 3) The intent analysis engine extracts important service information based on the intent entered by the user in natural language, converts the intent data into a format that can be processed by the model, extracts intent keywords from the intent data in a format that conforms to the specifications using a tokenizer, and obtains the service scope desired by the user, the network functions included in the service, and the service lifecycle information. 4) The acquired intent keywords, current network resources, and status are input to the service orchestrator, the service orchestrator generates a network policy conforming to the Frenetic network programming language specification using large-scale language model code generation technology, and the network policy is distributed to the user service by selecting the appropriate domain. 5) For domain controllers with different protocols, the network policy is further processed in different ways. In an OpenFlow domain consisting of network element devices that support the OpenFlow protocol, the Frenetic controller compiles the network policy as an OpenFlow flow table. In a NETCONF domain consisting of network element devices that support the NETCONF protocol, the network policy is extracted and transformed by a policy translator, and further combined with lower-level network element information to generate a YANG model file corresponding to the device. 6) The global SDN controller performs final deployment of the user network service after it has been implemented, in accordance with the service lifecycle described in the user intent, and activates the closed-loop control module as the service goes live; the global SDN controller monitors the service QoS with network state data returned by the domain controller; and if the service is unable to meet the user intent, the global SDN controller initiates a request to the service orchestrator, which then regenerates and deploys the service based on the current network resources and state.

[0011] In step 5) above, the step of converting the network policy in the NETCONF domain into a YANG model file corresponding to a specific device using a policy translator is: 5-1) The extractor extracts functions that describe network functions in a network policy and the objects of their processing based on data from a deterministic finite state automaton, 5-2) A step of comparing network function functions with a database containing lower-level network element functions, and mapping the functions and their processing targets in the network policy to the information necessary for the functions and configuration services of a specific network element device, 5-3) A step in which a policy generator built on context-free grammar generates a "content generation formula" and a "structure generation formula" based on the data after data transformation and specific network element configuration information, wherein the content generation formula is used to include the data in the correct XML tags, and the structure generation formula is responsible for organizing the overall XML structure and ensuring the correct relationships between each part, 5-4) Includes the step of outputting lower network element configuration information, i.e., a YANG model file.

[0012] The beneficial effects of this invention are as follows: This invention discloses a method for setting intents in heterogeneous networks based on a network programming language, realizing intelligent translation from user service requirements to device configurations, simplifying the service implementation process, and improving network performance. Specifically, by combining intent implementation, network programming language, and large-scale language modeling technology, and designing an intelligent retranslation method from user intent to network policy, flexible service design for each user is realized. Furthermore, by designing a policy translator that converts network policies conforming to the Frenetic network programming language specification into network element device configurations based on automaton theory and context-free grammar, the problem of incompatibility between the Frenetic network programming language and the NETCONF protocol is resolved, enabling intelligent organization, integrated management, and maintenance of user services in heterogeneous network environments. Finally, based on the concept of closed-loop control, the QoS of network services is guaranteed by monitoring and analyzing deployed services and designing a dynamic service adjustment process. Embodiments of the present invention can be applied to the network architecture of network service providers, solving the problems that still exist in existing network architectures, such as the difficulty in deploying and maintaining heterogeneous networks, the low degree of intelligence in their organization, and the inability to meet user service requirements in a timely manner. This significantly improves the diversity of network services, as well as the flexibility, accuracy, and stability of the network, and effectively ensures service delivery and response speed. [Brief explanation of the drawing]

[0013] [Figure 1] This is a schematic diagram illustrating a method for realizing intelligent intent setting in heterogeneous networks based on the network programming language of the present invention.

[0014] [Figure 2] This is a diagram illustrating the configuration of the policy translator in the NETCONF domain. [Modes for carrying out the invention]

[0015] In the following, the present invention will be further described while referring to the drawings and embodiments. Embodiment

[0016] A certain network service provider manages a heterogeneous network architecture composed of multiple domains, and each domain is composed of devices that support the OpenFlow protocol or the NETCONF protocol. The network service provider plans to implement intention acquisition, intelligent orchestration, and integrated deployment to perform closed-loop control when a network user issues a service requirement, and the method is as shown in FIG. 1.

[0017] 1-1) Before providing the service to the network user, the network service provider needs to clarify the current network resources and status. The domain controller establishes a connection with the lower-layer network element devices and network links through the OpenFlow protocol or the NETCONF protocol, and distributes query commands or requests. After receiving the request, the lower-layer network element device returns device topology information, port status information, fault and alarm information, network function information, and device configuration data. After completing the synchronization of the resource status, the domain controller aggregates the data to the global SDN controller.

[0018] 1-2) The network service provider provides an interface to the user, and the user expresses his intention in the form of text or voice. Assume that what the user inputs in the form of natural language is "Please prohibit all hosts in the network from accessing the entertainment website on business days." 1-3) After receiving the intention from the user, the intention analysis engine first converts the input natural language into a format that conforms to the specification. For example, { “role”:“user”:, "content": "Please prohibit access to the entertainment websites of all hosts in the network on business days.": } Next, the tokenizer extracts intent keywords from the intended data in a format that conforms to the specifications, including the service scope desired by the user, the network functions included in the service, and the life cycle of the service. For example, Service scope desired by the user: All hosts; Network functions included in the service: Prohibit access to entertainment websites on business days; Service life cycle: Business days.

[0019] 1 - 4) In the service orchestrator, using the intent keywords obtained through the above processing, as well as the lower - level network resources and states collected by the global SDN controller as input, a network policy that conforms to the Frenetic network programming language specification is generated by a fine - tuned large - language model. For example, import frenetic from frenetic.syntax import * class MyApp(frenetic.App): p1 = IP4DstEq(addr1,mask) / / Matches packets with the destination IP address of addr1 p2 = IP4DstEq(addr2,mask) …… pn = IP4DstEq(addrn,mask) / / addr is the IP address of the entertainment website that needs to be blocked selectPacket = Or([p1, p2,..., pn]) Drop(selectPacket) app = MyApp() app.start_event_loop().

[0020] 1-5) Determine the domain to which user services will be deployed based on user information and intent, and then distribute the generated network policy to the corresponding domain controller.

[0021] In an OpenFlow domain composed of OpenFlow network element devices, network policies are compiled directly as OpenFlow flow entries by the Frenetic controller. These are as follows, where 220.181.38.148 is one of the specific IP addresses in the above network policy's addr. Flow Table Entry: Match Fields: - IPsrc != 220.181.38.148 - IPdst = 220.181.38.148 Instructions: - Drop.

[0022] In a NETCONF domain composed of NETCONF network element devices, network policies are further translated into lower-level network element configuration information by a policy translator, and the specific process is as follows, as shown in Figure 2. 1-5-1) The extractor analyzes the network policy based on data from a deterministic finite state automaton and extracts functions that describe network functions in the network policy and the targets of their operations, specifically including IP4DstEq, the drop function, device IP, and port information.

[0023] 1-5-2) Compare network function functions with a database containing lower-level network element functions, and convert the functions and their targets in the network policy into information necessary for the functions and configuration services of a specific network element device, for example, <access-lists> <access-list> <name> BlockEntertainment< / name> <aces> <ace> <name> DenyEntertainmentSite1< / name> <matches> <ipv4> <destination-ip-address> 220.181.38.148< / destination-ip-address> < / ipv4> < / matches> <actions> <forwarding> drop< / forwarding> < / actions> < / ace> <!-- Add more ACEs for other entertainment sites --> < / aces> < / access-list> < / access-lists> .

[0024] 1-5-3) A policy generator built on context-free grammar generates "content generation formulas" and "structure generation formulas" based on the data after it has been transformed by the data converter and the configuration information of specific network elements. The content generation formulas are used to enclose the data in accurate XML tags, and the structure generation formulas are used to packetize other tags. Finally, based on the content of both, it outputs lower-level network element configuration information, i.e., a YANG model file.

[0025] 1-6) Based on the service lifecycle described in the user intent, the embodied network services are deployed. For OpenFlow domains, the domain controller updates the OpenFlow flow table, and for NETCONF domains, the domain controller distributes the YANG file model via NETCONF. As soon as the network services go live, the global SDN controller activates the closed-loop control module and monitors and analyzes network resources and status based on port status information, fault and alarm information returned by the domain controller to determine whether user service QoS is guaranteed. Once the service can no longer satisfy the user intent, the global SDN controller requests the service orchestrator to regenerate and deploy the service based on the current network resources and status.

[0026] The embodiments described above can be further combined or swapped, and the embodiments are merely illustrative of preferred embodiments of the present invention and do not limit the concept or scope of the present invention. Any changes and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the design concept of the present invention will all fall within the scope of the present invention. The scope of the present invention is indicated by the appended claims and all equivalent technical solutions thereof.

Claims

1. A method for realizing intelligent configuration of intents in heterogeneous networks based on the network programming language, which combines intent realization, network programming language and policy translation technology, The global SDN controller acquires information on the resources of the heterogeneous network and the status of those resources. The steps include: the user describing network service requirements in voice or text format, The intent analysis engine analyzes the network service requirements and extracts intent keywords. The service orchestrator receives information about the resources and resource status of the heterogeneous network from the global SDN controller, combines the information about the resources and resource status of the heterogeneous network based on the intent keyword, and generates a network policy that conforms to the Frenetic network programming language specification. The steps include: the global SDN controller selecting a corresponding domain controller according to the network service requirements and distributing the network policy; The OpenFlow domain controller, which supports the OpenFlow protocol, compiles the network policy as an OpenFlow flow table, and the NETCONF domain controller, which supports the NETCONF protocol, converts the network policy to a YANG data model using a policy translator. The process includes the steps of: the OpenFlow domain controller distributing the OpenFlow flow table to lower network element devices that support the OpenFlow protocol; and the NETCONF domain controller distributing the YANG data model to lower network element devices that support the NETCONF protocol, thereby enabling service placement and initiating closed-loop control. The closed-loop control includes the transmission of information about the resources of the heterogeneous network and the status of those resources from the OpenFlow domain controller and the NETCONF domain controller to the global SDN controller, and the transmission of information about the resources of the heterogeneous network and the status of those resources from the global SDN controller to the service orchestrator. A method for realizing intelligent intent setting in heterogeneous networks based on a network programming language, characterized by the above.

2. 1) Each of the OpenFlow domain controller and the NETCONF domain controller delivers an inquiry command or request to the lower network element device, analyzes the response data returned by the lower network element device, extracts necessary information including device topology information, port status information, fault and alarm information, network function information and device configuration, and then the global SDN controller of the heterogeneous network integrates the data from the OpenFlow domain controller and the NETCONF domain controller. 2) A step in which the user inputs the intent in the form of text or voice, wherein the intent includes the network service requirements and a high-level description of the network, and represents the network service objectives that the user wants to achieve across the entire heterogeneous network, 3) The intent analysis engine extracts important service information based on the intent entered by the user in natural language, converts the intent data into a format that can be processed by the model, extracts intent keywords from the intent data in a format that conforms to the specifications using a tokenizer, and obtains the service scope, network functions included in the service, and service lifecycle information desired by the user. 4) Inputting the acquired intent keyword, the current resources of the heterogeneous network, and the status of the resources into the service orchestrator; the service orchestrator generating a network policy conforming to the Frenetic network programming language specification using large-scale language model code generation technology; and distributing the network policy by selecting an appropriate domain for the scope of user services; 5) For domain controllers with different protocols, the network policy is further processed in different ways; in an OpenFlow domain consisting of lower network element devices that support the OpenFlow protocol, the OpenFlow domain controller compiles the network policy as an OpenFlow flow table; in a NETCONF domain consisting of lower network element devices that support the NETCONF protocol, the NETCONF domain controller extracts and transforms the network policy using a policy translator, and further combines it with information from the lower network element devices to generate a YANG model file corresponding to the network element device; 6) The global SDN controller performs the final deployment to the user network service after it has been implemented, in accordance with the service lifecycle described in the user intent. The method according to claim 1, characterized by including the steps of: starting a closed-loop control module simultaneously with the start of production operation; the global SDN controller monitoring service QoS with data on the resources and resource status information of the heterogeneous networks returned by each domain controller; and, once the service is unable to satisfy the user intent, the global SDN controller initiating a request to the service orchestrator, which then regenerates and deploys the service based on the current information on the resources and resource status of the heterogeneous networks.

3. In step 5) above, the step of converting the network policy in the NETCONF domain to the YANG model file corresponding to a specific network element device using the policy translator is: 5-1) The extractor extracts functions that describe network functions in the network policy and the objects to be processed, based on the data of the deterministic finite state automaton. 5-2) A step of comparing network function functions with a database containing lower-level network element functions, and mapping the functions and their processing targets in the network policy to information necessary for the functions and configuration services of a specific network element device, 5-3) A step in which a policy generator built on context-free grammar generates a "content generation formula" and a "structure generation formula" based on the data after data transformation and specific network element configuration information, wherein the content generation formula is used to include the data in accurate XML tags, and the structure generation formula is responsible for organizing the overall XML structure and ensuring the correct relationships between each part, 5-4) The method according to claim 2, characterized by comprising the step of outputting lower network element configuration information, i.e., the YANG model file.

Images