CN-116684269-B - Automatic production network configuration generation method and system based on intention

CN116684269BCN 116684269 BCN116684269 BCN 116684269BCN-116684269-B

Abstract

The invention provides an automatic generating method and system for production network configuration based on intention, comprising the steps of obtaining topology of a designated communication network and inputting a plurality of intention strategies to be executed, wherein the topology comprises network autonomous domain division, obtaining link information and basic information of each node in the topology, obtaining required ports and priorities of each intention strategy, combining all intention strategies to obtain conflict-free intention expression according to whether port conflicts exist between each intention strategy and the priorities of each intention strategy, dividing the topology into global topology and internal topology by taking a network autonomous domain as a basic unit, decomposing the intention expression, taking the priorities of each intention strategy as an objective function, setting link constraint and path uniqueness constraint, obtaining all paths conforming to the intention strategies in the topology through linear programming, and generating all network configuration files of the topology through a network protocol of the production network according to the paths obtained by the intention to ensure that the network state conforms to all intention strategies.

Inventors

ZHANG MENG
YU JINPING
BI JINGPING

Assignees

中国科学院计算技术研究所

Dates

Publication Date: 20260505
Application Date: 20230510

Claims (9)

1. An automatic production network configuration generation method based on intention, which is characterized by comprising the following steps: An initial step of acquiring a topology of a specified communication network and inputting a plurality of first intention strategies to be executed in an opening, wherein the topology nodes comprise basic information of routing nodes, host nodes and network function nodes, the topology comprises network autonomous domain division, and the first intention strategies are expressed by connectivity among network areas taking the network autonomous domain as a basic unit; The information processing step is used for analyzing the topology information to obtain link information and basic information of each node in the topology; An intention synthesizing step, namely combining the first intention strategies to obtain conflict-free intention expression according to whether port conflicts exist between the first intention strategies and the priority of the first intention strategies; Dividing the topology into global topology by taking the network autonomous domain as a basic unit, taking the network autonomous domain internal nodes and the connection relations thereof as internal topology, decomposing the intention expression on the basis of the global topology and the internal topology, taking the priority of each first intention strategy as an objective function, setting link constraint and path uniqueness constraint, and obtaining intention paths conforming to all first intention strategies in a topological graph through linear programming; and a configuration file generation step, namely generating all network configuration files of the topology through a network protocol of the production network according to the paths obtained by intention, and ensuring that the network state accords with all first intention strategies.
2. The automatic intent-based production network configuration generation method of claim 1, further comprising a consistency update step of: the method comprises the steps of obtaining a plurality of second intention strategies to be updated, obtaining a network configuration file of a topological graph, generating network topology basic information, carrying out autonomous domain division through an intra-domain protocol number and a boundary protocol, generating an optimal path of a global topological path, an internal topological path and a synthetic path according to the second intention strategies and the network topology basic information, and generating the updated configuration file of the topological graph through a gateway protocol of a production network by the optimal path so as to configure the generated network topology basic information to operate according to the second intention strategies.
3. The method for automatically generating an intent based production network configuration as recited in claim 2, wherein the step of updating the consistency includes: Step S1, acquiring a plurality of intention strategies to be updated, acquiring a network configuration file of the topological graph, generating network topology basic information, and carrying out autonomous domain division through an intra-domain protocol number and a boundary protocol; Step S2, according to the optimal path and the network configuration file, performing connectivity verification on the head and tail nodes of the optimal path, if the verification is passed, no updating operation is needed, and if the verification is not passed, the step S3 is executed; step S3, obtaining a host node, a boundary node and an internal topology path node of the starting point autonomous domain of the optimal path, so as to test the connectivity of the internal topology path node and the boundary node of the starting point autonomous domain and the connectivity of the host node and the boundary node, if both the nodes are connected, executing step S4, otherwise, carrying out configuration update on the nodes in the starting point autonomous domain, and obtaining the update configuration file; step S4, connectivity detection is carried out on the boundary node of the starting point autonomous domain and the tail node of the optimal path, if the boundary node is communicated with the tail node, updating operation is not needed, otherwise, step S5 is executed; step S5, carrying out configuration file book identification on the boundary node of the starting point autonomous domain, judging whether the configuration information of the network segment where the blocking path end point is located exists in the identification result, if so, updating the boundary node of the starting point autonomous domain to obtain the updated configuration file so as to carry out incremental configuration on the boundary node of the starting point autonomous domain, otherwise, executing step S6; Step S6, obtaining a host node, a boundary node and an internal topology path node of the terminal autonomous domain of the optimal path so as to test connectivity of the internal topology path node and the boundary node of the terminal autonomous domain and connectivity of the host node and the boundary node, if both are connected, executing step S7, otherwise, carrying out configuration update on the node in the terminal autonomous domain to obtain the update configuration file; Step S7, carrying out configuration file book identification on the boundary node of the terminal autonomous domain, judging whether configuration information of a network segment where a blocked path terminal is located exists in an identification result, if so, updating the boundary node of the terminal autonomous domain to obtain an updated configuration file, and carrying out global topology path reverse sequence increment configuration on the boundary node of the terminal autonomous domain to obtain the updated configuration file; Step S8, performing reverse sequence traversal text recognition on other nodes except the starting point autonomous domain and the end point autonomous domain in the optimal path, if all the intended network segments are matched, performing global topology path reverse sequence incremental configuration on the other nodes to obtain the updated configuration file, otherwise, executing step S9; And S9, replacing the path in the starting point autonomous domain in the optimal path with the path selected by initial configuration, and configuring the global topology path in reverse order to obtain the updated configuration file.
4. The automatic intent-based production network configuration generation method of claim 1, wherein the designated communication network is a telecommunications core network and a data center network.
5. An intent-based production network configuration auto-generation system, comprising: The method comprises the steps that an initial module is used for designing network topology by itself or obtaining the topology of a production network and inputting a plurality of first intention strategies to be executed in an opening mode, wherein topology nodes comprise basic information of routing nodes, host nodes and network function nodes, the topology comprises network autonomous domain division, and the first intention strategies are expressed by connectivity among network areas taking the network autonomous domain as a basic unit; The information processing module analyzes the topology information to obtain link information and basic information of each node in the topology; The intention synthesizing module combines the first intention strategies to obtain conflict-free intention expression according to whether port conflicts exist between the first intention strategies and the priority of the first intention strategies; The topology splitting path solving module divides the topology into global topology by taking the network autonomous domain as a basic unit, takes the internal nodes of the network autonomous domain and the connection relation thereof as internal topology, decomposes the intention expression on the basis of the global topology and the internal topology, takes the priority of each first intention strategy as an objective function, sets link constraint and path uniqueness constraint, and obtains intention paths conforming to all first intention strategies in a topological graph through linear programming; and the configuration file generation module is used for generating all network configuration files of the topology through a network protocol of the production network according to the paths obtained by intention, so as to ensure that the network state accords with all first intention strategies.
6. The intent based production network configuration auto-generation system of claim 5, further comprising a consistency update module: the method comprises the steps of obtaining a plurality of second intention strategies to be updated, obtaining a network configuration file of a topological graph, generating network topology basic information, carrying out autonomous domain division through an intra-domain protocol number and a boundary protocol, generating an optimal path of a global topological path, an internal topological path and a synthetic path according to the second intention strategies and the network topology basic information, and generating the updated configuration file of the topological graph through a gateway protocol of a production network by the optimal path so as to configure the generated network topology basic information to operate according to the second intention strategies.
7. The automatic intent-based production network configuration generation system of claim 6, wherein the consistency update module comprises: The module 1 acquires a plurality of intention strategies to be updated, acquires a network configuration file of the topological graph, generates network topology basic information, and performs autonomous domain division through an intra-domain protocol number and a boundary protocol; the module 2 performs connectivity verification on the head and tail nodes of the optimal path according to the optimal path and the network configuration file, and if the verification is passed, no updating operation is needed; A module 3, acquiring a host node, a boundary node and an internal topology path node of the starting point autonomous domain of the optimal path, so as to test connectivity of the internal topology path node and the boundary node of the starting point autonomous domain and connectivity of the host node and the boundary node, if both are connected, executing the module 4, otherwise, carrying out configuration update on the node in the starting point autonomous domain, and obtaining the update configuration file; The module 4 performs connectivity detection on the boundary node of the starting point autonomous domain and the tail node of the optimal path, if the boundary node is communicated with the tail node, updating operation is not needed, otherwise, the module 5 is executed; The module 5, carrying out configuration file book identification on the boundary node of the starting point autonomous domain, judging whether the configuration information of the network segment where the blocking path end point is located exists in the identification result, if yes, updating the boundary node of the starting point autonomous domain to obtain the updated configuration file so as to carry out incremental configuration on the boundary node of the starting point autonomous domain, otherwise, executing the module 6; a module 6, acquiring a host node, a boundary node and an internal topology path node of the terminal autonomous domain of the optimal path, so as to test connectivity of the internal topology path node and the boundary node of the terminal autonomous domain and connectivity of the host node and the boundary node, if both are connected, executing a module 7, otherwise, carrying out configuration update on the node in the terminal autonomous domain, and obtaining the update configuration file; The module 7, carrying out configuration file book identification on the boundary node of the terminal autonomous domain, judging whether the configuration information of the network segment where the blocking path terminal is located exists in the identification result, if yes, updating the boundary node of the terminal autonomous domain to obtain the updated configuration file, and carrying out global topology path reverse sequence increment configuration on the boundary node of the terminal autonomous domain to obtain the updated configuration file; The module 8 performs reverse sequence traversal text recognition on other nodes except the starting point autonomous domain and the end point autonomous domain in the optimal path, if all the intended network segments are matched, the other nodes are subjected to global topology path reverse sequence incremental configuration to obtain the updated configuration file, otherwise, the module 9 is executed; And a module 9, replacing the path in the starting point autonomous domain in the optimal path with the path selected by initial configuration, and configuring the global topology path in reverse order to obtain the updated configuration file.
8. A storage medium storing a program for executing the automatic intention-based production network configuration generation method according to any one of claims 1 to 4.
9. A client for the intent-based production network configuration auto-generation system of any one of claims 5 to 7.

Description

Automatic production network configuration generation method and system based on intention Technical Field The invention relates to the technical field of planning and configuration of production network switches, operation and maintenance configuration of a telecommunication core network, a data center network and the like. Background In the traditional network environment, in order to solve the problems of low efficiency, easy error, high cost, difficult maintenance and the like of manual configuration, various manufacturers and academia gradually adopt an automatic mode to realize various network functions, and given network requirements are converted into actual network configuration. Among the commonly used network functions are techniques such as network configuration generation and updating. Network configuration generation refers to the process of translating a given network requirement into an actual network configuration. In recent years, a number of automated network configuration generation techniques have emerged, such as automated synthesis of network configurations using techniques of model detection, constraint programming, automated planning, symbolic execution, and the like. The techniques can greatly improve the efficiency and accuracy of network configuration and can automatically manage and maintain the network. The network configuration update refers to a process of adjusting the configuration of the network device due to service requirements or daily operation and maintenance update, but the network configuration update needs to ensure that the normal operation of other services is not affected in the update process and after the update. Automated network configuration updating techniques include using automated scripts, automated tools, network programming interfaces, network automation platforms, continuous integration and deployment, network security auditing, and the like to automatically update and manage network configurations. The techniques can greatly improve the efficiency and accuracy of network configuration updating, and can automatically manage and maintain the network. In practical application, in order to provide a network operation environment in which a non-professional network operator can quickly get on hand, it is necessary to set an input intention with simplicity, easy operation and strong expression capability in the process of generating and updating network configuration. "intent" herein is a higher level abstract representation that is independent of the particular underlying network, and an intent graph composed of nodes and edges of the higher level abstract may be utilized to represent the network administrator's administrative intent for the underlying network. For example, as shown in fig. 1, there are office areas ZoneA and ZoneB, where the gray boxes represent designated passing network function nodes, including Firewall (FW), load Balancer (LB), etc. It is intended to communicate to two areas and specify the Port, bandwidth BW and Weight information Weight passed, which is intended to be expressed in the form of a graph. The intention mode has the advantages that the complex and complicated configuration of the bottom layer and the huge scale can be represented by a simple and small amount of intention strategies, so that the purposes of reducing the workload of a network manager and improving the working efficiency are achieved. When performing configuration updates in a traditional network environment, additional installations and configurations are required for different network scenarios, such as preparing the correct device list, the correct templates, the correct commands, etc., before using the automation tools, which is very inefficient. The current network configuration generation and updating techniques suffer from the disadvantage that 1) parsing for high level abstract intent does not support generating configurations that can be directly performed by different vendor (e.g., cisco, huawei) devices in the production network. Different vendor devices often have different configuration commands and formats, and there is currently no work directly from intent to configuration of a vendor device. 2) Other configuration generation methods are tightly coupled with the bottom production network, and do not support a description mode which is higher in abstraction and is close to natural language, so that an administrator is required to have relevant expertise and know the current network situation very well. This significantly increases the difficulty of network configuration management and as the network scale increases, the amount of work required to input intent plans may double or even exponentially increase. For example, inputs require explicit path and parameter designs for various protocols and configuration sketches closely related to the underlying network devices. Because of the complex input, the method is unfavorable for being applied