Method, device and medium for arranging and adjusting infrastructure interactive network topology

Abstract

The application relates to the technical field of data processing, and discloses a method and equipment for arranging and adjusting an infrastructure interactive network topology and a medium, which comprises the following steps: in response to an arrangement request triggered by a target object for a topology structure, performing semantic analysis on the arrangement request to obtain topology feature information and an operation mode corresponding to the arrangement request; obtaining a first topology state object corresponding to the operation mode; determining a topology arrangement task corresponding to the topology structure arrangement request based on the topology feature information and the operation mode; calling an atomic operation tool corresponding to each topology arrangement task to perform arrangement processing on the first topology state object to obtain a second topology state object; and performing serialization processing on the second topology state object to obtain a target topology structure corresponding to the arrangement request. Through implementation of the above method provided by the application, a topology structure meeting user requirements can be automatically generated according to a user input arrangement request, and the arrangement and adjustment efficiency of the network topology is improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method, device, and medium for orchestrating and adjusting the topology of an interactive infrastructure network. Background Technology

[0002] A network topology diagram is a graph structure model that describes the nodes and devices in a network, their interconnections, link attributes, deployment constraints, and communication paths. It is used to characterize the network architecture and resource allocation scheme. In network planning and operation, the network topology diagram is not only the foundation for visualization but also the data source for subsequent configuration generation, deployment, and status management.

[0003] In related technologies, network engineers need to manually plan nodes such as routers, switches, firewalls, servers, and cloud gateways according to user requirements documents, determine the number of nodes, device types, connection relationships, network segments, bandwidth, mirrors, and deployment constraints, and then write configurations in network simulation platforms, cloud platforms, or infrastructure-as-code tools to achieve topology construction for network scenarios.

[0004] However, since this method of manually constructing topologies mainly relies on human experience, engineers usually need to switch back and forth between understanding requirements, drawing topologies, and converting configurations. Especially in medium or large network scenarios, where there are many nodes and complex connection relationships, engineers need to manually maintain the correspondence between node identifiers, link endpoint references, and attribute values, resulting in low efficiency in network topology orchestration. Summary of the Invention

[0005] This invention provides a method, device, and medium for orchestrating and adjusting interactive network topology of infrastructure, in order to solve the problem of low efficiency in related technologies when relying solely on human experience for topology construction.

[0006] In a first aspect, the present invention provides a method for orchestrating and adjusting the topology of an interactive infrastructure network, comprising: responding to an orchestration request triggered by a target object for a topology structure, performing semantic parsing on the orchestration request to obtain topology feature information and an operation mode corresponding to the orchestration request; obtaining a first topology state object corresponding to the operation mode; determining at least one topology orchestration task corresponding to the topology structure orchestration request based on the topology feature information and the mode information of the operation mode; invoking atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object to obtain a second topology state object; and performing serialization processing on the second topology state object to obtain the target topology structure corresponding to the orchestration request.

[0007] The method for orchestrating and adjusting interactive network topology of infrastructure provided in this embodiment of the invention establishes the network topology orchestration process on an interactive framework driven by atomic operation tools and based on intermediate topology state objects. Users only need to input an orchestration request, and the target topology structure is automatically generated through steps such as semantic parsing of the orchestration request, decomposing the topology generation process into serialized calls of atomic operation tools, and performing step-by-step orchestration calls on the state object. This significantly improves the efficiency of network topology generation.

[0008] In some optional implementations, obtaining the first topology state object corresponding to the operation mode includes: if the operation mode is a build mode, then initializing an empty topology state object; the build mode is used to characterize the generation of a topology structure based on an orchestration request; the initial empty topology state object is used to determine the first topology state object corresponding to the build mode; if the operation mode is a modify mode, then reading the initial topology structure corresponding to the orchestration request from the target database; the modify mode is used to characterize the modification of a saved topology structure; the initial topology structure is deserialized to obtain the first topology state object corresponding to the modify mode.

[0009] In the infrastructure interactive network topology orchestration and adjustment method provided in this embodiment of the invention, different methods are used for the first topology state object for the construction mode and the modification mode, so that the system can flexibly adapt to the two scenarios of creating a new topology and modifying the topology, avoiding the inefficiency problem caused by having to rebuild the complete topology in the modification scenario.

[0010] In some optional implementations, the atomic operation tools corresponding to each topology orchestration task are invoked to orchestrate the first topology state object to obtain the second topology state object. This includes: obtaining topology constraint information; the topology constraint information is used to constrain the tool call parameters corresponding to the atomic operation tools to meet a first preset condition; loading the topology constraint information into a first agent, obtaining the tool call context, the first agent being used to decompose the topology orchestration task into multiple atomic operation steps with dependent execution order according to the topology constraint information, and determining the atomic operation tools and tool call parameters corresponding to each atomic operation step; and using the tool call context, invoking the atomic operation tools corresponding to each topology orchestration task to orchestrate the first topology state object to obtain the second topology state object.

[0011] In the method for orchestrating and adjusting the interactive network topology of infrastructure provided in this embodiment of the invention, topology structure constraint information is obtained and loaded into a first intelligent agent to obtain a tool invocation context, and atomic operation tools are invoked through the tool invocation context. At the same time, by decomposing the topology orchestration task into atomic operation steps with dependent execution order, the complex topology construction is transformed into a verifiable and error-correctable sequence of steps, thereby reducing the generation failure rate.

[0012] In some optional implementations, the topology orchestration task includes a node addition task, which calls the atomic operation tools corresponding to each topology orchestration task to orchestrate the first topology state object to obtain a second topology state object. This includes: calling the node addition tool corresponding to the node addition task to add a first node to the first topology state object; checking whether the node identifier, device type, and node attributes of the first node meet a second preset condition; and if the node identifier, device type, and node attributes of the first node all meet the second preset condition, then constructing the second topology state object based on the first topology state object and the first node.

[0013] In the method for orchestrating and adjusting the interactive network topology of infrastructure provided in this embodiment of the invention, since the verification process for newly added nodes is performed immediately after the atomic operation tool is executed, errors can be detected and corrected in a timely manner during the node addition stage, without the need for an overall check after the complete topology is generated, thus improving the reliability and efficiency of topology construction.

[0014] In some optional implementations, the topology orchestration task further includes a node batch addition task, which calls the atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object to obtain a second topology state object, including: obtaining a node list from the topology feature information; grouping and identifying each node in the node list to obtain at least one isomorphic node group; and in response to identifying the isomorphic node group, calling the node batch addition tool to create all the nodes contained in the isomorphic node group in the first topology state object to obtain the second topology state object.

[0015] In the method for orchestrating and adjusting the interactive network topology of infrastructure provided in this embodiment of the invention, each node in the node list is grouped and identified, and nodes that meet the homogeneity condition are grouped into homogeneous node groups. The node batch addition tool is then called to create all nodes at once. Compared with the method of calling the node addition tool one by one, this significantly reduces the number of tool calls and further improves the efficiency of orchestrating and adjusting the interactive network topology of infrastructure.

[0016] In some optional implementations, the nodes in the node list are grouped and identified to obtain at least one homogeneous node group, including: if multiple nodes simultaneously meet at least two of the following conditions: the node type is the same, the mirror identifier or specification parameters are the same, the number of nodes is greater than a preset threshold, and they have a continuous IP address range, then the multiple nodes are determined to belong to the same homogeneous node group.

[0017] In some optional implementations, the topology orchestration task further includes an edge-adding task, which is used to add a link between the first node and the second node. The edge-adding task calls the atomic operation tools corresponding to each topology orchestration task to orchestrate the first topology state object to obtain a second topology state object. This includes: determining whether the first node and the second node exist, and whether there is a link between the first node and the second node; if there is no link between the first node or the second node, calling the node-adding tool to add the first node or the second node to the first topology state object; calling the edge-adding task to add a first link between the first node and the second node; and constructing the second topology state object based on the first link and the first topology state object.

[0018] The method for orchestrating and adjusting the interactive network topology of infrastructure provided in this invention effectively prevents the generation of duplicate links in concurrent scenarios or under repeated requests by confirming that the link does not exist before adding the link. This ensures the uniqueness and correctness of the link set in the topology state, so that the target topology structure generated by subsequent serialization does not contain redundant link data, thereby improving the quality of orchestration and adjustment of the interactive network topology of infrastructure.

[0019] Secondly, the present invention provides an apparatus for orchestrating and adjusting the topology of an interactive infrastructure network, comprising: a parsing module, configured to perform semantic parsing on the orchestration request in response to an orchestration request triggered by a target object for a topology structure, to obtain topology feature information and an operation mode corresponding to the orchestration request; an acquisition module, configured to acquire a first topology state object corresponding to the operation mode; a determination module, configured to determine at least one topology orchestration task corresponding to the topology orchestration request based on the topology feature information and the mode information of the operation mode; an orchestration processing module, configured to call the atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object to obtain a second topology state object; and a serialization processing module, configured to perform serialization processing on the second topology state object to obtain the target topology structure corresponding to the orchestration request.

[0020] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the infrastructure interactive network topology arrangement and adjustment method of the first aspect or any corresponding embodiment described above.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the infrastructure interactive network topology arrangement and adjustment method of the first aspect or any corresponding embodiment described above.

[0022] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the method for orchestrating and adjusting the infrastructure interactive network topology of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first method for orchestrating and adjusting the interactive network topology of infrastructure according to an embodiment of the present invention; Figure 3 This is a second flowchart illustrating the method for orchestrating and adjusting the interactive network topology of infrastructure according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the architecture of a network topology generation system provided in an embodiment of the present invention; Figure 5 This is a structural block diagram of an infrastructure interactive network topology orchestration and adjustment device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0027] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0028] In related technologies, network engineers need to manually plan nodes such as routers, switches, firewalls, servers, and cloud gateways according to user requirements documents, determine the number of nodes, device types, connection relationships, network segments, bandwidth, mirrors, and deployment constraints, and then write configurations in network simulation platforms, cloud platforms, or infrastructure-as-code tools to achieve topology construction for network scenarios.

[0029] However, since this method of manually constructing topologies mainly relies on human experience, engineers usually need to switch back and forth between understanding requirements, drawing topologies, and converting configurations. Especially in medium or large network scenarios, where there are many nodes and complex connection relationships, engineers need to manually maintain the correspondence between node identifiers, link endpoint references, and attribute values, resulting in low efficiency in network topology orchestration.

[0030] Therefore, embodiments of the present invention provide a method for orchestrating and adjusting interactive network topology of infrastructure. By establishing the network topology orchestration process on an interactive framework driven by atomic operation tools and based on intermediate topology state objects, users only need to input orchestration requests. The target topology structure is automatically generated through steps such as semantic parsing of the orchestration request, decomposing the topology generation process into serialized calls of atomic operation tools, and performing step-by-step orchestration calls on the state objects. This significantly improves the efficiency of network topology generation.

[0031] As an optional application scenario of this invention, such as Figure 1 The diagram shown illustrates the system architecture used in the infrastructure interactive network topology orchestration and adjustment method provided in this embodiment of the invention. The system may include at least one electronic device and at least one server. Figure 1 The system is illustrated in the example, which includes an electronic device 101 and a server 102, with the electronic device 101 connected to the server 102 via a network 110.

[0032] In this context, electronic device 101 can be a workstation, personal computer, server, or high-performance computing cluster. Server 102 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranets, local area networks, wide area networks, mobile communication networks, and combinations thereof.

[0033] According to an embodiment of the present invention, an embodiment of an infrastructure interactive network topology orchestration and adjustment method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0034] This embodiment provides a method for orchestrating and adjusting the interactive network topology of infrastructure, which can be used in the aforementioned electronic devices. Figure 2 This is a flowchart of a method for orchestrating and adjusting the interactive network topology of infrastructure according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: In response to the orchestration request triggered by the target object for the topology structure, perform semantic parsing on the orchestration request to obtain topology feature information and the operation mode corresponding to the orchestration request.

[0035] The target object refers to the entity that initiates the topology orchestration request. This target object can be a user, such as a network engineer or operations and maintenance personnel.

[0036] Topology refers to the overall layout of various nodes and their connections in a network. Topology includes nodes and the connection links between nodes. Nodes can include network devices such as routers, switches, firewalls, servers, and cloud gateways. Connection links can include physical links or logical links between nodes.

[0037] An orchestration request is an instruction issued by a target object to create or adjust a topology. An orchestration request can include any one or more of the following: natural language text, structured commands, graphical interface operation signals, or application programming interface call parameters.

[0038] In this embodiment of the invention, when the system receives an orchestration request triggered by a target object regarding a topology, it first performs semantic parsing on the orchestration request to extract structured data from it. Through semantic parsing, topology feature information and the operation mode corresponding to the orchestration request can be obtained.

[0039] Among them, topology feature information refers to the various parameters and constraints related to the target topology structure extracted from the orchestration request. Topology feature information includes at least device type, number of devices, hierarchical structure, connection relationship, network area, bandwidth requirements, protocol type, mirror information, network segment information, and service constraints.

[0040] In some optional implementations, semantic parsing of orchestration requests may include the following steps: identifying the topology objects and operational intents involved in the orchestration request; extracting various parameters from the orchestration request; and determining whether the orchestration request belongs to a build mode or a modification mode. The topology objects may include nodes and links; the operational intents may include any one or more of the following: create, add, delete, and modify.

[0041] For example, device types can include routers, switches, firewalls, servers, cloud gateways, etc. Device quantity refers to the number of instances of each type of device. Hierarchical structure refers to the layered architecture of the network, such as core layer, aggregation layer, and access layer. Connection relationship refers to the interconnection method between nodes, such as point-to-point connection, star connection, dual-homed connection, and full mesh connection. Network area refers to the security domain or functional area to which the network belongs, such as intranet, isolation zone, production zone, and office zone. Bandwidth requirement refers to the transmission rate required by the link. Protocol type refers to the routing or communication protocol running on the network nodes, such as Open Shortest Path First (OSF) and Border Gateway Protocol (BGP). Image information refers to the firmware or operating system image identifier on which the device operates. Network segment information refers to the Internet Protocol (IP) address allocation range. Business constraints refer to the business rules that the topology must meet, such as redundancy requirements, isolation requirements, and deployment constraints.

[0042] Operation modes are used to characterize the operation type corresponding to an orchestration request. Operation modes include at least build modes and modify modes. The build mode refers to the mode of generating a completely new topology from scratch; the modify mode refers to the mode of making local adjustments based on an existing topology. Local adjustments can include any one or more of the following: adding nodes, deleting nodes, adding links, deleting links, and adjusting attributes.

[0043] Step S202: Obtain the first topology state object corresponding to the operation mode.

[0044] A topology state object is a structured data object maintained internally by the system to describe the current state of the topology. This topology state object includes at least a topology identifier, version number, node set, and link set. It may also include any one or more of node attributes, link attributes, protocol type, and topology tags. The topology state object is not the final output topology file itself, but rather an internal state carrier before the final file is generated; all operations on adding, deleting, or modifying the topology are first performed on this topology state object.

[0045] The first topology state object refers to the topology state object that already exists within the system before the topology orchestration operation is performed. The first topology state object has different meanings in different operation modes: for example, in build mode, since there is no existing topology, the first topology state object is an empty state object; in modify mode, the first topology state object is the state object corresponding to an existing topology.

[0046] Step S203: Based on topology feature information and operation mode pattern information, determine at least one topology orchestration task corresponding to the topology orchestration request.

[0047] Mode information refers to the rules and constraints associated with an operating mode, used to guide the generation of topology orchestration tasks. The content of mode information differs depending on the operating mode. For example, in build mode, mode information may include one or more of the following: topology initialization requirements, node and link addition order specifications, and a complete list of device types. In modify mode, mode information may include one or more of the following: methods for reading existing topologies, scope limitations for local operations, and cascading cleanup rules for deletion operations.

[0048] A topology orchestration task refers to the operational objective that needs to be executed on the first topology state object to achieve the target topology structure corresponding to the orchestration request. Each topology orchestration task corresponds to a high-level operational objective. An orchestration request is usually broken down into multiple topology orchestration tasks, and there are sequential dependencies between the topology orchestration tasks.

[0049] In this embodiment of the invention, the system can decompose the orchestration request into at least one topology orchestration task based on the parsed topology feature information and operation mode, combined with the mode information corresponding to the operation mode.

[0050] In some optional implementations, the node creation task can first be determined based on the device type, number of devices, and hierarchical structure in the topology feature information; then, the link creation task can be determined based on the connection relationship and network area in the topology feature information; and finally, the execution order of each task can be determined based on the mode information of the operation mode.

[0051] In some optional implementations, when the operation mode is build mode, the corresponding topology orchestration tasks may include: a topology initialization task, a node creation task, and a link creation task. The topology initialization task is used to create an empty topology framework, the node creation task is used to add nodes to the topology framework, and the link creation task is used to establish connections between the added nodes.

[0052] In some optional implementations, when the operation mode is modification mode, the corresponding topology orchestration task may include any one or more of the following tasks: node addition task, node deletion task, link addition task, link deletion task, and attribute adjustment task.

[0053] Step S204: Call the atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object to obtain the second topology state object.

[0054] Atomic operation tools refer to the smallest functional units that are pre-registered and maintained by the system. Each atomic operation tool corresponds to a change operation of the smallest granularity to the topology state object. The system pre-maintains a library of atomic operation tools, which may include the following tools: topology creation tool, node addition tool, batch node creation tool, node deletion tool, edge addition tool, edge deletion tool, and attribute update tool.

[0055] Among them, the Create Topology tool is used to reset and initialize an empty topology state; the Add Node tool is used to add a single node to the node set of the first topology state object; the Batch Create Node tool is used to create multiple isomorphic nodes at once according to the node template and quantity; the Delete Node tool is used to delete a specified node and simultaneously delete the associated links of that node; the Add Edge tool is used to add a link between two existing nodes; the Delete Edge tool is used to delete a link between two nodes; and the Update Attribute tool is used to update any one or more of node attributes, link attributes, or topology-level attributes.

[0056] In this embodiment of the invention, after determining at least one topology orchestration task corresponding to the topology orchestration request, atomic operation tools corresponding to each topology orchestration task can be invoked to perform orchestration processing on the first topology state object. Orchestration processing refers to the process of performing a series of modification operations on the first topology state object by sequentially executing one or more atomic operation tool invocations. Each topology orchestration task corresponds to one or more atomic operation tool invocations.

[0057] In some optional implementations, the atomic operation tools corresponding to each topology orchestration task are invoked, including: determining the atomic operation tool that matches the task type from the atomic operation tool library based on the task type and task parameters of the topology orchestration task; mapping the task parameters to the input parameters of the atomic operation tool; triggering the execution of the atomic operation tool; and receiving the execution result of the atomic operation tool.

[0058] The second topology state object refers to the updated topology state object obtained after performing all atomic operation tool calls on the first topology state object. The second topology state object reflects the latest state of the topology after the orchestration process is completed, and it has the same data structure as the first topology state object.

[0059] Step S205: Serialize the second topology state object to obtain the target topology structure corresponding to the orchestration request.

[0060] Serialization refers to the process of converting a second topological state object from a structured data representation within the system into a specific format that can be recognized and used by an external system or platform.

[0061] The target topology structure refers to the final topology output corresponding to the original orchestration request, obtained after serialization. The form of this target topology structure depends on the requirements of the subsequent use case and can include one or more of the following: a configuration file of the topology description in JSON (JavaScript Object Notation) format, a visual network diagram object, or a platform-specific topology import format. The target topology structure and the second topology state object are essentially the same topology data represented in different forms.

[0062] In this embodiment of the invention, the system can convert the second topology state object into a target topology structure of the corresponding format by performing serialization processing on the second topology state object.

[0063] The method for orchestrating and adjusting interactive network topology of infrastructure provided in this invention automatically transforms natural language requirements into structured topology features and operation modes through semantic parsing, replacing manual understanding and drawing. Atomic operation tools are used to orchestrate and verify the topology state step by step, avoiding engineers repeatedly switching between requirement understanding, topology drawing, and configuration conversion. Finally, it is automatically serialized into a standard format topology file, significantly reducing the burden of manually maintaining node identifiers, link references, and attribute correspondences, and effectively improving topology orchestration efficiency, especially in medium to large-scale complex network scenarios.

[0064] This embodiment provides a method for orchestrating and adjusting the interactive network topology of infrastructure, which can be used in the aforementioned electronic devices, etc. Figure 3 This is a flowchart of a method for orchestrating and adjusting the interactive network topology of infrastructure according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: In response to the orchestration request triggered by the target object regarding the topology, semantic parsing is performed on the orchestration request to obtain topology feature information and the operation mode corresponding to the orchestration request. For details, please refer to... Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0065] Step S302: Obtain the first topology state object corresponding to the operation mode.

[0066] Specifically, step S302 includes: Step S3021: If the operation mode is construction mode, then initialize an empty topology state object.

[0067] In this embodiment of the invention, if the operation mode is the build mode, an empty topology state object is initialized, and this initialized empty topology state object is determined as the first topology state object corresponding to the build mode. The build mode is used to characterize the operation type of generating a topology structure based on an orchestration request, i.e., creating a completely new topology from scratch.

[0068] Step S3022: Initialize the empty topology state object to determine the first topology state object corresponding to the construction mode.

[0069] In build mode, the first topology state object is an empty topology state object obtained during initialization. This empty topology state object does not contain any nodes or links, but only carries the topology identifier and initial version number assigned by the system.

[0070] Step S3023: If the operation mode is modification mode, then read the initial topology structure corresponding to the orchestration request from the target database.

[0071] If the operation mode is modification mode, the initial topology structure corresponding to the orchestration request is read from the target database, and the initial topology structure is deserialized to obtain the first topology state object corresponding to the modification mode.

[0072] The modification mode characterizes operations that modify an existing topology, i.e., perform local adjustments based on an existing topology. The target database refers to the database or file system that stores the saved topology. The initial topology refers to the saved topology description file or stored record, whose format may include JSON or other structured formats. Deserialization is the process of converting the initial topology from its storage format into a topology state object that is readable and writable within the system.

[0073] Step S3024: Deserialize the initial topology structure to obtain the first topology state object corresponding to the modified mode.

[0074] Deserialization refers to the process of converting the initial topology from a storage format into a topology state object that can be read and written by the system. That is, according to the preset data mapping rules, the node data, link data and attribute data in the storage format are mapped to the node set, link set and attribute field of the topology state object, respectively.

[0075] Through the above deserialization process, the stored data of the existing topology can be restored to the internal state of the system, so that subsequent atomic operation tools can directly perform incremental modifications on this state without rebuilding the existing parts, thereby improving the processing efficiency in the modification mode.

[0076] Step S303: Based on topology feature information and operation mode pattern information, determine at least one topology orchestration task corresponding to the topology orchestration request. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0077] Step S304: Invoke the atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object, obtaining the second topology state object. For details, please refer to [link to relevant documentation]. Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0078] In some optional implementations, after determining at least one topology orchestration task corresponding to the topology orchestration request, topology constraint information can also be obtained, and then the topology constraint information can be loaded into the first agent to obtain the tool invocation context. Then, the atomic operation tools corresponding to each topology orchestration task can be invoked using the tool invocation context to perform orchestration processing on the first topology state object to obtain the second topology state object.

[0079] Topology constraint information refers to the set of rules used to constrain the tool call parameters corresponding to atomic operation tools to meet the first preset condition. Topology constraint information serves as the context constraint when the large language model calls atomic operation tools, and also as the basis for verification after tool execution, reducing issues such as illegal device types, missing fields, non-existent link endpoints, and attributes not recognized by the platform.

[0080] In some optional implementations, topology constraint information may include: network graph specifications, device type constraints, and platform-supported node and link attributes. The network graph specifications define the overall structure of the topology data, including node fields, link fields, node identifier referencing methods, link directions, topology labels, grouping information, and the final serialization format. Device type constraints limit the types of devices supported by the system and the allowed attributes for each type. Device types include any one or more of cloud gateways, core routers, access routers, firewalls, and servers. Platform-supported node and link attributes limit the attribute fields and value ranges that the visualization or deployment platform can recognize, including any one or more of bandwidth, network segment, interface, protocol, virtual LAN, security domain, and image binding.

[0081] The first intelligent agent refers to an artificial intelligence agent equipped with a large language model, which is used to decompose the topology orchestration task into multiple atomic operation steps with dependent execution order based on the topology constraint information, and to determine the atomic operation tools and tool call parameters corresponding to each atomic operation step.

[0082] Loading topology constraint information into the first intelligent agent means injecting network graph specifications, device type constraints, and platform-supported node and link attributes into the first intelligent agent's system prompts or context windows, so that the first intelligent agent is subject to the common constraints of the above-mentioned constraint information when understanding user needs and performing task decomposition.

[0083] The tool invocation context refers to the structured information set generated by the first agent under the constraints of topological structure information, used to guide the invocation of atomic operation tools. The tool invocation context includes at least: the atomic operation tool identifier corresponding to each atomic operation step, the tool invocation parameters corresponding to each atomic operation tool, and the execution order among the atomic operation steps. Tool invocation parameters include any one or more of the following: tool name, applicable object, executable action, and input parameters.

[0084] For example, after loading device type constraints into the first agent, if a user requests to add a "load balancer" but the platform does not support this type, the first agent will be rejected by the verification module when calling the add node tool, or will be required to clarify and map to a supported type. As another example, after loading the network graph specification into the first agent, if the network graph specification stipulates that both ends of a link must reference an existing node, the first agent needs to first determine the atomic operation steps corresponding to the add node tool, and then determine the atomic operation steps corresponding to the add edge tool.

[0085] The system sequentially invokes the corresponding atomic operation tools according to the atomic operation steps and their execution order determined in the tool invocation context to orchestrate the first topology state object. Each atomic operation step corresponds to one atomic operation tool invocation, and the tool invocation parameters determined in the tool invocation context are passed in during the invocation.

[0086] In this embodiment of the invention, by obtaining topological constraint information and loading it into the first agent to obtain the tool invocation context, the first agent is subject to the joint constraints of network graph specifications, device type constraints, and platform attribute constraints during task decomposition and tool selection. This transforms the model output from free text generation into constrained tool invocation decisions, avoiding problems such as the model arbitrarily inventing illegal device types, ignoring platform field restrictions, and outputting unparseable content.

[0087] Meanwhile, by breaking down the topology orchestration task into atomic operation steps with execution order dependencies, complex topology construction is transformed into a verifiable and error-correctable sequence of steps, rather than generating a complete file all at once, thus reducing the generation failure rate.

[0088] Step S305: Serialize the second topology state object to obtain the target topology structure corresponding to the orchestration request. For details, please refer to [link to relevant documentation]. Figure 2 Step S205 of the illustrated embodiment will not be described again here.

[0089] In the infrastructure interactive network topology orchestration and adjustment method provided by the embodiments of the present invention, the first topology state object is obtained in different ways according to different operation modes. That is, an empty topology state object is initialized in the construction mode, and an existing topology structure is read from the target database in the modification mode for local adjustment. This enables the system to flexibly adapt to both the new topology and the modification scenario, avoiding the inefficiency caused by the need to rebuild the complete topology in the modification scenario.

[0090] In some optional implementations, when a node addition task is included in at least one topology orchestration task corresponding to a topology orchestration request, the first topology state object can be orchestrated in the following manner.

[0091] Specifically, firstly, the node adding tool corresponding to the node adding task is invoked to add the first node in the first topology state object; for the first node, the node identifier, device type and node attributes of the first node are checked to see if they meet the second preset conditions; if the node identifier, device type and node attributes of the first node all meet the second preset conditions, then the second topology state object is constructed based on the first topology state object and each node.

[0092] The node adding tool is used to add a single node to the node set of the first topology state object. The first agent can determine the node adding tool from the atomic operation tool library according to the type and parameters of the node adding task in the topology orchestration task, and then trigger the tool to execute after mapping the task parameters to the input parameters of the node adding tool.

[0093] The second preset condition refers to the set of verification rules that must be met after a node is added, used to ensure the legality of the new node. The second preset condition includes at least one or more of the following verification rules: the node identifier must not be duplicated with an existing node; the node type must belong to the preset device type set; and the node attribute fields must conform to the fields and value ranges supported by the platform. The second preset condition is jointly determined by the network diagram specifications, device type constraints, and the node and link attributes supported by the platform in the topology constraint information.

[0094] Specifically, each time the system calls the node addition tool, it iterates through the node set of the first topology state object for the added node, checking if the node identifier of the new node already exists in the node set. If a duplicate exists, the check fails. Additionally, it compares the device type of the new node with a preset set of device types to determine if the device type is one of the system's supported device types. If the device type is not in the preset set of device types, the check fails.

[0095] Additionally, the attribute fields of newly added nodes are compared with the set of node attribute fields supported by the platform to determine whether the attribute fields are fields that the platform can recognize and whether the attribute values ​​are within the allowed value range. Supported node attributes include mirror, IP address, group information, security domain, etc. If an attribute field is not recognized by the platform or its value exceeds the allowed range, the check fails.

[0096] In the implementation of this invention, when the node identifier, device type and node attributes of the newly added node all meet the second preset conditions, the first topology state object is merged with the newly added node to construct a second topology state object. The second topology state object is the updated state after adding a valid new node to the node set based on the first topology state object.

[0097] For example, when a user requests "add a branch router to the existing topology", the processing flow for the node addition task is as follows: The system calls the node addition tool to add a branch router node to the first topology state object; then it checks whether the node's identifier is duplicated with an existing node, whether the device type "router" belongs to the preset device type set, and whether the attribute fields are valid; after the verification is passed, the first topology state object is merged with the newly added node to construct the second topology state object.

[0098] The verification process in this embodiment of the invention is performed immediately after the atomic operation tool is executed, so that errors can be detected and corrected in a timely manner during the node addition stage, without the need for an overall check after the complete topology is generated, thus improving the reliability and efficiency of topology construction.

[0099] In some optional implementations, when at least one topology orchestration task corresponding to a topology orchestration request includes a node batch addition task, the first topology state object can be orchestrated in the following manner.

[0100] Specifically, first, the node list in the topology feature information is obtained; then, each node in the node list is grouped and identified to obtain at least one isomorphic node group; in response to the identification of the isomorphic node group, the node batch addition tool is called to create all the nodes contained in the isomorphic node group in the first topology state object to obtain the second topology state object.

[0101] The node list refers to the complete list of nodes to be created, extracted from the topology feature information. This node list contains all node information involved in the orchestration request, such as the device type, quantity, mirror, IP address range, and grouping information for each node. The system reads and extracts the node list from the topology feature information so that subsequent operations such as grouping identification and batch creation can be performed based on this node list.

[0102] A homogeneous node group refers to a set of nodes in a node list that meet the homogeneity criteria. Specifically, if multiple nodes simultaneously meet at least two of the following conditions: they have the same node type, the same mirror identifier or specification parameters, the number of nodes exceeds a preset threshold, and they have a continuous range of IP addresses, then the multiple nodes are determined to belong to the same homogeneous node group. The preset threshold is typically set to 2.

[0103] The system identifies and groups nodes in the node list, grouping nodes that meet the homogeneity criteria into the same homogeneous node group. For example, it iterates through the nodes in the node list, clustering them according to device type and image or specification information, grouping nodes with the same device type and the same image or specification into one group; for nodes grouped together, it further checks if the number of nodes in a group is greater than 1; if it is, the group is determined to be a homogeneous node group. Through group identification, the nodes in the node list can be divided into at least one homogeneous node group, as well as any possible single heterogeneous node.

[0104] The Batch Node Addition Tool refers to a tool in the Atomic Operations Tool Library used to create multiple homogeneous nodes at once. Taking a homogeneous node group as the input unit, the tool creates all nodes contained in the homogeneous node group at once, based on node templates within that group, such as device type, image, specifications, naming rules, and IP address range. These nodes are then uniformly added to the node set of the first topology state object.

[0105] After identifying a homogeneous node group, the system uses the node template information and quantity of that group as input parameters for the batch node addition tool, and then calls the tool to perform batch creation. During execution, the batch node addition tool generates node identifiers and node attributes for all nodes in the homogeneous node group according to the node template and quantity, and writes them all at once into the node set of the first topology state object, thus obtaining the second topology state object.

[0106] In some optional implementations, when at least one topology orchestration task corresponding to the topology orchestration request includes an edge addition task, the first topology state object can be orchestrated in the following manner to obtain the second topology state object.

[0107] First, determine whether the first node and the second node exist, and whether there is a link between the first node and the second node. If there is no link between the first node and the second node, call the node adding tool to add the first node or the second node to the first topology state object. Then, call the edge adding task to add a first link between the first node and the second node. Next, based on the first link and the first topology state object, construct the second topology state object.

[0108] The first node and the second node refer to the two endpoint nodes connected by the new link.

[0109] Specifically, after receiving an edge addition task, the system searches the node set of the first topology state object based on the first node identifier and the second node identifier carried in the task parameters to determine whether the first node and the second node already exist in the current topology state. Simultaneously, the system searches the link set of the first topology state object to determine whether a link already exists between the first node and the second node.

[0110] If a node matching the first node identifier (or second node identifier) ​​exists in the node set of the first topology state object, then the first node (or second node) is determined to exist; otherwise, the first node (or second node) is determined not to exist. If a link with the first node as the source node and the second node as the destination node (or vice versa) exists in the link set of the first topology state object, then a link between the first node and the second node is determined to exist; otherwise, a link between them is determined not to exist.

[0111] If the first node does not exist, the system calls the node adding tool to create the first node in the first topology state object; if the second node does not exist, the system calls the node adding tool to create the second node in the first topology state object; if neither exists, the system calls the node adding tool to create the first node and the second node in turn.

[0112] During the node completion process, the system can also use the node information carried in the edge addition task (such as node identifier, device type, node attributes, etc.) as input parameters for the node addition tool. After ensuring that the newly added node meets the second preset condition, the completed node is added to the node set of the first topology state object.

[0113] The edge addition tool is used to add a link between two existing nodes. After confirming that the first node and the second node both exist in the topology object, the system calls the edge addition tool, using the identifiers of the first node and the second node as input parameters, to add a link record to the link set of the first topology state object. This newly added link is the first link.

[0114] After the first link is successfully added, the system merges the first topology state object with the first link to construct the second topology state object. The second topology state object is an updated state based on the first topology state object, with the first link added to the link set.

[0115] If both the first node and the second node exist, and there is no link between the first node and the second node, then the edge addition task is called to add a first link between the first node and the second node; then, based on the first link and the first topology state object, the second topology state object is constructed.

[0116] In this embodiment of the invention, by determining whether endpoint nodes exist before executing the edge addition task and automatically calling the node addition tool to complete the task when endpoint nodes are missing, the problem of edge addition failure due to missing endpoints is reduced. Users do not need to manually create nodes one by one and then establish connections, thus reducing the complexity of user operations. In addition, checking whether links already exist between nodes before adding links reduces the generation of duplicate links and ensures the consistency and legality of the topology.

[0117] In some optional implementations, the state verification and self-correction module performs constraint checks on each modification. If a node is deleted, its associated links are automatically deleted; if a newly added link endpoint does not exist, an error is returned, prompting the model to either add the node or use the correct node identifier. Additionally, a second topology state object can be output after modification, retaining records of the differences before and after the modification for platform display, manual review, or generation of deployment change orders.

[0118] In existing natural language-driven topology generation solutions, system prompts only provide general role descriptions and do not inject network topology diagram specifications, domain terminology, device specifications, tool prerequisites, current topology status summary, platform field constraints, available mirror retrieval rules, constraint explicitness rules, and unified intermediate templates into the model context. This makes it difficult for the model to reliably determine whether to create the topology first, create nodes in batches, add links, or perform local modifications. The model output is usually a free text topology file or an unstructured configuration, which cannot be reliably parsed and executed by the system.

[0119] To this end, embodiments of the present invention employ a multi-component system prompt word strategy and a structured task planning mechanism to ensure that the first intelligent agent is subject to the joint constraints of knowledge domain, topological constraints, tool contracts, and verification rules when reading user natural language requirements. This transforms the output of the first intelligent agent from free text generation to tool invocation decisions constrained by system prompt words.

[0120] Specifically, before transmitting user requests to the first intelligent agent, the system preloads multiple components in the natural language intent parsing module. Each component provides different types of constraint information, and these constraint information are collectively injected into the system prompts of the first intelligent agent.

[0121] The natural language intent parsing module may include any one or more of the following components: domain knowledge injection component, few-sample parsing component, constraint explicitation component, ambiguity correction component, and mirror retrieval suggestion component.

[0122] The domain knowledge injection component provides network topology-related terms such as network zones, isolation zones, virtual LANs, security domains, routers, firewalls, switches, servers, workstations, images, and device specifications, along with their field meanings on the target platform. This enables the model to map non-standardized expressions from the user's natural language into platform-recognizable structured parameters.

[0123] The Few Samples Parsing Component is used to provide "user needs-structured task planning" mapping examples for typical scenarios such as office networks, campus networks, and industrial control networks, enabling the model to learn and understand the topology construction paradigm in different scenarios through few samples.

[0124] The constraint explicitization component is used to convert network design rules that users may omit into explicit constraints that the model must adhere to. Through the constraint explicitization component, the user's implicit design intent is transformed into hard constraints that the model cannot violate during task planning.

[0125] The ambiguity correction component is used to identify low-precision expressions in the user's natural language and generate confirmation information or default inference information. Low-precision expressions may include ambiguous terms such as multiple units, secure connection, and isolation.

[0126] The mirror search suggestion component is used to require large language models to generate mirror query objects when there is no explicit mirror or when the mirror may depend on industry scenarios, rather than arbitrarily creating mirror names based on static knowledge.

[0127] The unified intermediate template component specifies that the model can only generate structured task plans or tool call objects, and should not output free text topology files. The unified intermediate template component defines the data structure for structured task planning, including task type fields, task parameter fields, and task dependency fields, and forces the model's output format to conform to a system-parsable specification.

[0128] In practical applications, the system can input the user's natural language requirements and the system prompts composed of the above components into the first agent, so that the first agent can perform semantic understanding and task planning under the constraints of the system prompts.

[0129] Under the above constraints, the first intelligent agent can identify topology tags (such as topology name, project to which the topology belongs), protocol type (such as border network management protocol), device type (such as router, switch, firewall), number of devices, connection relationship (such as point-to-point, star, dual-homed), redundancy requirements (such as link redundancy, device redundancy), mirror information, IP address range and batch creation prompts from user requirements, and form a structured task plan based on the identification results.

[0130] The structured task planning described above can include at least a sequence of topology orchestration tasks, with each topology orchestration task arranged in a dependent execution order within the task sequence. For example, the first task in the structured task planning is a topology creation task, followed by node creation tasks and link creation tasks, where the link creation task depends on the prior completion of the node creation tasks corresponding to its endpoint nodes.

[0131] like Figure 4 The diagram shown is an architecture diagram of a network topology generation system provided in an embodiment of the present invention.

[0132] The system includes a natural language intent parsing module, a topology orchestration module, an atomic operation toolset, a state verification and self-correction module, a topology state management module, and a topology state serialization module.

[0133] The Natural Language Intent Parsing module performs semantic parsing on the natural language input from the user, identifies topological feature information and operation patterns, and passes the parsing results to the Topology Orchestration module.

[0134] After receiving the topology feature information and operation mode from the natural language intent parsing module, the topology orchestration module determines the topology orchestration task based on the mode information of the operation mode and breaks down each topology orchestration task into atomic operation steps.

[0135] The topology state management module maintains an internal topology state object on the server. This object stores the currently confirmed set of nodes, the set of links, topology labels, protocol types, version information, node attributes, and link attributes. All changes to the topology are first applied to this topology state object, rather than being directly written to the final file.

[0136] During the process of the topology orchestration module calling the atomic operation tool, after each atomic operation tool call is completed, the status verification and self-correction module performs constraint checks on the changed topology status.

[0137] If the check passes, the change takes effect, and the topology state object in the topology state management module is updated. If the check fails, the state verification and self-correction module returns a structured error reason, which is fed back to the topology orchestration module. The topology orchestration module triggers corresponding self-correction operations based on the error reason (such as adding missing nodes, correcting tool call parameters, etc.), and then continues to call atomic operation tools for repair. The above-mentioned cycle of "calling atomic operation tools - state verification - updating state or feeding back errors" continues until the atomic operation tool calls stop.

[0138] Once all atomic operation tools corresponding to the topology orchestration tasks have been invoked and verified, the topology state object in the topology state management module is passed to the topology state serialization module.

[0139] The topology state serialization module serializes topology state objects, converts them into target topology structures that the platform can consume, and outputs them for subsequent visualization rendering, storage, or deployment.

[0140] The processing flow of the state verification and self-correction module can include steps such as calling the large model generation tool, executing atomic operations by the tool dispatcher, verifying topological state constraints, and returning structured results.

[0141] Under the constraints of the system prompt words, the large model (i.e. the first agent) generates the corresponding atomic operation tool call request according to the current topology orchestration task. The request includes the tool name and tool call parameters, and the tool call request is passed to the tool dispatcher.

[0142] After receiving a tool invocation request generated from a large model, the tool dispatcher invokes the corresponding atomic operation tool from the atomic operation toolkit based on the tool name, and passes the tool invocation parameters to the tool. The atomic operation tool then performs the specific topology change operation, modifying the topology state objects maintained by the topology state management module.

[0143] After each atomic operation tool is executed, the state verification and self-correction module performs constraint verification on the changed topology state object, then generates structured return information based on the verification results, and feeds this information back to the first intelligent agent.

[0144] Through the aforementioned state verification and self-correction closed-loop mechanism, the system verifies the legality of the topology state immediately after each atomic operation, ensuring that only valid state changes can be included in the final topology file. Simultaneously, a structured error feedback mechanism enables the first agent to specifically correct erroneous steps, rather than regenerating the entire topology, effectively improving the success rate and efficiency of topology construction.

[0145] This embodiment also provides an apparatus for orchestrating and adjusting the interactive network topology of infrastructure. This apparatus is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0146] This embodiment provides an apparatus for orchestrating and adjusting the interactive network topology of infrastructure, such as... Figure 5 As shown, it includes: The parsing module 501 is used to respond to the orchestration request triggered by the target object for the topology structure, perform semantic parsing on the orchestration request, and obtain the topology feature information and the operation mode corresponding to the orchestration request. The acquisition module 502 is used to acquire the first topology state object corresponding to the operation mode.

[0147] The determination module 503 is used to determine at least one topology orchestration task corresponding to the topology orchestration request based on topology feature information and mode information of operation mode.

[0148] The orchestration processing module 504 is used to call the atomic operation tools corresponding to each topology orchestration task to perform orchestration processing on the first topology state object to obtain the second topology state object.

[0149] The serialization processing module 505 is used to serialize the second topology state object to obtain the target topology structure corresponding to the orchestration request.

[0150] In one optional implementation, the acquisition module 502 includes: The initialization unit is used to initialize an empty topology state object if the operation mode is build mode; the build mode is used to characterize the generation of topology structures based on orchestration requests.

[0151] The first determining unit is used to determine the first topology state object corresponding to the initial empty topology state object construction mode.

[0152] The read unit is used to read the initial topology structure corresponding to the orchestration request from the target database if the operation mode is modification mode; the modification mode is used to indicate that the saved topology structure is modified.

[0153] The deserialization processing unit is used to deserialize the initial topology structure to obtain the first topology state object corresponding to the modified mode.

[0154] In one alternative implementation, the orchestration processing module 504 includes: The first acquisition unit is used to acquire topological constraint information; the topological constraint information is used to constrain the tool call parameters corresponding to the atomic operation tool to meet the first preset condition.

[0155] The second acquisition unit is used to load the topology constraint information into the first agent and acquire the tool invocation context. The first agent is used to decompose the topology orchestration task into multiple atomic operation steps with dependent execution order based on the topology constraint information, and determine the atomic operation tool and tool invocation parameters corresponding to each atomic operation step.

[0156] The calling unit is used to utilize the tool calling context to call the atomic operation tools corresponding to each topology orchestration task, perform orchestration processing on the first topology state object, and obtain the second topology state object.

[0157] In one alternative implementation, the orchestration processing module 504 includes: The first addition unit is used to call the node addition tool corresponding to the node addition task and add the first node to the first topology state object.

[0158] The inspection unit is used to check whether the node identifier, device type and node attributes of the first node meet the second preset conditions.

[0159] The construction unit is used to construct a second topology state object based on the first topology state object and the first node if the node identifier, device type and node attributes of the first node all meet the second preset conditions.

[0160] In one optional implementation, the orchestration processing module 504 further includes: The third acquisition unit is used to acquire the list of nodes in the topological feature information.

[0161] The identification unit is used to group and identify each node in the node list to obtain at least one homogeneous node group.

[0162] The creation unit is used to respond to the identification of a homogeneous node group, call the node batch addition tool, create all the nodes contained in the homogeneous node group in the first topology state object, and obtain the second topology state object.

[0163] In one optional implementation, the identification unit includes: The sub-unit is determined when multiple nodes simultaneously meet at least two of the following conditions: they have the same node type, the same mirror identifier or specification parameters, the number of nodes is greater than a preset threshold, and they have a continuous range of IP addresses.

[0164] In one optional implementation, the orchestration processing module 504 further includes: The second determining unit is used to determine whether the first node and the second node exist, and whether there is a link between the first node and the second node.

[0165] The second adding unit is used to add the first or second node to the first topology state object by calling the node adding tool if there is no link between the first node and the second node.

[0166] The first newly added unit is used to call the edge to add tasks and add a first link between the first node and the second node.

[0167] The first building unit is used to build the second topology state object based on the first link and the first topology state object.

[0168] In one optional implementation, the orchestration processing module 504 further includes: The second newly added unit is used to call the edge addition task to add a first link between the first node and the second node if both the first node and the second node exist and there is no link between the first node and the second node.

[0169] The second building unit is used to build a second topology state object based on the first link and the first topology state object.

[0170] The infrastructure interactive network topology orchestration and adjustment apparatus provided in this embodiment of the invention can execute the infrastructure interactive network topology orchestration and adjustment method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0171] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0172] The following is a detailed reference. Figure 6 This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory 602 or a program loaded from a memory 608 into a random access memory 603. The random access memory 603 also stores various programs and data required for the operation of the electronic device. The processor 601, the read-only memory 602, and the random access memory 603 are interconnected via a bus 604. An input / output interface 605 is also connected to the bus 604.

[0173] Typically, the following devices can be connected to the input / output interface 605: input devices 606 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; memory devices 608 including, for example, magnetic tape, hard disk, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0174] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a read-only memory 602. When the computer program is executed by the processor 601, it performs the functions defined in the infrastructure interactive network topology orchestration and adjustment method of the embodiments of the present invention.

[0175] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0176] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the infrastructure interactive network topology orchestration and adjustment method shown in the above embodiments is implemented.

[0177] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0178] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for orchestrating and adjusting the topology of an interactive infrastructure network, characterized in that, The method includes: In response to an orchestration request triggered by a target object for a topology structure, the orchestration request is semantically parsed to obtain topology feature information and the operation mode corresponding to the orchestration request; Obtain the first topology state object corresponding to the operation mode; Based on the topology feature information and the mode information of the operation mode, at least one topology orchestration task corresponding to the topology orchestration request is determined. The atomic operation tools corresponding to each of the topology orchestration tasks are invoked to orchestrate the first topology state object, thereby obtaining the second topology state object; The second topology state object is serialized to obtain the target topology structure corresponding to the orchestration request.

2. The method according to claim 1, characterized in that, The step of obtaining the first topology state object corresponding to the operation mode includes: If the operation mode is the build mode, then an empty topology state object is initialized; the build mode is used to characterize the generation of topology structure based on orchestration requests. The initialized empty topology state object is used to determine the first topology state object corresponding to the construction mode; If the operation mode is modification mode, the initial topology corresponding to the orchestration request is read from the target database; the modification mode is used to indicate that the saved topology is modified. The initial topology is deserialized to obtain the first topology state object corresponding to the modified mode.

3. The method according to claim 1 or 2, characterized in that, The step of invoking the atomic operation tools corresponding to each of the topology orchestration tasks to perform orchestration processing on the first topology state object to obtain the second topology state object includes: Obtain topological constraint information; the topological constraint information is used to constrain the tool call parameters corresponding to the atomic operation tool to meet a first preset condition; The topology constraint information is loaded into the first agent, and the tool invocation context is obtained. The first agent is used to decompose the topology orchestration task into multiple atomic operation steps with dependent execution order according to the topology constraint information, and determine the atomic operation tool and tool invocation parameters corresponding to each atomic operation step. Using the tool invocation context, the atomic operation tools corresponding to each of the topology orchestration tasks are invoked to orchestrate the first topology state object, thereby obtaining the second topology state object.

4. The method according to claim 1, characterized in that, The topology orchestration task includes a node addition task. The step of calling the atomic operation tools corresponding to each of the topology orchestration tasks to orchestrate the first topology state object and obtain a second topology state object includes: Invoke the node addition tool corresponding to the node addition task to add the first node in the first topology state object; For the first node, check whether the node identifier, device type, and node attributes of the first node meet the second preset condition; If the node identifier, device type, and node attributes of the first node all meet the second preset conditions, then the second topology state object is constructed based on the first topology state object and the first node.

5. The method according to claim 4, characterized in that, The topology orchestration task also includes a batch node addition task. The step of calling the atomic operation tools corresponding to each of the topology orchestration tasks to orchestrate the first topology state object and obtain a second topology state object includes: Obtain the list of nodes from the topological feature information; The nodes in the node list are grouped and identified to obtain at least one homogeneous node group; In response to the identification of the homogeneous node group, the node batch addition tool is invoked to create all the nodes contained in the homogeneous node group in the first topology state object, thereby obtaining the second topology state object.

6. The method according to claim 5, characterized in that, The step of grouping and identifying the nodes in the node list to obtain at least one homogeneous node group includes: If multiple nodes simultaneously meet at least two of the following conditions: they have the same node type, the same mirror identifier or specification parameters, the number of nodes is greater than a preset threshold, or they have a continuous range of IP addresses, then the multiple nodes are determined to belong to the same homogeneous node group.

7. The method according to claim 4 or 5, characterized in that, The topology orchestration task also includes an edge addition task, which is used to add a new link between the first node and the second node. The step of calling the atomic operation tools corresponding to each of the topology orchestration tasks to orchestrate the first topology state object and obtain the second topology state object includes: Determine whether the first node and the second node exist, and whether there is a link between the first node and the second node; If there is no link between the first node and the second node, the node adding tool is invoked to add the first node or the second node to the first topology state object; The task is added by calling the edge to create a new first link between the first node and the second node; Based on the first link and the first topology state object, a second topology state object is constructed.

8. The method according to claim 7, characterized in that, The method further includes: If both the first node and the second node exist, and there is no link between the first node and the second node, then the edge addition task is invoked to add the first link between the first node and the second node; Based on the first link and the first topology state object, a second topology state object is constructed.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the method for orchestrating and adjusting the infrastructure interactive network topology as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method for orchestrating and adjusting the infrastructure interactive network topology as described in any one of claims 1 to 8.

Images