Cloud network connectivity detection method, electronic equipment, storage medium and program product

Abstract

The invention provides a cloud network connectivity detection method, electronic equipment, a storage medium and a program product, and relates to the technical field of cloud networks, the method comprises the following steps: obtaining problem information for a cloud network, and identifying the problem information; if it is determined that connectivity detection needs to be carried out according to the recognition result, a detection starting point and a detection ending point are determined from the recognition result; determining a detection path between the detection start point and the detection end point based on the cloud network knowledge graph; the cloud network knowledge graph represents the dynamic change of the association relationship between the resources in the cloud network; and performing connectivity detection in the cloud network according to the detection path to obtain a detection result. In the embodiment of the invention, as the cloud network knowledge graph represents the dynamic change of the association relationship between the resources in the cloud network, the connectivity detection based on the cloud network knowledge graph can improve the real-time performance and accuracy of the connectivity detection of the cloud network, thereby adapting to the dynamic change of the cloud network environment.

Description

Technical Field

[0001] The present application relates to the field of cloud network technology, and in particular to a cloud network connectivity detection method, electronic device, storage medium and program product. Background Art

[0002] With the rapid development of the Internet and cloud computing, cloud network connectivity analysis has become an increasingly important research field. Cloud network connectivity refers to the ability of different computing resources, services, and users to communicate and interact with each other in a cloud computing environment. In the current cloud computing environment, network connectivity analysis faces many challenges.

[0003] In related technologies, connectivity analysis relies on rule engines and static network data to determine the connectivity between nodes, and cannot dynamically respond to changes in the network environment. As a result, when the network topology changes or failures occur, the analysis results cannot be updated in time, and accurate connectivity information cannot be provided to users in a timely manner. Summary of the invention

[0004] The embodiments of the present application provide a cloud network connectivity detection method, electronic device, storage medium and program product to improve the real-time and accuracy of cloud network connectivity detection, thereby adapting to dynamic changes in the cloud network environment.

[0005] In a first aspect, an embodiment of the present application provides a cloud network connectivity detection method, including: Obtain problem information about the cloud network and identify the problem information; If it is determined according to the recognition result that a connectivity test is required, a test start point and a test end point are determined from the recognition result; Based on the cloud network knowledge graph, the detection path between the detection starting point and the detection end point is determined; the cloud network knowledge graph represents the dynamic changes of the relationship between resources in the cloud network and the resources; According to the detection path, connectivity detection is performed in the cloud network to obtain the detection result.

[0006] In a second aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor implements any of the above methods when executing the computer program.

[0007] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, any of the above methods is implemented.

[0008] In a fourth aspect, an embodiment of the present application provides a computer program product, the computer program product including a computer program, and the computer program implements any of the above methods when executed by a processor.

[0009] Compared with the prior art, this application has the following advantages: The present application provides a cloud network connectivity detection method, electronic device, storage medium and program product. First, problem information for the cloud network is obtained and the problem information is identified. If it is determined that a connectivity detection is required based on the identification result, the detection starting point and the detection end point are determined from the identification result. Then, based on the cloud network knowledge graph, the detection path between the detection starting point and the detection end point is determined. The cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network. According to the detection path, a connectivity detection is performed in the cloud network to obtain the detection result. In the embodiment of the present application, the detection starting point and the detection end point are determined by identifying the problem information, and the detection path between the detection starting point and the detection end point is determined based on the cloud network knowledge graph. Since the cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network, connectivity detection based on the cloud network knowledge graph can improve the real-time and accuracy of cloud network connectivity detection, thereby adapting to the dynamic changes of the cloud network environment.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments according to the present application and should not be regarded as limiting the scope of the present application.

[0012] Figure 1 A schematic diagram of an application scenario of a cloud network connectivity detection method according to an embodiment of the present application.

[0013] Figure 2 This is a flow chart of a cloud network connectivity detection method according to an embodiment of the present application.

[0014] Figure 3 This is a flow chart of a cloud network connectivity detection method according to an embodiment of the present application.

[0015] Figure 4 This is a structural block diagram of a cloud network connectivity detection device according to an embodiment of the present application.

[0016] Figure 5 A block diagram of an electronic device used to implement an embodiment of the present application. DETAILED DESCRIPTION

[0017] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the concept or scope of the present application. Therefore, the drawings and descriptions are considered to be exemplary in nature and not restrictive.

[0018] To facilitate understanding of the technical solutions of the embodiments of the present application, the following describes the related technologies of the embodiments of the present application. The following related technologies can be combined with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application.

[0019] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0020] The Large Language Model (LLM) of one or more embodiments of the present application can be a natural language processing model based on deep learning, containing more than one billion parameters, and the underlying converter of the large language model contains a series of neural networks. In one or more embodiments, an encoder and / or a decoder may be included, and have a self-attention function. These two types of modules extract meaning from the input text and clarify the structural relationship between context words. The model is pre-trained on a public dataset containing more than 1TB of text data to learn general language representations.

[0021] In order to adapt to the specific application scenario of the present invention - cloud network fault handling (for example, cloud network connectivity detection), we fine-tuned the large language model on a dataset containing thousands of cloud network fault information and solutions. The large language model receives the problem information about the cloud network as input, performs entity recognition, intent recognition, root cause location, and generates solutions as reply information output.

[0022] Figure 1A schematic diagram of an application scenario of the cloud network connectivity detection method provided in this application. In this embodiment, the agent is an intelligent system that can perceive data / tools, make decisions based on the context, and take actions to achieve specific goals. The agent in this embodiment can guide the LLM to reflect on itself, break down complex problems into multiple simple small problems, and improve the ability of the large language model to solve complex problems.

[0023] Pre-construct the cloud network knowledge graph, which is constructed in the following ways: based on the cloud network topology, map the resources in the cloud network (e.g., virtual machines, databases, storage devices, load balancers, and firewalls, etc.) as nodes in the cloud network knowledge graph, and map the associations between resources (e.g., subnets, routers, and network security policies, etc.) as edges in the cloud network knowledge graph; determine the attribute information of nodes and edges based on the real-time status data, fault handling data, and configuration data of the cloud network. Among them, the real-time status data includes the real-time integration of dynamic monitoring data of the cloud network, such as traffic usage, network latency, and fault information, to ensure that the data in the cloud network knowledge graph is always up to date, thereby further improving the accuracy and reliability of connectivity analysis. Fault handling data includes: configuration suggestions to avoid faults, historical records of fault cases, etc., such as fault cause analysis and effective solutions, which can provide users with intelligent guidance and help users make better decisions when facing complex network problems. Configuration data includes: detailed configuration information of resources, such as virtual private cloud (VPC) settings, subnet division, Internet Protocol (IP) address allocation, routing tables and bandwidth settings, etc., to ensure that the network environment reflected in the cloud network knowledge graph is consistent with the actual operating status.

[0024] The specific processing process of the cloud network connectivity detection method provided by this application is as follows: First, the interaction layer receives the problem information sent by the user terminal, for example, "Why VM1 cannot access the Internet". The interaction layer can be a user interaction interface.

[0025] Secondly, the intelligent agent is used to format the problem information, and after processing it into a preset format, it is input into the LLM for problem recognition, including: entity recognition and intent recognition. For example, the entity recognition results are "VM1" and "Internet", and the intent recognition result is "check connectivity". "VM1" is used as the detection starting point, and "Internet" is used as the detection end point.

[0026] Then, the cloud network knowledge graph is used to perform path identification. Specifically, based on the nodes corresponding to the detection starting point and the detection end point in the cloud network knowledge graph, and the edges between the nodes, the possible path between the detection starting point and the detection end point is determined. For example, the possible path is "virtual machine 1VM1->subnetSubnet->...->InternetInternet". According to the attribute information of the nodes in the possible path (for example, IP address, subnet, etc.) and the attribute information of the edges between the nodes, the detection path is determined in the possible path. For example, the detection path is "virtual machine 1VM1->virtual switch 1VS1->...->high power server HPS->InternetInternet".

[0027] After identifying the detection path from "VM1" to "Internet", perform network detection on the detection path, that is, perform connectivity detection. Specifically, prepare necessary detection tools, such as Internet Control Message Protocol (ICMP) requests (such as Ping requests), or Transmission Control Protocol (TCP) connection test requests, or application layer protocol detection requests, such as HyperText Transfer Protocol (HTTP) requests. These detection tools can effectively evaluate the actual connectivity from "VM1" to "Internet" and ensure the accuracy of the detection results. The specific type of request to be sent can be determined based on the attribute information of each node and edge in the detection path or user selection.

[0028] In actual detection, first determine at least one target node in the detection path, the target node includes but is not limited to a network gateway, a DNS server or a final Internet target (such as a preset web page service), send a connection request between the starting node and the target node, and establish a communication link between the starting node and the target node; based on the communication link, send a detection data packet to obtain the detection result. During the detection process, record the request response time, packet loss rate, and return status. If it is determined based on the real-time status data of the cloud network that the communication link cannot send the detection data packet, establish a backup communication link between the backup nodes corresponding to the starting node and the target node; based on the backup communication link, send a detection data packet to obtain the detection result. Specifically, monitor the network status during the detection process and perform dynamic data updates (such as traffic load, network delay and other real-time parameters). If it is found during the detection process that the real-time conditions of the current path are insufficient to support the connection, quickly switch to other potential detection paths, such as using redundant network services or backup gateways.

[0029] After the connectivity test is completed, the test results are summarized and analyzed, including: Connectivity status evaluation: Based on the response results of each test request, determine whether "VM1" is connected to "Internet". If an intermediate node (such as a gateway) is not connected, determine the fault location, for example, the fault location is a disconnected node. Performance evaluation: In addition to connectivity, test performance indicators such as response time and packet loss rate are also analyzed. The test performance indicators provide users with a deeper perspective on network performance and help identify network bottlenecks or latency issues.

[0030] If the detection results show that there is a problem with the connection from "VM1" to "Internet", the fault location and detection performance indicators are input into the large language model, and the large language model identifies the fault object and fault characteristics, such as high response time, increased packet loss rate, or complete disconnection. Based on the fault object and fault characteristics, fault knowledge retrieval is performed in the fault knowledge base of the cloud network knowledge graph (that is, the fault processing data in the attribute information of nodes and edges in the cloud network knowledge graph) to retrieve matching fault information; using the large language model, root cause location and solution are determined based on the matching fault information. Specifically, LLM uses machine learning algorithms to classify fault characteristics and compare them with the fault processing data stored in the fault knowledge base to ensure rapid location of similar fault cases. When searching for relevant fault information in the fault knowledge base, the contextual information of the problem information can also be considered to improve the accuracy of the recommended solution. LLM infers the possible root cause of the fault based on the matching fault information.

[0031] After the root cause is located, a solution recommendation for the current fault is generated. The solution can be determined based on the practical experience and successful cases recorded in the fault knowledge base. Finally, the agent performs role-playing, plays the role of a problem-solving expert, generates a detailed problem report with the root cause and solution, and sends it to the user terminal as a reply message so that the user can fully understand the network problem and its solution.

[0032] In the embodiment of the present application, a large language model and a cloud network knowledge graph are combined to optimize cloud network connectivity detection. The powerful semantic understanding and generation capabilities of the large language model enable it to process user question information and provide intelligent answers. At the same time, the cloud network knowledge graph establishes a dynamic representation of network resources and connection status, which can reflect changes in the network environment in real time. By combining the two, dynamic analysis of complex network status is achieved, which can not only quickly respond to topology changes, but also provide users with easy-to-understand connectivity information and problem solutions, making it easier for ordinary users to manage and optimize their networks, and improving the intelligence and flexibility of cloud network management.

[0033] An embodiment of the present application provides a cloud network connectivity detection method. The method in this embodiment can be applied to servers, terminal devices, platforms, devices, etc. with computing and processing capabilities, wherein the server can be a server cluster or a single server, a server deployed in the cloud or a local server.

[0034] like Figure 2 The flowchart of a cloud network connectivity detection method according to an embodiment of the present application is shown, including: Step S201, obtaining problem information about the cloud network and identifying the problem information.

[0035] In this embodiment, the execution subject is taken as an example to introduce the server. The server receives the question information sent by the user terminal, identifies the question information, and obtains the identification result.

[0036] Step S202: If it is determined according to the recognition result that a connectivity test is required, a test start point and a test end point are determined from the recognition result.

[0037] Determine whether it is a cloud network connectivity problem based on the identification results. If so, a connectivity test is required. Connectivity refers to the access capability from one network node to another in a cloud network environment, which usually involves analysis of the path and connection status.

[0038] If a connectivity test is required, the network nodes that need to be tested for connectivity are extracted from the identification results of the problem information as the test start point and the test end point.

[0039] Step S203, based on the cloud network knowledge graph, determine the detection path between the detection starting point and the detection end point; the cloud network knowledge graph represents the dynamic changes of the relationship between resources and resources in the cloud network.

[0040] The resources in the cloud network include, but are not limited to, virtual machines, databases, storage devices, load balancers, and firewalls, and the relationships between resources include, but are not limited to, subnets, routers, and network security policies (e.g., whitelist policies).

[0041] The cloud network knowledge graph can automatically adjust according to changes in the cloud network, so that it can represent the dynamic changes in the relationship between resources and resources. For example, when a new device joins the network or an existing device leaves the network, the knowledge graph can automatically add or delete the corresponding node.

[0042] In the cloud network knowledge graph, the detection path can be obtained by querying from the detection starting point to the detection end point.

[0043] Step S204: Perform connectivity detection in the cloud network according to the detection path to obtain a detection result.

[0044] In addition to the detection starting point and the detection end point, the detection path can also include at least one intermediate node. In the cloud network, connectivity detection is performed one by one on each node in the detection path to obtain a detection result. The detection result can indicate whether each node in the detection path is connected. If not, it indicates that a fault may have occurred and fault handling is required.

[0045] The cloud network connectivity detection method provided in the embodiment of the present application first obtains the problem information for the cloud network and identifies the problem information; if it is determined that a connectivity detection is required based on the identification result, the detection starting point and the detection end point are determined from the identification result; then, based on the cloud network knowledge graph, the detection path between the detection starting point and the detection end point is determined; the cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network; according to the detection path, a connectivity detection is performed in the cloud network to obtain the detection result. In the embodiment of the present application, the detection starting point and the detection end point are determined by identifying the problem information, and the detection path between the detection starting point and the detection end point is determined based on the cloud network knowledge graph. Since the cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network, connectivity detection based on the cloud network knowledge graph can improve the real-time and accuracy of cloud network connectivity detection, thereby adapting to the dynamic changes of the cloud network environment.

[0046] The following introduces the specific implementation process of the above steps through multiple implementation methods: In one implementation, in step S201, the question information is identified, including: inputting the question information into a large language model, and using the large language model to perform entity recognition and intent recognition; in step S202, it is determined based on the recognition result that a connectivity check is required, and then a detection start point and a detection end point are determined from the recognition result, including: determining that a connectivity check is required based on the recognition result of intent recognition; and determining a detection start point and a detection end point based on the recognition result of entity recognition.

[0047] The problem information is identified using a large language model to identify entities and intents, and whether a connectivity test is needed is determined based on the intent. If so, the identified entities are determined as the detection start point and detection end point. The detection start point and detection end point can be resources in the cloud network, such as virtual machines, databases, storage devices, load balancers, and firewalls.

[0048] In the related art, the user intention recognition is not intelligent enough and the interaction is cumbersome because of the reliance on command line operation or complex interface. In this embodiment, the natural language processing capability based on the large language model is used to identify the question information, so that the user can directly ask questions in natural language, realize the intuitive interaction experience, greatly simplify the operation process, and improve the user experience.

[0049] In one implementation, the cloud network knowledge graph is constructed in the following manner: based on the cloud network topology structure, resources are mapped as nodes in the cloud network knowledge graph, and association relationships are mapped as edges in the cloud network knowledge graph; based on the real-time status data, fault handling data, and configuration data of the cloud network, the attribute information of the nodes and edges is determined.

[0050] In practical applications, a cloud network knowledge graph is pre-constructed, and the cloud network knowledge graph is constructed in the following way: First, based on the cloud network topology, the resources in the cloud network (for example, virtual machines, databases, storage devices, load balancers, and firewalls, etc.) are mapped as nodes in the cloud network knowledge graph, and the associations between resources (for example, subnets, routers, and network security policies, etc.) are mapped as edges in the cloud network knowledge graph. The cloud network topology structure can be a topology structure composed of resources in the cloud network and the physical connections between resources.

[0051] Then, based on the real-time status data, fault handling data, and configuration data of the cloud network, the attribute information of the nodes and edges is determined. Among them, the real-time status data includes the real-time integration of the dynamic monitoring data of the cloud network, such as traffic usage, network latency, and fault information, etc., to ensure that the data in the cloud network knowledge graph is always kept up to date, thereby further improving the accuracy and reliability of connectivity analysis. Fault handling data includes: configuration suggestions to avoid faults, historical records of fault cases, etc., such as fault cause analysis and effective solutions, etc., which can provide users with intelligent guidance and help users make better decisions when facing complex network problems. Configuration data includes: detailed configuration information of resources, such as VPC settings, subnet division, IP address allocation, routing tables, and bandwidth settings, etc., to ensure that the network environment reflected in the cloud network knowledge graph is consistent with the actual operating status.

[0052] In this embodiment, the cloud network knowledge graph integrates the cloud network topology, real-time status data, fault handling data and configuration data to ensure the consistency and accuracy of the data and can flexibly adapt to the dynamic cloud network environment.

[0053] In one implementation, step S203 determines the detection path between the detection starting point and the detection end point based on the cloud network knowledge graph, including: determining the possible path between the detection starting point and the detection end point based on the nodes corresponding to the detection starting point and the detection end point in the cloud network knowledge graph, and the edges between the nodes; determining the detection path in the possible path according to the attribute information of the nodes in the possible path and the attribute information of the edges between the nodes.

[0054] From the detection starting point to the detection end point, the corresponding nodes and edges are queried in the cloud network knowledge graph to obtain a possible path, namely the logical path. The actual detection path, namely the forwarding path, is further determined through the attribute information of the nodes and edges in the possible path.

[0055] For example, in the possible path L1, node A and node B can be connected. According to the relevant configuration in the attribute information of node A and node B, node A and node B are not connected, so the possible path L1 cannot be used as a detection path.

[0056] In this embodiment, the detection path is determined based on the cloud network knowledge graph. Since the cloud network knowledge graph can reflect the dynamic changes of the cloud network, the accuracy of determining the detection path can be improved.

[0057] In one implementation, step S204, performing connectivity detection in the cloud network according to the detection path to obtain a detection result, includes: step S2041, determining at least one target node in the detection path, sending a connection request between the starting node and at least one target node in the detection path, the connection request is used to establish a communication link between the starting node and at least one target node; step S2042, sending a detection data packet based on the communication link to obtain a detection result.

[0058] The target node includes but is not limited to a network gateway, a DNS server or a final Internet target (such as a preset web page service). A connection request is sent between the starting node and the target node to establish a communication link.

[0059] The connection request includes but is not limited to: Internet Control Message Protocol (ICMP) request (such as Ping request), or Transmission Control Protocol (TCP) connection test request, or application layer protocol detection request, such as Hypertext Transfer Protocol (HTTP) request. The specific type of request sent can be determined based on the attribute information of each node and edge in the detection path or user selection.

[0060] The detection results may include: whether the nodes are connected, the location of disconnection, and detection performance indicators such as request response time, packet loss rate and return status.

[0061] In this embodiment, by sending a request to establish a communication link and performing connectivity detection, the accuracy of the detection result can be ensured.

[0062] In one implementation, step S2042, based on the communication link, sends a detection data packet to obtain a detection result, including: if it is determined based on the real-time status data of the cloud network that the communication link cannot send a detection data packet, then establish a backup communication link between the backup nodes corresponding to the starting node and the target node; based on the backup communication link, send a detection data packet to obtain a detection result.

[0063] Specifically, during the detection process, the network status is monitored and dynamic data is updated, such as traffic load, network delay and other real-time parameters. The traffic load on each path is continuously monitored to ensure that there is no overload. The delays of different paths are measured regularly, and the optimal path is selected based on the results. If the detection process finds that the real-time conditions of the current path (for example, traffic overload or long network delay) are insufficient to support the connection, quickly switch to other potential detection paths, such as using redundant network services or backup gateways.

[0064] In this embodiment, when the real-time status data cannot satisfy the connectivity detection, the connectivity detection is performed based on the standby node, which can improve the detection efficiency and the accuracy of the detection result.

[0065] In one implementation, in step S204, connectivity detection is performed in the cloud network according to the detection path. After the detection result is obtained, the cloud network connectivity detection method further includes: if it is determined according to the detection result that there is no connectivity between the detection starting point and the detection end point, then the fault location and detection performance index are obtained from the detection result; based on the fault location and the detection performance index, the root cause of the problem and the solution are determined.

[0066] In actual applications, after the connectivity test is completed, the test results are summarized and analyzed, including: Connectivity status evaluation: Based on the response results of each test request, determine whether the test start point and the test end point are connected. If an intermediate node (such as a gateway) is not connected, determine the fault location, for example, the fault location is a disconnected node. Performance evaluation: In addition to connectivity, test performance indicators such as response time and packet loss rate are also analyzed. Based on the fault location and test performance indicators, analyze the problem, query the root cause of the problem and the solution.

[0067] In this embodiment, not only the connectivity test is performed, but also the root cause of the problem can be checked and an effective solution can be provided to the user.

[0068] In one implementation, the root cause and solution of the problem are determined based on the fault location and the detection performance index, including: inputting the fault location and the detection performance index into a large language model to determine the fault object and fault characteristics; and determining the root cause and solution of the problem based on the fault object and the fault characteristics.

[0069] If the test results show that there is a problem with the connection from the test start point to the test end point, the fault location and test performance indicators are input into the large language model, and the large language model is used to identify the fault object and fault characteristics, such as high response time, increased packet loss rate, or complete disconnection. According to the specific object of the fault and the characteristics of the fault, the root cause of the problem is located and a solution is obtained.

[0070] In this embodiment, the natural language understanding capability of a large language model is used to locate the root cause of the problem, which can accurately locate the root cause of the problem, reduce troubleshooting time, and improve the speed and accuracy of problem handling.

[0071] In one implementation, the root cause and solution of the problem are determined based on the fault object and fault characteristics, including: based on the fault object and fault characteristics, retrieving matching fault information in the fault knowledge base of the cloud network knowledge graph; using a large language model to determine the root cause and solution of the problem based on the matching fault information as reply information to the problem information.

[0072] Specifically, the large language model uses machine learning algorithms to classify fault features and compares them with the fault handling data stored in the fault knowledge base of the cloud network knowledge graph to ensure rapid location of similar fault cases. LLM infers possible root causes of faults based on matching fault information. For example, network configuration errors: such as incorrect routing settings and IP address conflicts. Hardware failures: such as unreachable due to damaged network equipment. Traffic overload: network bottlenecks that cause delayed response times.

[0073] After the root cause is located, generate a solution recommendation for the current fault. The solution can be determined based on the practical experience and success cases recorded in the fault knowledge base. For example, direct solutions: such as restarting related devices, reconfiguring routing, or increasing bandwidth. Long-term improvement measures: recommended actions for users, such as optimizing network configuration, adding redundant paths, or implementing network monitoring tools. Fault prevention measures: provide follow-up monitoring and management suggestions to prevent similar faults from happening again.

[0074] Finally, a detailed problem analysis report is generated with the root cause and solution of the problem and sent to the user terminal as a reply message so that the user can fully understand the network problem and its solution.

[0075] In one implementation, searching for matching fault information in a fault knowledge base based on the fault object and the fault feature also includes searching for matching fault information in the fault knowledge base based on the fault object, the fault feature and context information of the problem information.

[0076] Specifically, when searching for relevant fault information in the fault knowledge base, the context information of the problem information may also be considered. The context information may include information related to the specific details of the problem. When retrieving matching fault information, in addition to considering the fault object and fault characteristics, adding matching retrieval of context information can improve the accuracy of locating the root cause of the problem and recommending solutions.

[0077] An embodiment of the present application provides a cloud network connectivity detection method. The method in this embodiment can be applied to servers, terminal devices, platforms, devices, etc. with computing and processing capabilities, wherein the server can be a server cluster or a single server, a server deployed in the cloud or a local server.

[0078] like Figure 3 The flowchart of a cloud network connectivity detection method according to an embodiment of the present application is shown, including: Step S301, obtain question information for the cloud network, input the question information into a large language model, and use the large language model to perform entity recognition and intent recognition.

[0079] Step S302, determining the need for connectivity detection based on the recognition result of intent recognition, and determining the detection start point and detection end point based on the recognition result of entity recognition.

[0080] Step S303, based on the nodes corresponding to the detection starting point and the detection end point in the cloud network knowledge graph, and the edges between the nodes, determine the possible path between the detection starting point and the detection end point.

[0081] Step S304: determining a detection path in the possible paths according to the attribute information of the nodes in the possible paths and the attribute information of the edges between the nodes.

[0082] Step S305, determining at least one target node in the detection path, sending a connection request between the starting node and the at least one target node in the detection path, the connection request being used to establish a communication link between the starting node and the at least one target node.

[0083] Step S306: Send a detection data packet based on the communication link to obtain a detection result.

[0084] Step S307: If it is determined according to the detection result that the detection starting point and the detection end point are not connected, the fault location and the detection performance index are obtained from the detection result.

[0085] Step S308: input the fault location and detection performance index into the large language model to determine the fault object and fault characteristics.

[0086] Step S309: Based on the fault object and fault characteristics, matching fault information is retrieved from the fault knowledge base of the cloud network knowledge graph.

[0087] Step S310, using the large language model, according to the matched fault information, determine the root cause of the problem and the solution, and generate a problem analysis report as the reply information to the problem information.

[0088] The problem analysis report includes: Root cause of the problem: For example, network configuration error: such as incorrect routing settings, IP address conflicts. Hardware failure: such as unreachable due to damage to network equipment. Traffic overload: Network bottleneck that causes delayed response time. Solution: For example, direct solution: such as restarting related equipment, reconfiguring routing or increasing bandwidth. Long-term improvement measures: Recommended actions for users, such as optimizing network configuration, adding redundant paths or implementing network monitoring tools. Fault prevention measures: Provide follow-up monitoring and management suggestions to prevent similar faults from happening again.

[0089] The specific process of each step in the embodiment of the present application can be found in the corresponding description in the above method, which will not be repeated here.

[0090] In the embodiment of the present application, with the help of the natural language processing capabilities of LLM, the user's intentions are accurately identified, the response speed is improved, and a seamless user interaction experience is achieved. The cloud network knowledge graph is used to integrate the cloud network topology, real-time monitoring data and configuration data to ensure the real-time and accuracy of information and adapt to the dynamic changes in the cloud environment. Quickly identify fault characteristics, accurately locate the root cause of the problem, and significantly reduce the troubleshooting time based on the unique environmental conditions of the cloud network. Provide immediate repair measures and long-term optimization suggestions in the cloud environment to ensure the high availability and security of the cloud network. Generate easy-to-understand user reports, combined with cloud architecture characteristics, to help users quickly grasp problems and solutions, and facilitate subsequent maintenance and management. Corresponding to the application scenario and method of the method provided in the embodiment of the present application, the embodiment of the present application also provides a cloud network connectivity detection device. Figure 4 The figure is a structural block diagram of a cloud network connectivity detection device according to an embodiment of the present application, the device comprising: The problem identification module 401 is used to obtain problem information about the cloud network and identify the problem information.

[0091] The first determination module 402 is configured to determine a detection start point and a detection end point from the recognition result if it is determined that a connectivity detection is required according to the recognition result.

[0092] The second determination module 403 is used to determine the detection path between the detection start point and the detection end point based on the cloud network knowledge graph; the cloud network knowledge graph represents the dynamic changes of the relationship between resources and resources in the cloud network.

[0093] The connectivity detection module 404 is used to perform connectivity detection in the cloud network according to the detection path to obtain a detection result.

[0094] The cloud network connectivity detection device provided in the embodiment of the present application first obtains problem information for the cloud network and identifies the problem information; if it is determined that a connectivity detection is required based on the identification result, the detection starting point and the detection end point are determined from the identification result; then, based on the cloud network knowledge graph, the detection path between the detection starting point and the detection end point is determined; the cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network; according to the detection path, a connectivity detection is performed in the cloud network to obtain a detection result. In the embodiment of the present application, the detection starting point and the detection end point are determined by identifying the problem information, and the detection path between the detection starting point and the detection end point is determined based on the cloud network knowledge graph. Since the cloud network knowledge graph represents the dynamic changes of the association relationship between resources in the cloud network, connectivity detection based on the cloud network knowledge graph can improve the real-time and accuracy of cloud network connectivity detection, thereby adapting to the dynamic changes of the cloud network environment.

[0095] In one implementation, when identifying the problem information, the problem identification module 401 is used to: Input the question information into the large language model and use the large language model to perform entity recognition and intent recognition; If it is determined that a connectivity test is required based on the identification result, the detection start point and the detection end point are determined from the identification result, including: Determine the need for connectivity detection based on the recognition result of intent recognition; The detection starting point and the detection end point are determined according to the recognition result of the entity recognition.

[0096] In one implementation, the cloud network knowledge graph is constructed in the following way: Based on the cloud network topology, resources are mapped as nodes in the cloud network knowledge graph, and associations are mapped as edges in the cloud network knowledge graph; Determine the attribute information of nodes and edges based on the real-time status data, fault handling data, and configuration data of the cloud network.

[0097] In one implementation, the second determining module 403 is configured to: Based on the nodes corresponding to the detection starting point and the detection end point in the cloud network knowledge graph, and the edges between the nodes, determine the possible path between the detection starting point and the detection end point; A detection path is determined in the possible paths according to the attribute information of the nodes in the possible paths and the attribute information of the edges between the nodes.

[0098] In one implementation, the connectivity detection module 404 is configured to: Determine at least one target node in the detection path, send a connection request between the starting node and the at least one target node in the detection path, the connection request is used to establish a communication link between the starting node and the at least one target node; Based on the communication link, a detection data packet is sent to obtain the detection result.

[0099] In one implementation, when the connectivity detection module 404 sends a detection data packet based on the communication link and obtains the detection result, it is used to: If it is determined based on the real-time status data of the cloud network that the communication link cannot send the detection data packet, a backup communication link is established between the start node and the backup node corresponding to the target node; Based on the backup communication link, a detection data packet is sent to obtain a detection result.

[0100] In one implementation, the device is also used to: perform connectivity detection in the cloud network according to the detection path, and after obtaining the detection result, if it is determined according to the detection result that there is no connectivity between the detection starting point and the detection end point, then obtain the fault location and detection performance indicators from the detection result; based on the fault location and the detection performance indicators, determine the root cause of the problem and the solution.

[0101] In one implementation, the device, when determining the root cause and solution of the problem based on the fault location and the detection performance indicator, is used to: Input the fault location and detection performance index into the large language model to determine the fault object and fault characteristics; Determine the root cause and solution based on the fault object and fault characteristics.

[0102] In one implementation, when determining the root cause and solution of a problem based on a fault object and fault characteristics, the device is configured to: Based on the fault object and fault characteristics, the matching fault information is retrieved from the fault knowledge base of the cloud network knowledge graph; Using a large language model, the root cause and solution of the problem are determined according to the matching fault information as the reply information of the problem information.

[0103] In one implementation, when the device retrieves matching fault information from a fault knowledge base based on the fault object and the fault feature, it is further configured to: Based on the fault object, fault characteristics and context information of the problem information, matching fault information is retrieved in the fault knowledge base.

[0104] The functions of each module in the embodiment of the present application can be referred to the corresponding description in the above method, and have corresponding beneficial effects, which will not be repeated here.

[0105] Figure 5 is a block diagram of an electronic device used to implement an embodiment of the present application. Figure 5 As shown, the electronic device includes: a memory 510 and a processor 520. The memory 510 stores a computer program that can be run on the processor 520. When the processor 520 executes the computer program, the method in the above embodiment is implemented. The number of the memory 510 and the processor 520 can be one or more.

[0106] The electronic device also includes: The communication interface 530 is used to communicate with external devices and perform data exchange transmission.

[0107] If the memory 510, the processor 520 and the communication interface 530 are implemented independently, the memory 510, the processor 520 and the communication interface 530 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0108] Optionally, in a specific implementation, if the memory 510, the processor 520 and the communication interface 530 are integrated on a chip, the memory 510, the processor 520 and the communication interface 530 can communicate with each other through an internal interface.

[0109] An embodiment of the present application provides a computer-readable storage medium storing a computer program, which implements the method provided in the embodiment of the present application when the program is executed by a processor.

[0110] An embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method provided in the embodiment of the present application is implemented.

[0111] An embodiment of the present application also provides a chip, which includes a processor for calling and executing instructions stored in the memory from the memory, so that a communication device equipped with the chip executes the method provided by the embodiment of the present application.

[0112] An embodiment of the present application also provides a chip, including: an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected via an internal connection path, and the processor is used to execute the code in the memory. When the code is executed, the processor is used to execute the method provided in the embodiment of the application.

[0113] It should be understood that the above processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0114] Further, optionally, the above-mentioned memory may include a read-only memory and a random access memory. The memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may include a random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting description, many forms of RAM are available. For example, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (DR RAM).

[0115] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0116] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0117] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0118] Any process or method described in the flow chart or otherwise described herein can be understood as a module, fragment or portion of a code representing one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.

[0119] The logic and / or steps described in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processor or other system that can fetch instructions from an instruction execution system, apparatus or device and execute instructions), or used in combination with these instruction execution systems, apparatuses or devices.

[0120] It should be understood that the various parts of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented with software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium, and when the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0121] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.

[0122] The above is only an exemplary embodiment of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope recorded in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A cloud network connectivity detection method, characterized in that: include: Obtaining problem information about the cloud network and identifying the problem information; If it is determined according to the recognition result that a connectivity test is required, determining a test start point and a test end point from the recognition result; Based on the cloud network knowledge graph, determine the detection path between the detection starting point and the detection end point; The cloud network knowledge graph represents the dynamic changes of resources in the cloud network and the association relationships between the resources; According to the detection path, a connectivity detection is performed in the cloud network to obtain a detection result.

2. The method according to claim 1, characterized in that The identifying the problem information includes: Inputting the question information into a large language model, and using the large language model to perform entity recognition and intent recognition; Determining that connectivity detection is required according to the recognition result, and determining a detection start point and a detection end point from the recognition result, includes: Determining the need for connectivity detection according to the recognition result of the intent recognition; The detection starting point and the detection end point are determined according to the recognition result of the entity recognition.

3. The method according to claim 1, characterized in that The cloud network knowledge graph is constructed in the following way: Based on the cloud network topology, the resources are mapped into nodes in the cloud network knowledge graph, and the association relationships are mapped into edges in the cloud network knowledge graph; Attribute information of the nodes and the edges is determined according to the real-time status data, fault handling data and configuration data of the cloud network.

4. The method according to claim 1 or 3, characterized in that: The determining of the detection path between the detection starting point and the detection end point based on the cloud network knowledge graph includes: Determine a possible path between the detection starting point and the detection end point based on the nodes corresponding to the detection starting point and the detection end point in the cloud network knowledge graph, and the edges between the nodes; The detection path is determined in the possible paths according to the attribute information of the nodes in the possible paths and the attribute information of the edges between the nodes.

5. The method according to claim 1, characterized in that The performing connectivity detection in the cloud network according to the detection path to obtain a detection result includes: Determine at least one target node in the detection path, and send a connection request between a start node in the detection path and the at least one target node, wherein the connection request is used to establish a communication link between the start node and the at least one target node; Based on the communication link, a detection data packet is sent to obtain a detection result.

6. The method according to claim 5, characterized in that The sending of a detection data packet based on the communication link to obtain a detection result includes: If it is determined based on the real-time status data of the cloud network that the communication link cannot send the detection data packet, a backup communication link is established between the start node and the backup node corresponding to the target node; Based on the standby communication link, a detection data packet is sent to obtain a detection result.

7. The method according to claim 1, characterized in that After performing connectivity detection in the cloud network according to the detection path and obtaining the detection result, the method further includes: If it is determined according to the detection result that the detection starting point and the detection end point are not connected, then obtaining a fault location and a detection performance index from the detection result; Based on the fault location and the detection performance indicator, a root cause of the problem and a solution are determined.

8. The method according to claim 7, characterized in that The determining the root cause and solution of the problem based on the fault location and the detection performance indicator includes: Inputting the fault location and the detection performance index into a large language model to determine the fault object and fault characteristics; Based on the fault object and the fault characteristics, the root cause of the problem and a solution are determined.

9. The method according to claim 8, characterized in that The determining the root cause and solution of the problem based on the fault object and the fault feature includes: Based on the fault object and the fault feature, searching for matching fault information in the fault knowledge base of the cloud network knowledge graph; By using a large language model, the root cause of the problem and a solution are determined according to the matched fault information as reply information to the problem information.

10. The method according to claim 9, characterized in that The retrieving matching fault information in a fault knowledge base based on the fault object and the fault feature further includes: Based on the fault object, the fault characteristics and the context information of the problem information, matching fault information is retrieved from a fault knowledge base.

11. An electronic device, characterized in that: The electronic device comprises a memory, a processor and a computer program stored in the memory, and the processor implements the method according to any one of claims 1 to 10 when executing the computer program.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

13. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images