Network operation and maintenance system and electronic equipment

Abstract

The invention provides a network operation and maintenance system and electronic equipment. The system comprises a multi-agent group, an environment perception tool and an agent long-term memory tool, the multi-agent group further comprises an assistant agent, a monitoring agent, a resource agent and an expert agent, and each agent in the multi-agent group is realized based on a network operation and maintenance large model. That is to say, a multi-agent group is designed by simulating the working mode of a human network operation and maintenance team in order to solve the network operation and maintenance problem, and the network operation and maintenance system is built by combining an environment sensing tool and an agent long-term memory tool on the basis of the multi-agent group. Therefore, the network operation and maintenance system has team cooperation capability, environment perception capability and long-term memory capability among intelligent agents, can adapt to a complex network environment, and is competent for complex network operation and maintenance tasks such as network fault diagnosis.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a network operation and maintenance system and electronic equipment. Background Art

[0002] The intelligent network operation and maintenance system based on big models is the core of modern network management. It monitors network status in real time, quickly diagnoses faults, and optimizes resource allocation through automation and intelligent means, thereby significantly improving network stability and operating efficiency, reducing operation and maintenance costs, and ensuring business continuity. However, due to the complex network environment, the large number of network equipment models and versions, and the complexity of operation and maintenance problems, the complexity of operation and maintenance problems also increases greatly with the scale of the network. How to apply big models to more complex network operation and maintenance work is a difficult point in the application of big models in the field of network operation and maintenance. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present application provides a network operation and maintenance system and electronic equipment.

[0004] According to a first aspect of an embodiment of the present application, a network operation and maintenance system is provided, the system comprising: a multi-agent group, an environment perception tool and an agent long-term memory tool, the multi-agent group comprising an assistant agent, a monitoring agent, a resource agent and an expert agent, each agent in the multi-agent group is implemented based on a network operation and maintenance large model;

[0005] The assistant agent is used to determine the user's operation and maintenance needs by interacting with the user;

[0006] The expert agent is used to convert the operation and maintenance requirements into operation and maintenance tasks, split the operation and maintenance tasks to obtain a first subtask, a second subtask and a third subtask; obtain a first result corresponding to the first subtask by interacting with the monitoring agent, the first result includes monitoring alarm data; obtain a second result corresponding to the second subtask by interacting with the resource agent, the second result includes business application data; obtain a third result by interacting with the environment perception tool, the third result includes environment data; comprehensively analyze the first result, the second result and the third result to obtain the operation and maintenance result corresponding to the operation and maintenance task; call the assistant agent to feed back the operation and maintenance result to the user;

[0007] The agent long-term memory tool is used to record the interaction data between the user and the assistant agent, the interaction data between any agents in the multi-agent group, and is also used for any agent in the multi-agent group to call to implement model training.

[0008] According to a second aspect of an embodiment of the present application, an electronic device is provided, wherein the electronic device is equipped with the network operation and maintenance system as described above.

[0009] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0010] The embodiment of the present application aims to solve network operation and maintenance problems, designs a multi-agent group by simulating the working mode of a human network operation and maintenance team, and builds a network operation and maintenance system based on the multi-agent group, combining environmental perception tools and agent long-term memory tools. The network operation and maintenance system has team collaboration capabilities, environmental perception capabilities, and long-term memory capabilities among agents, so it can adapt to complex network environments and be competent for complex network operation and maintenance tasks such as network fault diagnosis.

[0011] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings, which are incorporated in the specification and constitute a part of this application, illustrate embodiments consistent with the application and, together with the description, serve to explain the principles of the application.

[0013] Figure 1 This is a schematic diagram of a traditional network operation and maintenance solution;

[0014] Figure 2 Schematic diagram of a network operation and maintenance system provided in an embodiment of the present application Figure 1 ;

[0015] Figure 3 Schematic diagram of a network operation and maintenance system provided in an embodiment of the present application Figure 2 ;

[0016] Figure 4 A schematic diagram of an intelligent agent dialogue provided in an embodiment of the present application;

[0017] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application. In the description of the embodiments of the present application, the terms used in the following embodiments are only for the purpose of describing specific embodiments and are not intended to be limiting of the present application.

[0019] It should be noted that in this application, "at least one" means one or more, and "more than one" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The terms "first", "second", "third", etc. (if any) in the specification, claims and drawings of this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0020] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0021] First, let's explain the professional terms and English abbreviations involved in this article: ARP is the full spelling of Address Resolution Protocol, which means Address Resolution Protocol in Chinese;

[0022] VLAN stands for Virtual Local Area Network, which means virtual local area network in Chinese.

[0023] IVL stands for Independent VLAN Learning, which means independent VLAN learning in Chinese.

[0024] MAC address is a full spelling of Media Access Control Address, which means media access control address in Chinese.

[0025] The full spelling of API interface is Application Programming Interface, which means application programming interface in Chinese.

[0026] In the embodiments of the present application, an intelligent agent refers to a system or program with a network operation and maintenance large model as its core, which has autonomous decision-making capabilities, language interaction capabilities, learning and evolution capabilities, collaborative work capabilities, and independent task execution capabilities.

[0027] As mentioned above, how to apply large models to more complex network operation and maintenance tasks is the difficulty of applying large models in the field of network operation and maintenance. For example, checking the types of existing network equipment and equipment feature information is crucial for locating network faults. The following are examples of problems caused by different equipment types:

[0028] Existing network: Figure 1 As shown in the figure, optical transmission equipment is used between the headquarters and the two branch backbone networks, and the network topology is connected through optical fiber. The one-node aggregation network device at the headquarters is an S3610 three-layer switch, and VLAN 10 and VLAN 20 are set, which are the VLANs of branch 1 and branch 2 respectively. The connected ports are the E1 / 1 port and E1 / 2 port of the Ethernet card of the optical transmission equipment respectively. The gateway address of VLAN 10 is: 10.10.10.1 / 24, and the gateway address of VLAN 20 is: 10.10.20.1 / 24.

[0029] Fault phenomenon: Due to the damage of the headquarters node aggregation switch S3610, it was temporarily replaced with an S3600-EI switch, and the fault occurred. The headquarters' S3600-EI switch can sometimes ping the test node address of the branch switch, but the branch switch often cannot ping the various gateway addresses of the headquarters switch, and the network is paralyzed. Check and confirm that the device configuration and the ARP table of the three-layer switch are normal. Replace another S3600-EI, and the phenomenon still exists. After replacing the new S3610, the fault disappears. Based on the above situation, it can be judged that there is no problem with the link and network equipment, and the fault is caused by updating the switch.

[0030] Cause of the problem: The S3610 switch has multiple MAC addresses, and each VLAN gateway can be assigned a different MAC address. However, the S3600-EI three-layer switch has only one MAC address, and each VLAN virtual interface uses the same MAC address to send messages. The Ethernet boards of early optical transmission equipment had poor support for the IVL protocol. When the same network device (such as a low-end three-layer switch, which usually only has one fixed MAC address) is connected through different ports of the same Ethernet board, it does not support the "MAC address + port" identification method and cannot distinguish data sent from different ports using the same MAC address, resulting in MAC address drift.

[0031] Therefore, in order to make the large model adapt to the complex network environment, in addition to the use of network diagnostic tools, it is equally important to understand the current status of the network and analyze the current status and fault problems of the complex network.

[0032] Traditional network operation and maintenance solutions based on large models are mainly based on product knowledge questions and answers and information queries. The core is to call API interfaces through large models to achieve single operation and maintenance functions, such as alarm monitoring queries and service status queries. In the process of observation-thinking-action, this solution skips the observation step and directly enters the thinking and action links, and the thinking link only relies on the thinking ability of the large model. However, due to the hallucination problem of the large model, there is a large uncertainty in its thinking accuracy. In addition, the solution lacks the acquisition and observation of environmental information, and the accuracy of the actions selected by the large model will also have a large deviation, which will lead to poor final task execution.

[0033] In summary, the main disadvantages of traditional network operation and maintenance solutions based on large models are as follows:

[0034] (1) The complexity of tasks that can be completed by a single call to an API using a large model is limited by the capabilities provided by the API;

[0035] (2) The memory and environmental perception capabilities of the large model have not been expanded, resulting in unreliable operation and maintenance results.

[0036] In response to the above problems, the present application provides a network operation and maintenance system and electronic equipment.

[0037] Next, the embodiments of the present application are described in detail.

[0038] The present application embodiment provides a network operation and maintenance system, such as Figure 2 As shown, the system may include: a multi-agent group, an environmental perception tool, and an agent long-term memory tool, wherein the multi-agent group further includes an assistant agent, an expert agent, a monitoring agent, and a resource agent.

[0039] Specifically, the embodiment of the present application is based on the large model of network operation and maintenance, aims to solve network operation and maintenance problems, and refers to the division of responsibilities of the human network operation and maintenance team to create assistant intelligent agents, monitoring intelligent agents, resource intelligent agents and expert intelligent agents respectively. Each intelligent agent performs its own duties in the entire network operation and maintenance work, calls each other, and works together to complete complex network operation and maintenance tasks.

[0040] It should be noted that the network operation and maintenance big model mentioned in the embodiment of this application refers to a big model that has been specially trained with network operation and maintenance knowledge (such as basic network knowledge, network equipment configuration manual, network fault diagnosis, troubleshooting, network optimization, product knowledge, maintenance guidance, etc.). On this basis, combined with the big model's own natural language understanding ability and tool use ability, in the field of network operation and maintenance, the big model can be applied to simple network operation and maintenance work such as network data monitoring and analysis, log query analysis, etc., to assist network operation and maintenance personnel in completing part of the work to a certain extent.

[0041] Therefore, each agent in the multi-agent group of the embodiment of the present application is implemented based on a network operation and maintenance big model, which has a knowledge base close to that of a network operation and maintenance engineer and has the ability to understand and judge basic network problems.

[0042] As described above, the embodiment of the present application designs a multi-agent architecture with reference to the working mode of the human network operation and maintenance team. Figure 3 As shown in the figure, the assistant agent is responsible for communicating and understanding the user's needs, including determining the user's operation and maintenance needs and feeding back the operation and maintenance results to the user; the expert agent is responsible for converting the user's operation and maintenance needs into specific operation and maintenance tasks, and coordinating the monitoring agent, resource agent, assistant agent, etc. to jointly complete the operation and maintenance tasks; the monitoring agent and resource agent are respectively responsible for the execution of monitoring tasks and resource query tasks.

[0043] In practical applications, state diagrams can be used to organize the workflow of intelligent agents, decompose the tasks of intelligent agents into multiple nodes, and set the flow logic between nodes.

[0044] The following is an introduction to each agent in the agent group:

[0045] Assistant agent: Responsible for interacting with users. On the one hand, it determines the operation and maintenance requirements and passes them to the expert agent. On the other hand, it feeds back the operation and maintenance results of the expert agent to the user.

[0046] Specifically, the assistant agent has the ability to understand network knowledge, can communicate directly with the user, understand the user's fault description, determine the user's operation and maintenance needs, and can also generate system responses based on the operation and maintenance results generated by the expert agent, polish the system responses, and finally provide feedback to the user.

[0047] As a specific implementation method, in order to collect and understand user needs more efficiently, the assistant agent converges the user's fault description by setting a problem classification template, assisting the expert agent to better judge the fault problem; through the demand template, it assists the expert agent to collect user needs more comprehensively.

[0048] The embodiment of the present application provides examples of requirement templates of question classification templates, as shown in Table 1 and Table 2 respectively:

[0049] Table 1

[0050]

[0051]

[0052] Table 2

[0053] Serial number information 1 Failure time 2 Fault location ……

[0054] It is understandable that in actual applications, the specific contents of the problem classification template and the operation and maintenance requirement template can be set and adjusted according to the requirements of the application scenario, and the embodiments of the present application do not limit this.

[0055] Monitoring agent: responsible for assisting the expert agent in collecting monitoring alarm data, and can also provide the expert agent with the first preliminary analysis results.

[0056] The above monitoring alarm data includes but is not limited to:

[0057] Device status: the operating status of network devices (such as switches, routers, servers, etc.), including key indicators such as CPU utilization, memory usage, temperature, etc.

[0058] Link status: performance indicators of network links such as bandwidth utilization, delay, and packet loss rate;

[0059] Interface status: network interface connection status, traffic statistics, error counts and other information;

[0060] Alarm information: When the monitoring indicators exceed the preset threshold, detailed alarm information is generated, including alarm level, alarm time, alarm reason, etc.

[0061] In addition, the monitoring agent can identify the most relevant fault source object based on the correlation analysis of alarm information. For example, if multiple alarms point to a port of a switch, then the port may be the source of the fault. In addition, the monitoring agent can also filter and combine multiple alarm information with the same fault source object to form advanced alarm information for fault location, and generate preliminary fault location conclusions based on the filtered and combined advanced alarm information. For example, it is preliminarily determined that the fault occurred on a specific interface of a device.

[0062] The monitoring agent of the embodiment of the present application, on the one hand, provides real-time monitoring alarm information to the expert agent, helping the expert agent to quickly understand the overall operating status of the network; on the other hand, the monitoring agent can further summarize, classify and preliminarily analyze the alarm information, thereby extracting key information and providing the expert agent with preliminary fault location results (i.e., the first preliminary analysis results).

[0063] Resource agent: Responsible for assisting the expert agent in collecting business application data and can also provide the expert agent with a second preliminary analysis result.

[0064] The above business application data includes but is not limited to:

[0065] Service status: including key indicators such as application availability, response time, and error rate;

[0066] Performance indicators: performance indicators of application services, such as CPU and memory usage, number of database connections, thread pool status, etc.

[0067] Configuration information: configuration files and parameter settings of business applications;

[0068] Log files: Log files of application services.

[0069] In addition, the resource agent can combine the collected service status, performance indicators and log information to preliminarily analyze the possible causes of the failure. It can also compare the data before and after the failure, such as the changes in performance indicators, abnormal information in the log, etc., to find the differences. Based on the preliminary analysis results, it can determine the location of the failure, such as a service instance, database, middleware or network link. In addition, the resource agent can also check whether the application's dependencies are correctly installed and configured, and confirm whether there are any missing dependencies or version incompatibility issues.

[0070] The resource agent of the embodiment of the present application can deeply analyze the internal state of the business application based on the preliminary alarm information provided by the monitoring agent, and assist the expert agent in locating the specific cause of the fault. On the one hand, it provides business application information to the expert agent, and on the other hand, the resource agent can further analyze the error information and abnormal behavior in the log file, and provide preliminary troubleshooting results (i.e., the second preliminary analysis results) for the expert agent's troubleshooting work.

[0071] Expert agent: As the core and decision-maker of the multi-agent group, it obtains the operation and maintenance requirements transmitted by the assistant agent, converts the operation and maintenance requirements into operation and maintenance tasks, splits the operation and maintenance tasks to obtain the first subtask, the second subtask and the third subtask; obtains the first result corresponding to the first subtask by interacting with the monitoring agent, and the first result includes monitoring alarm data; obtains the second result corresponding to the second subtask by interacting with the resource agent, and the second result includes business application data; obtains the third result by interacting with the environmental perception tool, and the third result includes environmental data; finally, comprehensively analyzes the first result, the second result, and the third result to obtain the operation and maintenance result corresponding to the operation and maintenance task, and calls the assistant agent to feed back the operation and maintenance result to the user.

[0072] It is worth mentioning that the operation and maintenance results output by the embodiment of the present application mainly refer to the network fault location result and the network fault troubleshooting result. Among them, the network fault location result represents the specific location of the fault, and the network fault troubleshooting result represents the root cause of the fault.

[0073] Furthermore, the monitoring agent and the resource agent can also generate a first preliminary analysis result and a second preliminary analysis result respectively under the arrangement of the expert agent for reference by the expert agent.

[0074] It is understandable that in the above interaction process, agents share information, feedback results, and adjust their own behaviors and strategies according to the output of other agents. For example, after the monitoring agent finds abnormal data, it sends the abnormal data to the expert agent. After the expert agent investigates the cause, it feeds back the analysis results to the assistant agent. The user tries to repair it according to the prompt of the assistant agent, and then the assistant agent feeds back the results of the attempted repair to the monitoring agent, and so on, until the problem is solved. This multi-round, iterative interaction mode can give full play to the advantages of each agent, achieve efficient collaboration, and ensure the smooth resolution of complex operation and maintenance tasks.

[0075] As a specific implementation method, agents can interact with each other through a dialogue window. Specifically, agents can communicate with other agents in the dialogue window by @ing other agents. Since all agents can receive the information in the dialogue window, multiple agents can communicate with each other in this way. For example, if an expert agent determines that it needs to query the interface status of an AP, it can do so by @ing the monitoring agent in the dialogue window:

[0076] @Monitoring agent, query the interface status of AP

[0077] After receiving the @ message from the dialogue window, each agent determines whether it is related to itself and then triggers the corresponding action. The judgment is based on "@" and "agent name".

[0078] In addition to the multi-agent group, the network operation and maintenance system of the embodiment of the present application also includes an agent long-term memory tool and an environment perception tool.

[0079] Considering that large models are limited by their context windows (i.e., the number of tokens they can remember at one time), in actual network operation and maintenance scenarios, this memory space will soon be filled with interaction records in multiple rounds of dialogue, lengthy tool outputs, or additional context that the agent relies on. Therefore, it is crucial to establish an effective memory management strategy, so the network operation and maintenance system of the embodiment of the present application adds an agent long-term memory tool within the framework of the agent.

[0080] The agent long-term memory tool is responsible for the storage and reading of the agent's long-term memory. The long-term memory includes the interaction records between the assistant agent and the user, the interaction records between agents, etc. The agent long-term memory tool saves this data as long-term memory for all agents to query. This allows each agent to maintain context over time, expands the agent's memory capacity, and optimizes its response based on previous interactive exchanges.

[0081] As a preferred implementation, in order to improve storage efficiency, the agent long-term memory tool can use a large model to summarize and refine historical interaction records.

[0082] In addition, if Figure 3 As shown, the agent long-term memory tool can also be used to record agent descriptions, expert knowledge bases, operation and maintenance knowledge bases, environmental data collected by environmental perception tools, etc.

[0083] The agent description refers to the introduction of the agent's role, functions, application scenarios, and features, so that other agents can understand when to switch to the agent. An example is as follows:

[0084] #Agent Description

[0085] This is an intelligent agent for analyzing the root cause of network problems. Its input is the user's problem description, environmental information obtained from various sources, monitoring alarm information, etc. Its output is the analysis conclusion of the root cause of the problem. The output conclusion can be used as a reference for answering users and helping them troubleshoot the problem.

[0086] The environmental perception tool consists of basic modules such as device model query, device version query, device configuration query, and device log query. After the user asks a question, it assists the intelligent agent in completing the environmental information query.

[0087] The environmental perception tool can respond to requests from other intelligent agents, start the environmental information collection workflow, and implement tasks such as device model query, device version query, device configuration query, and device log query.

[0088] It can be seen from the above technical solutions that the embodiment of the present application aims to solve the network operation and maintenance problems, designs a multi-agent group by simulating the working mode of the human network operation and maintenance team, and builds a network operation and maintenance system based on the multi-agent group in combination with the environmental perception tool and the agent long-term memory tool. The network operation and maintenance system has the team collaboration ability between agents (each performs its duties, team discussion, and mutual call), environmental perception ability (used to grasp the current status of the network) and long-term memory ability (used to expand the memory capacity of the agent), so it can adapt to complex network environments and be competent for complex network operation and maintenance tasks such as network fault diagnosis.

[0089] An embodiment of the present application provides an electronic device, wherein the electronic device is equipped with the network operation and maintenance system as described above. The electronic device may include: a memory and one or more processors. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device may perform various functions or steps of the above method embodiment.

[0090] The structure of the electronic device can refer to Figure 5The structure of the electronic device 100 is shown.

[0091] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0092] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0093] In the several embodiments provided in the present application, it should be understood that the disclosed method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of modules or units, which can be electrical, mechanical or other forms.

[0094] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0095] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A network operation and maintenance system, characterized in that: The system includes: a multi-agent group, an environment perception tool and an agent long-term memory tool, wherein the multi-agent group includes an assistant agent, a monitoring agent, a resource agent and an expert agent, and each agent in the multi-agent group is implemented based on a network operation and maintenance large model; The assistant agent is used to determine the user's operation and maintenance needs by interacting with the user; The expert agent is used to convert the operation and maintenance requirements into operation and maintenance tasks, split the operation and maintenance tasks to obtain a first subtask, a second subtask and a third subtask; obtain a first result corresponding to the first subtask by interacting with the monitoring agent, the first result includes monitoring alarm data; obtain a second result corresponding to the second subtask by interacting with the resource agent, the second result includes business application data; obtain a third result by interacting with the environment perception tool, the third result includes environment data; comprehensively analyze the first result, the second result and the third result to obtain the operation and maintenance result corresponding to the operation and maintenance task; call the assistant agent to feed back the operation and maintenance result to the user; The agent long-term memory tool is used to record the interaction data between the user and the assistant agent, the interaction data between any agents in the multi-agent group, and is also used for any agent in the multi-agent group to call to implement model training.

2. The system according to claim 1, characterized in that The assistant agent collects the user's fault description through the problem classification template, and collects the user's demand description through the demand template; and determines the user's operation and maintenance needs by analyzing the fault description and the demand description.

3. The system according to claim 1, characterized in that The expert agent is also used to split the operation and maintenance task to obtain a fourth subtask; obtain a fourth result corresponding to the fourth subtask by interacting with a network diagnostic tool; and perform a comprehensive analysis on the first result, the second result, the third result, and the fourth result to obtain an operation and maintenance result corresponding to the operation and maintenance task.

4. The system according to claim 1, characterized in that The agents in the multi-agent group interact with each other through a dialogue window.

5. The system according to claim 1, characterized in that The environment perception tool is used to collect any one or more of the following environment data: device model, device version, device configuration, and device log.

6. The system according to claim 1, characterized in that The agent long-term memory tool is also used to store expert knowledge base, operation and maintenance knowledge base or agent description.

7. The system according to any one of claims 1 to 6, characterized in that: The monitoring agent is further used to perform a preliminary analysis on the monitoring alarm data collected by itself to obtain a first preliminary analysis result, wherein the first result includes the first preliminary analysis result; The resource agent is further used to perform a preliminary analysis on the business application data collected by itself to obtain a second preliminary analysis result, and the second result includes the second preliminary analysis result.

8. The system according to claim 1, characterized in that The operation and maintenance results generated by the expert agent include network fault location results and network fault troubleshooting results.

9. The system according to claim 1, characterized in that The monitoring alarm data collected by the monitoring agent includes any one or more of the following: device status, link status, interface status, and alarm information.

10. The system according to claim 1, characterized in that The business application data collected by the resource agent includes any one or more of the following: service status, performance indicators, configuration information, and log files.

11. An electronic device, characterized in that: The electronic device is equipped with a network operation and maintenance system as described in any one of claims 1-10.

Images